@fleetagent/pi-ai 0.1.4 → 0.1.8

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (73) hide show
  1. package/dist/image-models.generated.d.ts +64 -19
  2. package/dist/image-models.generated.d.ts.map +1 -1
  3. package/dist/image-models.generated.js +90 -45
  4. package/dist/image-models.generated.js.map +1 -1
  5. package/dist/index.d.ts +1 -0
  6. package/dist/index.d.ts.map +1 -1
  7. package/dist/index.js +1 -0
  8. package/dist/index.js.map +1 -1
  9. package/dist/models.generated.d.ts +4110 -2145
  10. package/dist/models.generated.d.ts.map +1 -1
  11. package/dist/models.generated.js +2488 -708
  12. package/dist/models.generated.js.map +1 -1
  13. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  14. package/dist/providers/amazon-bedrock.js +41 -2
  15. package/dist/providers/amazon-bedrock.js.map +1 -1
  16. package/dist/providers/anthropic.d.ts.map +1 -1
  17. package/dist/providers/anthropic.js +1 -1
  18. package/dist/providers/anthropic.js.map +1 -1
  19. package/dist/providers/azure-openai-responses.d.ts.map +1 -1
  20. package/dist/providers/azure-openai-responses.js +6 -16
  21. package/dist/providers/azure-openai-responses.js.map +1 -1
  22. package/dist/providers/google-vertex.d.ts.map +1 -1
  23. package/dist/providers/google-vertex.js +1 -1
  24. package/dist/providers/google-vertex.js.map +1 -1
  25. package/dist/providers/google.d.ts.map +1 -1
  26. package/dist/providers/google.js +1 -1
  27. package/dist/providers/google.js.map +1 -1
  28. package/dist/providers/mistral.d.ts.map +1 -1
  29. package/dist/providers/mistral.js +1 -1
  30. package/dist/providers/mistral.js.map +1 -1
  31. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  32. package/dist/providers/openai-codex-responses.js +89 -50
  33. package/dist/providers/openai-codex-responses.js.map +1 -1
  34. package/dist/providers/openai-completions.d.ts.map +1 -1
  35. package/dist/providers/openai-completions.js +13 -4
  36. package/dist/providers/openai-completions.js.map +1 -1
  37. package/dist/providers/openai-responses-shared.d.ts.map +1 -1
  38. package/dist/providers/openai-responses-shared.js +157 -181
  39. package/dist/providers/openai-responses-shared.js.map +1 -1
  40. package/dist/providers/openai-responses.d.ts.map +1 -1
  41. package/dist/providers/openai-responses.js +6 -16
  42. package/dist/providers/openai-responses.js.map +1 -1
  43. package/dist/providers/simple-options.d.ts +3 -2
  44. package/dist/providers/simple-options.d.ts.map +1 -1
  45. package/dist/providers/simple-options.js +11 -2
  46. package/dist/providers/simple-options.js.map +1 -1
  47. package/dist/types.d.ts +4 -2
  48. package/dist/types.d.ts.map +1 -1
  49. package/dist/types.js.map +1 -1
  50. package/dist/utils/error-body.d.ts +12 -0
  51. package/dist/utils/error-body.d.ts.map +1 -0
  52. package/dist/utils/error-body.js +70 -0
  53. package/dist/utils/error-body.js.map +1 -0
  54. package/dist/utils/estimate.d.ts +17 -0
  55. package/dist/utils/estimate.d.ts.map +1 -0
  56. package/dist/utils/estimate.js +95 -0
  57. package/dist/utils/estimate.js.map +1 -0
  58. package/dist/utils/oauth/device-code.d.ts +2 -0
  59. package/dist/utils/oauth/device-code.d.ts.map +1 -1
  60. package/dist/utils/oauth/device-code.js +22 -8
  61. package/dist/utils/oauth/device-code.js.map +1 -1
  62. package/dist/utils/oauth/github-copilot.d.ts.map +1 -1
  63. package/dist/utils/oauth/github-copilot.js +3 -2
  64. package/dist/utils/oauth/github-copilot.js.map +1 -1
  65. package/dist/utils/overflow.d.ts +1 -0
  66. package/dist/utils/overflow.d.ts.map +1 -1
  67. package/dist/utils/overflow.js +3 -0
  68. package/dist/utils/overflow.js.map +1 -1
  69. package/dist/utils/retry.d.ts +3 -0
  70. package/dist/utils/retry.d.ts.map +1 -0
  71. package/dist/utils/retry.js +58 -0
  72. package/dist/utils/retry.js.map +1 -0
  73. package/package.json +4 -3
@@ -70,6 +70,23 @@ export const MODELS = {
70
70
  contextWindow: 300000,
71
71
  maxTokens: 8192,
72
72
  },
73
+ "anthropic.claude-fable-5": {
74
+ id: "anthropic.claude-fable-5",
75
+ name: "Claude Fable 5",
76
+ api: "bedrock-converse-stream",
77
+ provider: "amazon-bedrock",
78
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
79
+ reasoning: true,
80
+ input: ["text", "image"],
81
+ cost: {
82
+ input: 10,
83
+ output: 50,
84
+ cacheRead: 1,
85
+ cacheWrite: 12.5,
86
+ },
87
+ contextWindow: 1000000,
88
+ maxTokens: 128000,
89
+ },
73
90
  "anthropic.claude-haiku-4-5-20251001-v1:0": {
74
91
  id: "anthropic.claude-haiku-4-5-20251001-v1:0",
75
92
  name: "Claude Haiku 4.5",
@@ -208,6 +225,23 @@ export const MODELS = {
208
225
  contextWindow: 1000000,
209
226
  maxTokens: 64000,
210
227
  },
228
+ "anthropic.claude-sonnet-5": {
229
+ id: "anthropic.claude-sonnet-5",
230
+ name: "Claude Sonnet 5",
231
+ api: "bedrock-converse-stream",
232
+ provider: "amazon-bedrock",
233
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
234
+ reasoning: true,
235
+ input: ["text", "image"],
236
+ cost: {
237
+ input: 2,
238
+ output: 10,
239
+ cacheRead: 0.2,
240
+ cacheWrite: 2.5,
241
+ },
242
+ contextWindow: 1000000,
243
+ maxTokens: 128000,
244
+ },
211
245
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
212
246
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
213
247
  name: "Claude Haiku 4.5 (AU)",
@@ -294,6 +328,23 @@ export const MODELS = {
294
328
  contextWindow: 1000000,
295
329
  maxTokens: 128000,
296
330
  },
331
+ "au.anthropic.claude-sonnet-5": {
332
+ id: "au.anthropic.claude-sonnet-5",
333
+ name: "Claude Sonnet 5 (AU)",
334
+ api: "bedrock-converse-stream",
335
+ provider: "amazon-bedrock",
336
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
337
+ reasoning: true,
338
+ input: ["text", "image"],
339
+ cost: {
340
+ input: 2,
341
+ output: 10,
342
+ cacheRead: 0.2,
343
+ cacheWrite: 2.5,
344
+ },
345
+ contextWindow: 1000000,
346
+ maxTokens: 128000,
347
+ },
297
348
  "deepseek.r1-v1:0": {
298
349
  id: "deepseek.r1-v1:0",
299
350
  name: "DeepSeek-R1",
@@ -371,10 +422,10 @@ export const MODELS = {
371
422
  reasoning: true,
372
423
  input: ["text", "image"],
373
424
  cost: {
374
- input: 1,
375
- output: 5,
376
- cacheRead: 0.1,
377
- cacheWrite: 1.25,
425
+ input: 1.1,
426
+ output: 5.5,
427
+ cacheRead: 0.11,
428
+ cacheWrite: 1.375,
378
429
  },
379
430
  contextWindow: 200000,
380
431
  maxTokens: 64000,
@@ -388,10 +439,10 @@ export const MODELS = {
388
439
  reasoning: true,
389
440
  input: ["text", "image"],
390
441
  cost: {
391
- input: 5,
392
- output: 25,
393
- cacheRead: 0.5,
394
- cacheWrite: 6.25,
442
+ input: 5.5,
443
+ output: 27.5,
444
+ cacheRead: 0.55,
445
+ cacheWrite: 6.875,
395
446
  },
396
447
  contextWindow: 200000,
397
448
  maxTokens: 64000,
@@ -483,6 +534,23 @@ export const MODELS = {
483
534
  contextWindow: 1000000,
484
535
  maxTokens: 64000,
485
536
  },
537
+ "eu.anthropic.claude-sonnet-5": {
538
+ id: "eu.anthropic.claude-sonnet-5",
539
+ name: "Claude Sonnet 5 (EU)",
540
+ api: "bedrock-converse-stream",
541
+ provider: "amazon-bedrock",
542
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
543
+ reasoning: true,
544
+ input: ["text", "image"],
545
+ cost: {
546
+ input: 2.2,
547
+ output: 11,
548
+ cacheRead: 0.22,
549
+ cacheWrite: 2.75,
550
+ },
551
+ contextWindow: 1000000,
552
+ maxTokens: 128000,
553
+ },
486
554
  "global.anthropic.claude-fable-5": {
487
555
  id: "global.anthropic.claude-fable-5",
488
556
  name: "Claude Fable 5 (Global)",
@@ -621,6 +689,23 @@ export const MODELS = {
621
689
  contextWindow: 1000000,
622
690
  maxTokens: 64000,
623
691
  },
692
+ "global.anthropic.claude-sonnet-5": {
693
+ id: "global.anthropic.claude-sonnet-5",
694
+ name: "Claude Sonnet 5 (Global)",
695
+ api: "bedrock-converse-stream",
696
+ provider: "amazon-bedrock",
697
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
698
+ reasoning: true,
699
+ input: ["text", "image"],
700
+ cost: {
701
+ input: 2,
702
+ output: 10,
703
+ cacheRead: 0.2,
704
+ cacheWrite: 2.5,
705
+ },
706
+ contextWindow: 1000000,
707
+ maxTokens: 128000,
708
+ },
624
709
  "google.gemma-3-27b-it": {
625
710
  id: "google.gemma-3-27b-it",
626
711
  name: "Google Gemma 3 27B Instruct",
@@ -655,6 +740,23 @@ export const MODELS = {
655
740
  contextWindow: 128000,
656
741
  maxTokens: 4096,
657
742
  },
743
+ "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
744
+ id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
745
+ name: "Claude Haiku 4.5 (JP)",
746
+ api: "bedrock-converse-stream",
747
+ provider: "amazon-bedrock",
748
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
749
+ reasoning: true,
750
+ input: ["text", "image"],
751
+ cost: {
752
+ input: 1,
753
+ output: 5,
754
+ cacheRead: 0.1,
755
+ cacheWrite: 1.25,
756
+ },
757
+ contextWindow: 200000,
758
+ maxTokens: 64000,
759
+ },
658
760
  "jp.anthropic.claude-opus-4-7": {
659
761
  id: "jp.anthropic.claude-opus-4-7",
660
762
  name: "Claude Opus 4.7 (JP)",
@@ -724,6 +826,23 @@ export const MODELS = {
724
826
  contextWindow: 1000000,
725
827
  maxTokens: 64000,
726
828
  },
829
+ "jp.anthropic.claude-sonnet-5": {
830
+ id: "jp.anthropic.claude-sonnet-5",
831
+ name: "Claude Sonnet 5 (JP)",
832
+ api: "bedrock-converse-stream",
833
+ provider: "amazon-bedrock",
834
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
835
+ reasoning: true,
836
+ input: ["text", "image"],
837
+ cost: {
838
+ input: 2,
839
+ output: 10,
840
+ cacheRead: 0.2,
841
+ cacheWrite: 2.5,
842
+ },
843
+ contextWindow: 1000000,
844
+ maxTokens: 128000,
845
+ },
727
846
  "meta.llama3-1-70b-instruct-v1:0": {
728
847
  id: "meta.llama3-1-70b-instruct-v1:0",
729
848
  name: "Llama 3.1 70B Instruct",
@@ -1527,6 +1646,23 @@ export const MODELS = {
1527
1646
  contextWindow: 1000000,
1528
1647
  maxTokens: 64000,
1529
1648
  },
1649
+ "us.anthropic.claude-sonnet-5": {
1650
+ id: "us.anthropic.claude-sonnet-5",
1651
+ name: "Claude Sonnet 5 (US)",
1652
+ api: "bedrock-converse-stream",
1653
+ provider: "amazon-bedrock",
1654
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1655
+ reasoning: true,
1656
+ input: ["text", "image"],
1657
+ cost: {
1658
+ input: 2,
1659
+ output: 10,
1660
+ cacheRead: 0.2,
1661
+ cacheWrite: 2.5,
1662
+ },
1663
+ contextWindow: 1000000,
1664
+ maxTokens: 128000,
1665
+ },
1530
1666
  "us.deepseek.r1-v1:0": {
1531
1667
  id: "us.deepseek.r1-v1:0",
1532
1668
  name: "DeepSeek-R1 (US)",
@@ -1612,6 +1748,23 @@ export const MODELS = {
1612
1748
  contextWindow: 1040000,
1613
1749
  maxTokens: 8192,
1614
1750
  },
1751
+ "xai.grok-4.3": {
1752
+ id: "xai.grok-4.3",
1753
+ name: "Grok 4.3",
1754
+ api: "bedrock-converse-stream",
1755
+ provider: "amazon-bedrock",
1756
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1757
+ reasoning: true,
1758
+ input: ["text", "image"],
1759
+ cost: {
1760
+ input: 1.25,
1761
+ output: 2.5,
1762
+ cacheRead: 0.2,
1763
+ cacheWrite: 0,
1764
+ },
1765
+ contextWindow: 1000000,
1766
+ maxTokens: 131072,
1767
+ },
1615
1768
  "zai.glm-4.7": {
1616
1769
  id: "zai.glm-4.7",
1617
1770
  name: "GLM-4.7",
@@ -1665,142 +1818,6 @@ export const MODELS = {
1665
1818
  },
1666
1819
  },
1667
1820
  "anthropic": {
1668
- "claude-3-5-haiku-20241022": {
1669
- id: "claude-3-5-haiku-20241022",
1670
- name: "Claude Haiku 3.5",
1671
- api: "anthropic-messages",
1672
- provider: "anthropic",
1673
- baseUrl: "https://api.anthropic.com",
1674
- reasoning: false,
1675
- input: ["text", "image"],
1676
- cost: {
1677
- input: 0.8,
1678
- output: 4,
1679
- cacheRead: 0.08,
1680
- cacheWrite: 1,
1681
- },
1682
- contextWindow: 200000,
1683
- maxTokens: 8192,
1684
- },
1685
- "claude-3-5-haiku-latest": {
1686
- id: "claude-3-5-haiku-latest",
1687
- name: "Claude Haiku 3.5 (latest)",
1688
- api: "anthropic-messages",
1689
- provider: "anthropic",
1690
- baseUrl: "https://api.anthropic.com",
1691
- reasoning: false,
1692
- input: ["text", "image"],
1693
- cost: {
1694
- input: 0.8,
1695
- output: 4,
1696
- cacheRead: 0.08,
1697
- cacheWrite: 1,
1698
- },
1699
- contextWindow: 200000,
1700
- maxTokens: 8192,
1701
- },
1702
- "claude-3-5-sonnet-20240620": {
1703
- id: "claude-3-5-sonnet-20240620",
1704
- name: "Claude Sonnet 3.5",
1705
- api: "anthropic-messages",
1706
- provider: "anthropic",
1707
- baseUrl: "https://api.anthropic.com",
1708
- reasoning: false,
1709
- input: ["text", "image"],
1710
- cost: {
1711
- input: 3,
1712
- output: 15,
1713
- cacheRead: 0.3,
1714
- cacheWrite: 3.75,
1715
- },
1716
- contextWindow: 200000,
1717
- maxTokens: 8192,
1718
- },
1719
- "claude-3-5-sonnet-20241022": {
1720
- id: "claude-3-5-sonnet-20241022",
1721
- name: "Claude Sonnet 3.5 v2",
1722
- api: "anthropic-messages",
1723
- provider: "anthropic",
1724
- baseUrl: "https://api.anthropic.com",
1725
- reasoning: false,
1726
- input: ["text", "image"],
1727
- cost: {
1728
- input: 3,
1729
- output: 15,
1730
- cacheRead: 0.3,
1731
- cacheWrite: 3.75,
1732
- },
1733
- contextWindow: 200000,
1734
- maxTokens: 8192,
1735
- },
1736
- "claude-3-7-sonnet-20250219": {
1737
- id: "claude-3-7-sonnet-20250219",
1738
- name: "Claude Sonnet 3.7",
1739
- api: "anthropic-messages",
1740
- provider: "anthropic",
1741
- baseUrl: "https://api.anthropic.com",
1742
- reasoning: true,
1743
- input: ["text", "image"],
1744
- cost: {
1745
- input: 3,
1746
- output: 15,
1747
- cacheRead: 0.3,
1748
- cacheWrite: 3.75,
1749
- },
1750
- contextWindow: 200000,
1751
- maxTokens: 64000,
1752
- },
1753
- "claude-3-haiku-20240307": {
1754
- id: "claude-3-haiku-20240307",
1755
- name: "Claude Haiku 3",
1756
- api: "anthropic-messages",
1757
- provider: "anthropic",
1758
- baseUrl: "https://api.anthropic.com",
1759
- reasoning: false,
1760
- input: ["text", "image"],
1761
- cost: {
1762
- input: 0.25,
1763
- output: 1.25,
1764
- cacheRead: 0.03,
1765
- cacheWrite: 0.3,
1766
- },
1767
- contextWindow: 200000,
1768
- maxTokens: 4096,
1769
- },
1770
- "claude-3-opus-20240229": {
1771
- id: "claude-3-opus-20240229",
1772
- name: "Claude Opus 3",
1773
- api: "anthropic-messages",
1774
- provider: "anthropic",
1775
- baseUrl: "https://api.anthropic.com",
1776
- reasoning: false,
1777
- input: ["text", "image"],
1778
- cost: {
1779
- input: 15,
1780
- output: 75,
1781
- cacheRead: 1.5,
1782
- cacheWrite: 18.75,
1783
- },
1784
- contextWindow: 200000,
1785
- maxTokens: 4096,
1786
- },
1787
- "claude-3-sonnet-20240229": {
1788
- id: "claude-3-sonnet-20240229",
1789
- name: "Claude Sonnet 3",
1790
- api: "anthropic-messages",
1791
- provider: "anthropic",
1792
- baseUrl: "https://api.anthropic.com",
1793
- reasoning: false,
1794
- input: ["text", "image"],
1795
- cost: {
1796
- input: 3,
1797
- output: 15,
1798
- cacheRead: 0.3,
1799
- cacheWrite: 0.3,
1800
- },
1801
- contextWindow: 200000,
1802
- maxTokens: 4096,
1803
- },
1804
1821
  "claude-fable-5": {
1805
1822
  id: "claude-fable-5",
1806
1823
  name: "Claude Fable 5",
@@ -1852,26 +1869,9 @@ export const MODELS = {
1852
1869
  contextWindow: 200000,
1853
1870
  maxTokens: 64000,
1854
1871
  },
1855
- "claude-opus-4-0": {
1856
- id: "claude-opus-4-0",
1857
- name: "Claude Opus 4 (latest)",
1858
- api: "anthropic-messages",
1859
- provider: "anthropic",
1860
- baseUrl: "https://api.anthropic.com",
1861
- reasoning: true,
1862
- input: ["text", "image"],
1863
- cost: {
1864
- input: 15,
1865
- output: 75,
1866
- cacheRead: 1.5,
1867
- cacheWrite: 18.75,
1868
- },
1869
- contextWindow: 200000,
1870
- maxTokens: 32000,
1871
- },
1872
- "claude-opus-4-1": {
1873
- id: "claude-opus-4-1",
1874
- name: "Claude Opus 4.1 (latest)",
1872
+ "claude-opus-4-1": {
1873
+ id: "claude-opus-4-1",
1874
+ name: "Claude Opus 4.1 (latest)",
1875
1875
  api: "anthropic-messages",
1876
1876
  provider: "anthropic",
1877
1877
  baseUrl: "https://api.anthropic.com",
@@ -1903,23 +1903,6 @@ export const MODELS = {
1903
1903
  contextWindow: 200000,
1904
1904
  maxTokens: 32000,
1905
1905
  },
1906
- "claude-opus-4-20250514": {
1907
- id: "claude-opus-4-20250514",
1908
- name: "Claude Opus 4",
1909
- api: "anthropic-messages",
1910
- provider: "anthropic",
1911
- baseUrl: "https://api.anthropic.com",
1912
- reasoning: true,
1913
- input: ["text", "image"],
1914
- cost: {
1915
- input: 15,
1916
- output: 75,
1917
- cacheRead: 1.5,
1918
- cacheWrite: 18.75,
1919
- },
1920
- contextWindow: 200000,
1921
- maxTokens: 32000,
1922
- },
1923
1906
  "claude-opus-4-5": {
1924
1907
  id: "claude-opus-4-5",
1925
1908
  name: "Claude Opus 4.5 (latest)",
@@ -2010,40 +1993,6 @@ export const MODELS = {
2010
1993
  contextWindow: 1000000,
2011
1994
  maxTokens: 128000,
2012
1995
  },
2013
- "claude-sonnet-4-0": {
2014
- id: "claude-sonnet-4-0",
2015
- name: "Claude Sonnet 4 (latest)",
2016
- api: "anthropic-messages",
2017
- provider: "anthropic",
2018
- baseUrl: "https://api.anthropic.com",
2019
- reasoning: true,
2020
- input: ["text", "image"],
2021
- cost: {
2022
- input: 3,
2023
- output: 15,
2024
- cacheRead: 0.3,
2025
- cacheWrite: 3.75,
2026
- },
2027
- contextWindow: 200000,
2028
- maxTokens: 64000,
2029
- },
2030
- "claude-sonnet-4-20250514": {
2031
- id: "claude-sonnet-4-20250514",
2032
- name: "Claude Sonnet 4",
2033
- api: "anthropic-messages",
2034
- provider: "anthropic",
2035
- baseUrl: "https://api.anthropic.com",
2036
- reasoning: true,
2037
- input: ["text", "image"],
2038
- cost: {
2039
- input: 3,
2040
- output: 15,
2041
- cacheRead: 0.3,
2042
- cacheWrite: 3.75,
2043
- },
2044
- contextWindow: 200000,
2045
- maxTokens: 64000,
2046
- },
2047
1996
  "claude-sonnet-4-5": {
2048
1997
  id: "claude-sonnet-4-5",
2049
1998
  name: "Claude Sonnet 4.5 (latest)",
@@ -2058,7 +2007,7 @@ export const MODELS = {
2058
2007
  cacheRead: 0.3,
2059
2008
  cacheWrite: 3.75,
2060
2009
  },
2061
- contextWindow: 200000,
2010
+ contextWindow: 1000000,
2062
2011
  maxTokens: 64000,
2063
2012
  },
2064
2013
  "claude-sonnet-4-5-20250929": {
@@ -2075,7 +2024,7 @@ export const MODELS = {
2075
2024
  cacheRead: 0.3,
2076
2025
  cacheWrite: 3.75,
2077
2026
  },
2078
- contextWindow: 200000,
2027
+ contextWindow: 1000000,
2079
2028
  maxTokens: 64000,
2080
2029
  },
2081
2030
  "claude-sonnet-4-6": {
@@ -2094,7 +2043,24 @@ export const MODELS = {
2094
2043
  cacheWrite: 3.75,
2095
2044
  },
2096
2045
  contextWindow: 1000000,
2097
- maxTokens: 64000,
2046
+ maxTokens: 128000,
2047
+ },
2048
+ "claude-sonnet-5": {
2049
+ id: "claude-sonnet-5",
2050
+ name: "Claude Sonnet 5",
2051
+ api: "anthropic-messages",
2052
+ provider: "anthropic",
2053
+ baseUrl: "https://api.anthropic.com",
2054
+ reasoning: true,
2055
+ input: ["text", "image"],
2056
+ cost: {
2057
+ input: 2,
2058
+ output: 10,
2059
+ cacheRead: 0.2,
2060
+ cacheWrite: 2.5,
2061
+ },
2062
+ contextWindow: 1000000,
2063
+ maxTokens: 128000,
2098
2064
  },
2099
2065
  },
2100
2066
  "azure-openai-responses": {
@@ -2700,6 +2666,95 @@ export const MODELS = {
2700
2666
  contextWindow: 1050000,
2701
2667
  maxTokens: 128000,
2702
2668
  },
2669
+ "gpt-5.6": {
2670
+ id: "gpt-5.6",
2671
+ name: "GPT-5.6",
2672
+ api: "azure-openai-responses",
2673
+ provider: "azure-openai-responses",
2674
+ baseUrl: "",
2675
+ reasoning: true,
2676
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2677
+ input: ["text", "image"],
2678
+ cost: {
2679
+ input: 5,
2680
+ output: 30,
2681
+ cacheRead: 0.5,
2682
+ cacheWrite: 6.25,
2683
+ },
2684
+ contextWindow: 1050000,
2685
+ maxTokens: 128000,
2686
+ },
2687
+ "gpt-5.6-luna": {
2688
+ id: "gpt-5.6-luna",
2689
+ name: "GPT-5.6 Luna",
2690
+ api: "azure-openai-responses",
2691
+ provider: "azure-openai-responses",
2692
+ baseUrl: "",
2693
+ reasoning: true,
2694
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2695
+ input: ["text", "image"],
2696
+ cost: {
2697
+ input: 1,
2698
+ output: 6,
2699
+ cacheRead: 0.1,
2700
+ cacheWrite: 1.25,
2701
+ },
2702
+ contextWindow: 1050000,
2703
+ maxTokens: 128000,
2704
+ },
2705
+ "gpt-5.6-sol": {
2706
+ id: "gpt-5.6-sol",
2707
+ name: "GPT-5.6 Sol",
2708
+ api: "azure-openai-responses",
2709
+ provider: "azure-openai-responses",
2710
+ baseUrl: "",
2711
+ reasoning: true,
2712
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2713
+ input: ["text", "image"],
2714
+ cost: {
2715
+ input: 5,
2716
+ output: 30,
2717
+ cacheRead: 0.5,
2718
+ cacheWrite: 6.25,
2719
+ },
2720
+ contextWindow: 1050000,
2721
+ maxTokens: 128000,
2722
+ },
2723
+ "gpt-5.6-terra": {
2724
+ id: "gpt-5.6-terra",
2725
+ name: "GPT-5.6 Terra",
2726
+ api: "azure-openai-responses",
2727
+ provider: "azure-openai-responses",
2728
+ baseUrl: "",
2729
+ reasoning: true,
2730
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2731
+ input: ["text", "image"],
2732
+ cost: {
2733
+ input: 2.5,
2734
+ output: 15,
2735
+ cacheRead: 0.25,
2736
+ cacheWrite: 3.125,
2737
+ },
2738
+ contextWindow: 1050000,
2739
+ maxTokens: 128000,
2740
+ },
2741
+ "gpt-realtime-2.1": {
2742
+ id: "gpt-realtime-2.1",
2743
+ name: "GPT-Realtime-2.1",
2744
+ api: "azure-openai-responses",
2745
+ provider: "azure-openai-responses",
2746
+ baseUrl: "",
2747
+ reasoning: true,
2748
+ input: ["text", "image"],
2749
+ cost: {
2750
+ input: 4,
2751
+ output: 24,
2752
+ cacheRead: 0.4,
2753
+ cacheWrite: 0,
2754
+ },
2755
+ contextWindow: 128000,
2756
+ maxTokens: 32000,
2757
+ },
2703
2758
  "o1": {
2704
2759
  id: "o1",
2705
2760
  name: "o1",
@@ -2838,6 +2893,23 @@ export const MODELS = {
2838
2893
  },
2839
2894
  },
2840
2895
  "cerebras": {
2896
+ "gemma-4-31b": {
2897
+ id: "gemma-4-31b",
2898
+ name: "Gemma 4 31B IT",
2899
+ api: "openai-completions",
2900
+ provider: "cerebras",
2901
+ baseUrl: "https://api.cerebras.ai/v1",
2902
+ reasoning: true,
2903
+ input: ["text", "image"],
2904
+ cost: {
2905
+ input: 0.99,
2906
+ output: 1.49,
2907
+ cacheRead: 0,
2908
+ cacheWrite: 0,
2909
+ },
2910
+ contextWindow: 131072,
2911
+ maxTokens: 40960,
2912
+ },
2841
2913
  "gpt-oss-120b": {
2842
2914
  id: "gpt-oss-120b",
2843
2915
  name: "GPT OSS 120B",
@@ -3169,6 +3241,23 @@ export const MODELS = {
3169
3241
  contextWindow: 1000000,
3170
3242
  maxTokens: 64000,
3171
3243
  },
3244
+ "claude-sonnet-5": {
3245
+ id: "claude-sonnet-5",
3246
+ name: "Claude Sonnet 5",
3247
+ api: "anthropic-messages",
3248
+ provider: "cloudflare-ai-gateway",
3249
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3250
+ reasoning: true,
3251
+ input: ["text", "image"],
3252
+ cost: {
3253
+ input: 2,
3254
+ output: 10,
3255
+ cacheRead: 0.2,
3256
+ cacheWrite: 2.5,
3257
+ },
3258
+ contextWindow: 1000000,
3259
+ maxTokens: 128000,
3260
+ },
3172
3261
  "gpt-4": {
3173
3262
  id: "gpt-4",
3174
3263
  name: "GPT-4",
@@ -3738,6 +3827,24 @@ export const MODELS = {
3738
3827
  contextWindow: 131072,
3739
3828
  maxTokens: 131072,
3740
3829
  },
3830
+ "@cf/zai-org/glm-5.2": {
3831
+ id: "@cf/zai-org/glm-5.2",
3832
+ name: "Glm 5.2",
3833
+ api: "openai-completions",
3834
+ provider: "cloudflare-workers-ai",
3835
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3836
+ compat: { "sendSessionAffinityHeaders": true },
3837
+ reasoning: true,
3838
+ input: ["text"],
3839
+ cost: {
3840
+ input: 1.4,
3841
+ output: 4.4,
3842
+ cacheRead: 0.26,
3843
+ cacheWrite: 0,
3844
+ },
3845
+ contextWindow: 262144,
3846
+ maxTokens: 262144,
3847
+ },
3741
3848
  },
3742
3849
  "deepseek": {
3743
3850
  "deepseek-v4-flash": {
@@ -3792,7 +3899,7 @@ export const MODELS = {
3792
3899
  cost: {
3793
3900
  input: 0.14,
3794
3901
  output: 0.28,
3795
- cacheRead: 0.03,
3902
+ cacheRead: 0.028,
3796
3903
  cacheWrite: 0,
3797
3904
  },
3798
3905
  contextWindow: 1000000,
@@ -3834,6 +3941,24 @@ export const MODELS = {
3834
3941
  contextWindow: 202800,
3835
3942
  maxTokens: 131072,
3836
3943
  },
3944
+ "accounts/fireworks/models/glm-5p2": {
3945
+ id: "accounts/fireworks/models/glm-5p2",
3946
+ name: "GLM 5.2",
3947
+ api: "anthropic-messages",
3948
+ provider: "fireworks",
3949
+ baseUrl: "https://api.fireworks.ai/inference",
3950
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
3951
+ reasoning: true,
3952
+ input: ["text"],
3953
+ cost: {
3954
+ input: 1.4,
3955
+ output: 4.4,
3956
+ cacheRead: 0.14,
3957
+ cacheWrite: 0,
3958
+ },
3959
+ contextWindow: 1048575,
3960
+ maxTokens: 131072,
3961
+ },
3837
3962
  "accounts/fireworks/models/gpt-oss-120b": {
3838
3963
  id: "accounts/fireworks/models/gpt-oss-120b",
3839
3964
  name: "GPT OSS 120B",
@@ -3846,7 +3971,7 @@ export const MODELS = {
3846
3971
  cost: {
3847
3972
  input: 0.15,
3848
3973
  output: 0.6,
3849
- cacheRead: 0.01,
3974
+ cacheRead: 0.015,
3850
3975
  cacheWrite: 0,
3851
3976
  },
3852
3977
  contextWindow: 131072,
@@ -3978,6 +4103,24 @@ export const MODELS = {
3978
4103
  contextWindow: 202800,
3979
4104
  maxTokens: 131072,
3980
4105
  },
4106
+ "accounts/fireworks/routers/glm-5p2-fast": {
4107
+ id: "accounts/fireworks/routers/glm-5p2-fast",
4108
+ name: "GLM 5.2 Fast",
4109
+ api: "anthropic-messages",
4110
+ provider: "fireworks",
4111
+ baseUrl: "https://api.fireworks.ai/inference",
4112
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4113
+ reasoning: true,
4114
+ input: ["text"],
4115
+ cost: {
4116
+ input: 2.1,
4117
+ output: 6.6,
4118
+ cacheRead: 0.21,
4119
+ cacheWrite: 0,
4120
+ },
4121
+ contextWindow: 1048575,
4122
+ maxTokens: 131072,
4123
+ },
3981
4124
  "accounts/fireworks/routers/kimi-k2p6-fast": {
3982
4125
  id: "accounts/fireworks/routers/kimi-k2p6-fast",
3983
4126
  name: "Kimi K2.6 Fast",
@@ -4024,7 +4167,7 @@ export const MODELS = {
4024
4167
  reasoning: true,
4025
4168
  input: ["text", "image"],
4026
4169
  cost: {
4027
- input: 2,
4170
+ input: 1.9,
4028
4171
  output: 8,
4029
4172
  cacheRead: 0.38,
4030
4173
  cacheWrite: 0,
@@ -4206,6 +4349,25 @@ export const MODELS = {
4206
4349
  contextWindow: 1000000,
4207
4350
  maxTokens: 32000,
4208
4351
  },
4352
+ "claude-sonnet-5": {
4353
+ id: "claude-sonnet-5",
4354
+ name: "Claude Sonnet 5",
4355
+ api: "openai-completions",
4356
+ provider: "github-copilot",
4357
+ baseUrl: "https://api.individual.githubcopilot.com",
4358
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4359
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4360
+ reasoning: true,
4361
+ input: ["text", "image"],
4362
+ cost: {
4363
+ input: 2,
4364
+ output: 10,
4365
+ cacheRead: 0.2,
4366
+ cacheWrite: 2.5,
4367
+ },
4368
+ contextWindow: 1000000,
4369
+ maxTokens: 128000,
4370
+ },
4209
4371
  "gemini-2.5-pro": {
4210
4372
  id: "gemini-2.5-pro",
4211
4373
  name: "Gemini 2.5 Pro",
@@ -4260,7 +4422,7 @@ export const MODELS = {
4260
4422
  cacheRead: 0.2,
4261
4423
  cacheWrite: 0,
4262
4424
  },
4263
- contextWindow: 200000,
4425
+ contextWindow: 1000000,
4264
4426
  maxTokens: 64000,
4265
4427
  },
4266
4428
  "gemini-3.5-flash": {
@@ -4393,7 +4555,7 @@ export const MODELS = {
4393
4555
  cacheRead: 0.25,
4394
4556
  cacheWrite: 0,
4395
4557
  },
4396
- contextWindow: 400000,
4558
+ contextWindow: 1050000,
4397
4559
  maxTokens: 128000,
4398
4560
  },
4399
4561
  "gpt-5.4-mini": {
@@ -4450,30 +4612,106 @@ export const MODELS = {
4450
4612
  cacheRead: 0.5,
4451
4613
  cacheWrite: 0,
4452
4614
  },
4453
- contextWindow: 400000,
4615
+ contextWindow: 1050000,
4454
4616
  maxTokens: 128000,
4455
4617
  },
4456
- "raptor-mini": {
4457
- id: "raptor-mini",
4458
- name: "Raptor mini",
4459
- api: "openai-completions",
4618
+ "gpt-5.6-luna": {
4619
+ id: "gpt-5.6-luna",
4620
+ name: "GPT-5.6 Luna",
4621
+ api: "openai-responses",
4460
4622
  provider: "github-copilot",
4461
4623
  baseUrl: "https://api.individual.githubcopilot.com",
4462
4624
  headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4463
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4464
4625
  reasoning: true,
4626
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4465
4627
  input: ["text", "image"],
4466
4628
  cost: {
4467
- input: 0.25,
4468
- output: 2,
4469
- cacheRead: 0.025,
4629
+ input: 1,
4630
+ output: 6,
4631
+ cacheRead: 0.1,
4470
4632
  cacheWrite: 0,
4471
4633
  },
4472
- contextWindow: 400000,
4634
+ contextWindow: 1050000,
4473
4635
  maxTokens: 128000,
4474
4636
  },
4475
- },
4476
- "google": {
4637
+ "gpt-5.6-sol": {
4638
+ id: "gpt-5.6-sol",
4639
+ name: "GPT-5.6 Sol",
4640
+ api: "openai-responses",
4641
+ provider: "github-copilot",
4642
+ baseUrl: "https://api.individual.githubcopilot.com",
4643
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4644
+ reasoning: true,
4645
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4646
+ input: ["text", "image"],
4647
+ cost: {
4648
+ input: 5,
4649
+ output: 30,
4650
+ cacheRead: 0.5,
4651
+ cacheWrite: 0,
4652
+ },
4653
+ contextWindow: 1050000,
4654
+ maxTokens: 128000,
4655
+ },
4656
+ "gpt-5.6-terra": {
4657
+ id: "gpt-5.6-terra",
4658
+ name: "GPT-5.6 Terra",
4659
+ api: "openai-responses",
4660
+ provider: "github-copilot",
4661
+ baseUrl: "https://api.individual.githubcopilot.com",
4662
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4663
+ reasoning: true,
4664
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4665
+ input: ["text", "image"],
4666
+ cost: {
4667
+ input: 2.5,
4668
+ output: 15,
4669
+ cacheRead: 0.25,
4670
+ cacheWrite: 0,
4671
+ },
4672
+ contextWindow: 1050000,
4673
+ maxTokens: 128000,
4674
+ },
4675
+ "kimi-k2.7-code": {
4676
+ id: "kimi-k2.7-code",
4677
+ name: "Kimi K2.7 Code",
4678
+ api: "openai-completions",
4679
+ provider: "github-copilot",
4680
+ baseUrl: "https://api.individual.githubcopilot.com",
4681
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4682
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4683
+ reasoning: true,
4684
+ input: ["text", "image"],
4685
+ cost: {
4686
+ input: 0.95,
4687
+ output: 4,
4688
+ cacheRead: 0.19,
4689
+ cacheWrite: 0,
4690
+ },
4691
+ contextWindow: 256000,
4692
+ maxTokens: 32000,
4693
+ },
4694
+ "mai-code-1-flash-picker": {
4695
+ id: "mai-code-1-flash-picker",
4696
+ name: "MAI-Code-1-Flash",
4697
+ api: "openai-completions",
4698
+ provider: "github-copilot",
4699
+ baseUrl: "https://api.individual.githubcopilot.com",
4700
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4701
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4702
+ reasoning: true,
4703
+ input: ["text"],
4704
+ cost: {
4705
+ input: 0.75,
4706
+ output: 4.5,
4707
+ cacheRead: 0.075,
4708
+ cacheWrite: 0,
4709
+ },
4710
+ contextWindow: 256000,
4711
+ maxTokens: 128000,
4712
+ },
4713
+ },
4714
+ "google": {
4477
4715
  "gemini-2.0-flash": {
4478
4716
  id: "gemini-2.0-flash",
4479
4717
  name: "Gemini 2.0 Flash",
@@ -4755,42 +4993,6 @@ export const MODELS = {
4755
4993
  contextWindow: 262144,
4756
4994
  maxTokens: 32768,
4757
4995
  },
4758
- "gemma-4-E2B-it": {
4759
- id: "gemma-4-E2B-it",
4760
- name: "Gemma 4 E2B IT",
4761
- api: "google-generative-ai",
4762
- provider: "google",
4763
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4764
- reasoning: true,
4765
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4766
- input: ["text", "image"],
4767
- cost: {
4768
- input: 0,
4769
- output: 0,
4770
- cacheRead: 0,
4771
- cacheWrite: 0,
4772
- },
4773
- contextWindow: 131072,
4774
- maxTokens: 8192,
4775
- },
4776
- "gemma-4-E4B-it": {
4777
- id: "gemma-4-E4B-it",
4778
- name: "Gemma 4 E4B IT",
4779
- api: "google-generative-ai",
4780
- provider: "google",
4781
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4782
- reasoning: true,
4783
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4784
- input: ["text", "image"],
4785
- cost: {
4786
- input: 0,
4787
- output: 0,
4788
- cacheRead: 0,
4789
- cacheWrite: 0,
4790
- },
4791
- contextWindow: 131072,
4792
- maxTokens: 8192,
4793
- },
4794
4996
  },
4795
4997
  "google-vertex": {
4796
4998
  "gemini-1.5-flash": {
@@ -5116,7 +5318,7 @@ export const MODELS = {
5116
5318
  cost: {
5117
5319
  input: 0.075,
5118
5320
  output: 0.3,
5119
- cacheRead: 0.037,
5321
+ cacheRead: 0,
5120
5322
  cacheWrite: 0,
5121
5323
  },
5122
5324
  contextWindow: 131072,
@@ -5142,6 +5344,24 @@ export const MODELS = {
5142
5344
  },
5143
5345
  },
5144
5346
  "huggingface": {
5347
+ "MiniMaxAI/MiniMax-M2": {
5348
+ id: "MiniMaxAI/MiniMax-M2",
5349
+ name: "MiniMax-M2",
5350
+ api: "openai-completions",
5351
+ provider: "huggingface",
5352
+ baseUrl: "https://router.huggingface.co/v1",
5353
+ compat: { "supportsDeveloperRole": false },
5354
+ reasoning: true,
5355
+ input: ["text"],
5356
+ cost: {
5357
+ input: 0.3,
5358
+ output: 1.2,
5359
+ cacheRead: 0,
5360
+ cacheWrite: 0,
5361
+ },
5362
+ contextWindow: 204800,
5363
+ maxTokens: 128000,
5364
+ },
5145
5365
  "MiniMaxAI/MiniMax-M2.1": {
5146
5366
  id: "MiniMaxAI/MiniMax-M2.1",
5147
5367
  name: "MiniMax-M2.1",
@@ -5196,6 +5416,42 @@ export const MODELS = {
5196
5416
  contextWindow: 204800,
5197
5417
  maxTokens: 131072,
5198
5418
  },
5419
+ "MiniMaxAI/MiniMax-M3": {
5420
+ id: "MiniMaxAI/MiniMax-M3",
5421
+ name: "MiniMax-M3",
5422
+ api: "openai-completions",
5423
+ provider: "huggingface",
5424
+ baseUrl: "https://router.huggingface.co/v1",
5425
+ compat: { "supportsDeveloperRole": false },
5426
+ reasoning: true,
5427
+ input: ["text", "image"],
5428
+ cost: {
5429
+ input: 0.3,
5430
+ output: 1.2,
5431
+ cacheRead: 0,
5432
+ cacheWrite: 0,
5433
+ },
5434
+ contextWindow: 524288,
5435
+ maxTokens: 128000,
5436
+ },
5437
+ "Qwen/Qwen3-235B-A22B": {
5438
+ id: "Qwen/Qwen3-235B-A22B",
5439
+ name: "Qwen3 235B-A22B",
5440
+ api: "openai-completions",
5441
+ provider: "huggingface",
5442
+ baseUrl: "https://router.huggingface.co/v1",
5443
+ compat: { "supportsDeveloperRole": false },
5444
+ reasoning: true,
5445
+ input: ["text"],
5446
+ cost: {
5447
+ input: 0.2,
5448
+ output: 0.8,
5449
+ cacheRead: 0,
5450
+ cacheWrite: 0,
5451
+ },
5452
+ contextWindow: 40960,
5453
+ maxTokens: 16384,
5454
+ },
5199
5455
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5200
5456
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5201
5457
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5214,6 +5470,42 @@ export const MODELS = {
5214
5470
  contextWindow: 262144,
5215
5471
  maxTokens: 131072,
5216
5472
  },
5473
+ "Qwen/Qwen3-32B": {
5474
+ id: "Qwen/Qwen3-32B",
5475
+ name: "Qwen3 32B",
5476
+ api: "openai-completions",
5477
+ provider: "huggingface",
5478
+ baseUrl: "https://router.huggingface.co/v1",
5479
+ compat: { "supportsDeveloperRole": false },
5480
+ reasoning: true,
5481
+ input: ["text"],
5482
+ cost: {
5483
+ input: 0.29,
5484
+ output: 0.59,
5485
+ cacheRead: 0,
5486
+ cacheWrite: 0,
5487
+ },
5488
+ contextWindow: 131072,
5489
+ maxTokens: 16384,
5490
+ },
5491
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
5492
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
5493
+ name: "Qwen3-Coder 30B-A3B Instruct",
5494
+ api: "openai-completions",
5495
+ provider: "huggingface",
5496
+ baseUrl: "https://router.huggingface.co/v1",
5497
+ compat: { "supportsDeveloperRole": false },
5498
+ reasoning: false,
5499
+ input: ["text"],
5500
+ cost: {
5501
+ input: 0.07,
5502
+ output: 0.26,
5503
+ cacheRead: 0,
5504
+ cacheWrite: 0,
5505
+ },
5506
+ contextWindow: 262144,
5507
+ maxTokens: 65536,
5508
+ },
5217
5509
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5218
5510
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5219
5511
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5286,9 +5578,9 @@ export const MODELS = {
5286
5578
  contextWindow: 262144,
5287
5579
  maxTokens: 131072,
5288
5580
  },
5289
- "Qwen/Qwen3.5-397B-A17B": {
5290
- id: "Qwen/Qwen3.5-397B-A17B",
5291
- name: "Qwen3.5-397B-A17B",
5581
+ "Qwen/Qwen3.5-122B-A10B": {
5582
+ id: "Qwen/Qwen3.5-122B-A10B",
5583
+ name: "Qwen3.5 122B-A10B",
5292
5584
  api: "openai-completions",
5293
5585
  provider: "huggingface",
5294
5586
  baseUrl: "https://router.huggingface.co/v1",
@@ -5296,103 +5588,319 @@ export const MODELS = {
5296
5588
  reasoning: true,
5297
5589
  input: ["text", "image"],
5298
5590
  cost: {
5299
- input: 0.6,
5300
- output: 3.6,
5591
+ input: 0.4,
5592
+ output: 3.2,
5301
5593
  cacheRead: 0,
5302
5594
  cacheWrite: 0,
5303
5595
  },
5304
5596
  contextWindow: 262144,
5305
- maxTokens: 32768,
5597
+ maxTokens: 65536,
5306
5598
  },
5307
- "XiaomiMiMo/MiMo-V2-Flash": {
5308
- id: "XiaomiMiMo/MiMo-V2-Flash",
5309
- name: "MiMo-V2-Flash",
5599
+ "Qwen/Qwen3.5-27B": {
5600
+ id: "Qwen/Qwen3.5-27B",
5601
+ name: "Qwen3.5 27B",
5310
5602
  api: "openai-completions",
5311
5603
  provider: "huggingface",
5312
5604
  baseUrl: "https://router.huggingface.co/v1",
5313
5605
  compat: { "supportsDeveloperRole": false },
5314
5606
  reasoning: true,
5315
- input: ["text"],
5607
+ input: ["text", "image"],
5316
5608
  cost: {
5317
- input: 0.1,
5318
- output: 0.3,
5609
+ input: 0.3,
5610
+ output: 2.4,
5319
5611
  cacheRead: 0,
5320
5612
  cacheWrite: 0,
5321
5613
  },
5322
5614
  contextWindow: 262144,
5323
- maxTokens: 4096,
5615
+ maxTokens: 65536,
5324
5616
  },
5325
- "deepseek-ai/DeepSeek-R1-0528": {
5326
- id: "deepseek-ai/DeepSeek-R1-0528",
5327
- name: "DeepSeek-R1-0528",
5617
+ "Qwen/Qwen3.5-35B-A3B": {
5618
+ id: "Qwen/Qwen3.5-35B-A3B",
5619
+ name: "Qwen3.5 35B-A3B",
5328
5620
  api: "openai-completions",
5329
5621
  provider: "huggingface",
5330
5622
  baseUrl: "https://router.huggingface.co/v1",
5331
5623
  compat: { "supportsDeveloperRole": false },
5332
5624
  reasoning: true,
5333
- input: ["text"],
5625
+ input: ["text", "image"],
5334
5626
  cost: {
5335
- input: 3,
5336
- output: 5,
5627
+ input: 0.25,
5628
+ output: 2,
5337
5629
  cacheRead: 0,
5338
5630
  cacheWrite: 0,
5339
5631
  },
5340
- contextWindow: 163840,
5341
- maxTokens: 163840,
5632
+ contextWindow: 262144,
5633
+ maxTokens: 65536,
5342
5634
  },
5343
- "deepseek-ai/DeepSeek-V3.2": {
5344
- id: "deepseek-ai/DeepSeek-V3.2",
5345
- name: "DeepSeek-V3.2",
5635
+ "Qwen/Qwen3.5-397B-A17B": {
5636
+ id: "Qwen/Qwen3.5-397B-A17B",
5637
+ name: "Qwen3.5-397B-A17B",
5346
5638
  api: "openai-completions",
5347
5639
  provider: "huggingface",
5348
5640
  baseUrl: "https://router.huggingface.co/v1",
5349
5641
  compat: { "supportsDeveloperRole": false },
5350
5642
  reasoning: true,
5351
- input: ["text"],
5643
+ input: ["text", "image"],
5352
5644
  cost: {
5353
- input: 0.28,
5354
- output: 0.4,
5645
+ input: 0.6,
5646
+ output: 3.6,
5355
5647
  cacheRead: 0,
5356
5648
  cacheWrite: 0,
5357
5649
  },
5358
- contextWindow: 163840,
5359
- maxTokens: 65536,
5650
+ contextWindow: 262144,
5651
+ maxTokens: 32768,
5360
5652
  },
5361
- "deepseek-ai/DeepSeek-V4-Pro": {
5362
- id: "deepseek-ai/DeepSeek-V4-Pro",
5363
- name: "DeepSeek V4 Pro",
5653
+ "Qwen/Qwen3.5-9B": {
5654
+ id: "Qwen/Qwen3.5-9B",
5655
+ name: "Qwen3.5 9B",
5364
5656
  api: "openai-completions",
5365
5657
  provider: "huggingface",
5366
5658
  baseUrl: "https://router.huggingface.co/v1",
5367
5659
  compat: { "supportsDeveloperRole": false },
5368
5660
  reasoning: true,
5369
- input: ["text"],
5661
+ input: ["text", "image"],
5370
5662
  cost: {
5371
- input: 0.435,
5372
- output: 0.87,
5373
- cacheRead: 0.003625,
5663
+ input: 0.17,
5664
+ output: 0.25,
5665
+ cacheRead: 0,
5374
5666
  cacheWrite: 0,
5375
5667
  },
5376
- contextWindow: 1048576,
5377
- maxTokens: 393216,
5668
+ contextWindow: 262144,
5669
+ maxTokens: 65536,
5378
5670
  },
5379
- "moonshotai/Kimi-K2-Instruct": {
5380
- id: "moonshotai/Kimi-K2-Instruct",
5381
- name: "Kimi-K2-Instruct",
5671
+ "Qwen/Qwen3.6-27B": {
5672
+ id: "Qwen/Qwen3.6-27B",
5673
+ name: "Qwen3.6 27B",
5382
5674
  api: "openai-completions",
5383
5675
  provider: "huggingface",
5384
5676
  baseUrl: "https://router.huggingface.co/v1",
5385
5677
  compat: { "supportsDeveloperRole": false },
5386
- reasoning: false,
5387
- input: ["text"],
5678
+ reasoning: true,
5679
+ input: ["text", "image"],
5388
5680
  cost: {
5389
- input: 1,
5390
- output: 3,
5681
+ input: 0.47,
5682
+ output: 3.19,
5391
5683
  cacheRead: 0,
5392
5684
  cacheWrite: 0,
5393
5685
  },
5394
- contextWindow: 131072,
5395
- maxTokens: 16384,
5686
+ contextWindow: 262144,
5687
+ maxTokens: 65536,
5688
+ },
5689
+ "Qwen/Qwen3.6-35B-A3B": {
5690
+ id: "Qwen/Qwen3.6-35B-A3B",
5691
+ name: "Qwen3.6 35B-A3B",
5692
+ api: "openai-completions",
5693
+ provider: "huggingface",
5694
+ baseUrl: "https://router.huggingface.co/v1",
5695
+ compat: { "supportsDeveloperRole": false },
5696
+ reasoning: true,
5697
+ input: ["text", "image"],
5698
+ cost: {
5699
+ input: 0.15,
5700
+ output: 0.95,
5701
+ cacheRead: 0,
5702
+ cacheWrite: 0,
5703
+ },
5704
+ contextWindow: 262144,
5705
+ maxTokens: 65536,
5706
+ },
5707
+ "XiaomiMiMo/MiMo-V2-Flash": {
5708
+ id: "XiaomiMiMo/MiMo-V2-Flash",
5709
+ name: "MiMo-V2-Flash",
5710
+ api: "openai-completions",
5711
+ provider: "huggingface",
5712
+ baseUrl: "https://router.huggingface.co/v1",
5713
+ compat: { "supportsDeveloperRole": false },
5714
+ reasoning: true,
5715
+ input: ["text"],
5716
+ cost: {
5717
+ input: 0.1,
5718
+ output: 0.3,
5719
+ cacheRead: 0,
5720
+ cacheWrite: 0,
5721
+ },
5722
+ contextWindow: 262144,
5723
+ maxTokens: 4096,
5724
+ },
5725
+ "XiaomiMiMo/MiMo-V2.5-Pro": {
5726
+ id: "XiaomiMiMo/MiMo-V2.5-Pro",
5727
+ name: "MiMo-V2.5-Pro",
5728
+ api: "openai-completions",
5729
+ provider: "huggingface",
5730
+ baseUrl: "https://router.huggingface.co/v1",
5731
+ compat: { "supportsDeveloperRole": false },
5732
+ reasoning: true,
5733
+ input: ["text"],
5734
+ cost: {
5735
+ input: 1,
5736
+ output: 3,
5737
+ cacheRead: 0,
5738
+ cacheWrite: 0,
5739
+ },
5740
+ contextWindow: 1048576,
5741
+ maxTokens: 131072,
5742
+ },
5743
+ "deepseek-ai/DeepSeek-R1": {
5744
+ id: "deepseek-ai/DeepSeek-R1",
5745
+ name: "DeepSeek-R1",
5746
+ api: "openai-completions",
5747
+ provider: "huggingface",
5748
+ baseUrl: "https://router.huggingface.co/v1",
5749
+ compat: { "supportsDeveloperRole": false },
5750
+ reasoning: true,
5751
+ input: ["text"],
5752
+ cost: {
5753
+ input: 0.7,
5754
+ output: 2.5,
5755
+ cacheRead: 0,
5756
+ cacheWrite: 0,
5757
+ },
5758
+ contextWindow: 64000,
5759
+ maxTokens: 32768,
5760
+ },
5761
+ "deepseek-ai/DeepSeek-R1-0528": {
5762
+ id: "deepseek-ai/DeepSeek-R1-0528",
5763
+ name: "DeepSeek-R1-0528",
5764
+ api: "openai-completions",
5765
+ provider: "huggingface",
5766
+ baseUrl: "https://router.huggingface.co/v1",
5767
+ compat: { "supportsDeveloperRole": false },
5768
+ reasoning: true,
5769
+ input: ["text"],
5770
+ cost: {
5771
+ input: 3,
5772
+ output: 5,
5773
+ cacheRead: 0,
5774
+ cacheWrite: 0,
5775
+ },
5776
+ contextWindow: 163840,
5777
+ maxTokens: 163840,
5778
+ },
5779
+ "deepseek-ai/DeepSeek-V3.2": {
5780
+ id: "deepseek-ai/DeepSeek-V3.2",
5781
+ name: "DeepSeek-V3.2",
5782
+ api: "openai-completions",
5783
+ provider: "huggingface",
5784
+ baseUrl: "https://router.huggingface.co/v1",
5785
+ compat: { "supportsDeveloperRole": false },
5786
+ reasoning: true,
5787
+ input: ["text"],
5788
+ cost: {
5789
+ input: 0.28,
5790
+ output: 0.4,
5791
+ cacheRead: 0,
5792
+ cacheWrite: 0,
5793
+ },
5794
+ contextWindow: 163840,
5795
+ maxTokens: 65536,
5796
+ },
5797
+ "deepseek-ai/DeepSeek-V4-Flash": {
5798
+ id: "deepseek-ai/DeepSeek-V4-Flash",
5799
+ name: "DeepSeek V4 Flash",
5800
+ api: "openai-completions",
5801
+ provider: "huggingface",
5802
+ baseUrl: "https://router.huggingface.co/v1",
5803
+ compat: { "supportsDeveloperRole": false },
5804
+ reasoning: true,
5805
+ input: ["text"],
5806
+ cost: {
5807
+ input: 0.14,
5808
+ output: 0.28,
5809
+ cacheRead: 0,
5810
+ cacheWrite: 0,
5811
+ },
5812
+ contextWindow: 1048576,
5813
+ maxTokens: 384000,
5814
+ },
5815
+ "deepseek-ai/DeepSeek-V4-Pro": {
5816
+ id: "deepseek-ai/DeepSeek-V4-Pro",
5817
+ name: "DeepSeek V4 Pro",
5818
+ api: "openai-completions",
5819
+ provider: "huggingface",
5820
+ baseUrl: "https://router.huggingface.co/v1",
5821
+ compat: { "supportsDeveloperRole": false },
5822
+ reasoning: true,
5823
+ input: ["text"],
5824
+ cost: {
5825
+ input: 0.435,
5826
+ output: 0.87,
5827
+ cacheRead: 0.003625,
5828
+ cacheWrite: 0,
5829
+ },
5830
+ contextWindow: 1048576,
5831
+ maxTokens: 393216,
5832
+ },
5833
+ "google/gemma-4-26B-A4B-it": {
5834
+ id: "google/gemma-4-26B-A4B-it",
5835
+ name: "Gemma 4 26B A4B IT",
5836
+ api: "openai-completions",
5837
+ provider: "huggingface",
5838
+ baseUrl: "https://router.huggingface.co/v1",
5839
+ compat: { "supportsDeveloperRole": false },
5840
+ reasoning: true,
5841
+ input: ["text", "image"],
5842
+ cost: {
5843
+ input: 0.13,
5844
+ output: 0.4,
5845
+ cacheRead: 0,
5846
+ cacheWrite: 0,
5847
+ },
5848
+ contextWindow: 262144,
5849
+ maxTokens: 32768,
5850
+ },
5851
+ "google/gemma-4-31B-it": {
5852
+ id: "google/gemma-4-31B-it",
5853
+ name: "Gemma 4 31B IT",
5854
+ api: "openai-completions",
5855
+ provider: "huggingface",
5856
+ baseUrl: "https://router.huggingface.co/v1",
5857
+ compat: { "supportsDeveloperRole": false },
5858
+ reasoning: true,
5859
+ input: ["text", "image"],
5860
+ cost: {
5861
+ input: 0.14,
5862
+ output: 0.4,
5863
+ cacheRead: 0,
5864
+ cacheWrite: 0,
5865
+ },
5866
+ contextWindow: 262144,
5867
+ maxTokens: 32768,
5868
+ },
5869
+ "meta-llama/Llama-3.3-70B-Instruct": {
5870
+ id: "meta-llama/Llama-3.3-70B-Instruct",
5871
+ name: "Llama-3.3-70B-Instruct",
5872
+ api: "openai-completions",
5873
+ provider: "huggingface",
5874
+ baseUrl: "https://router.huggingface.co/v1",
5875
+ compat: { "supportsDeveloperRole": false },
5876
+ reasoning: false,
5877
+ input: ["text"],
5878
+ cost: {
5879
+ input: 0.59,
5880
+ output: 0.79,
5881
+ cacheRead: 0,
5882
+ cacheWrite: 0,
5883
+ },
5884
+ contextWindow: 131072,
5885
+ maxTokens: 4096,
5886
+ },
5887
+ "moonshotai/Kimi-K2-Instruct": {
5888
+ id: "moonshotai/Kimi-K2-Instruct",
5889
+ name: "Kimi-K2-Instruct",
5890
+ api: "openai-completions",
5891
+ provider: "huggingface",
5892
+ baseUrl: "https://router.huggingface.co/v1",
5893
+ compat: { "supportsDeveloperRole": false },
5894
+ reasoning: false,
5895
+ input: ["text"],
5896
+ cost: {
5897
+ input: 1,
5898
+ output: 3,
5899
+ cacheRead: 0,
5900
+ cacheWrite: 0,
5901
+ },
5902
+ contextWindow: 131072,
5903
+ maxTokens: 16384,
5396
5904
  },
5397
5905
  "moonshotai/Kimi-K2-Instruct-0905": {
5398
5906
  id: "moonshotai/Kimi-K2-Instruct-0905",
@@ -5466,6 +5974,168 @@ export const MODELS = {
5466
5974
  contextWindow: 262144,
5467
5975
  maxTokens: 262144,
5468
5976
  },
5977
+ "moonshotai/Kimi-K2.7-Code": {
5978
+ id: "moonshotai/Kimi-K2.7-Code",
5979
+ name: "Kimi K2.7 Code",
5980
+ api: "openai-completions",
5981
+ provider: "huggingface",
5982
+ baseUrl: "https://router.huggingface.co/v1",
5983
+ compat: { "supportsDeveloperRole": false },
5984
+ reasoning: true,
5985
+ input: ["text", "image"],
5986
+ cost: {
5987
+ input: 0.95,
5988
+ output: 4,
5989
+ cacheRead: 0,
5990
+ cacheWrite: 0,
5991
+ },
5992
+ contextWindow: 262144,
5993
+ maxTokens: 262144,
5994
+ },
5995
+ "openai/gpt-oss-120b": {
5996
+ id: "openai/gpt-oss-120b",
5997
+ name: "GPT OSS 120B",
5998
+ api: "openai-completions",
5999
+ provider: "huggingface",
6000
+ baseUrl: "https://router.huggingface.co/v1",
6001
+ compat: { "supportsDeveloperRole": false },
6002
+ reasoning: true,
6003
+ input: ["text"],
6004
+ cost: {
6005
+ input: 0.25,
6006
+ output: 0.69,
6007
+ cacheRead: 0,
6008
+ cacheWrite: 0,
6009
+ },
6010
+ contextWindow: 131072,
6011
+ maxTokens: 32768,
6012
+ },
6013
+ "openai/gpt-oss-20b": {
6014
+ id: "openai/gpt-oss-20b",
6015
+ name: "GPT OSS 20B",
6016
+ api: "openai-completions",
6017
+ provider: "huggingface",
6018
+ baseUrl: "https://router.huggingface.co/v1",
6019
+ compat: { "supportsDeveloperRole": false },
6020
+ reasoning: true,
6021
+ input: ["text"],
6022
+ cost: {
6023
+ input: 0.1,
6024
+ output: 0.5,
6025
+ cacheRead: 0,
6026
+ cacheWrite: 0,
6027
+ },
6028
+ contextWindow: 131072,
6029
+ maxTokens: 32768,
6030
+ },
6031
+ "stepfun-ai/Step-3.5-Flash": {
6032
+ id: "stepfun-ai/Step-3.5-Flash",
6033
+ name: "Step 3.5 Flash",
6034
+ api: "openai-completions",
6035
+ provider: "huggingface",
6036
+ baseUrl: "https://router.huggingface.co/v1",
6037
+ compat: { "supportsDeveloperRole": false },
6038
+ reasoning: true,
6039
+ input: ["text"],
6040
+ cost: {
6041
+ input: 0.1,
6042
+ output: 0.3,
6043
+ cacheRead: 0,
6044
+ cacheWrite: 0,
6045
+ },
6046
+ contextWindow: 262144,
6047
+ maxTokens: 256000,
6048
+ },
6049
+ "stepfun-ai/Step-3.7-Flash": {
6050
+ id: "stepfun-ai/Step-3.7-Flash",
6051
+ name: "Step 3.7 Flash",
6052
+ api: "openai-completions",
6053
+ provider: "huggingface",
6054
+ baseUrl: "https://router.huggingface.co/v1",
6055
+ compat: { "supportsDeveloperRole": false },
6056
+ reasoning: true,
6057
+ input: ["text", "image"],
6058
+ cost: {
6059
+ input: 0.2,
6060
+ output: 1.15,
6061
+ cacheRead: 0,
6062
+ cacheWrite: 0,
6063
+ },
6064
+ contextWindow: 262144,
6065
+ maxTokens: 256000,
6066
+ },
6067
+ "zai-org/GLM-4.5": {
6068
+ id: "zai-org/GLM-4.5",
6069
+ name: "GLM-4.5",
6070
+ api: "openai-completions",
6071
+ provider: "huggingface",
6072
+ baseUrl: "https://router.huggingface.co/v1",
6073
+ compat: { "supportsDeveloperRole": false },
6074
+ reasoning: true,
6075
+ input: ["text"],
6076
+ cost: {
6077
+ input: 0.6,
6078
+ output: 2.2,
6079
+ cacheRead: 0,
6080
+ cacheWrite: 0,
6081
+ },
6082
+ contextWindow: 131072,
6083
+ maxTokens: 98304,
6084
+ },
6085
+ "zai-org/GLM-4.5-Air": {
6086
+ id: "zai-org/GLM-4.5-Air",
6087
+ name: "GLM-4.5-Air",
6088
+ api: "openai-completions",
6089
+ provider: "huggingface",
6090
+ baseUrl: "https://router.huggingface.co/v1",
6091
+ compat: { "supportsDeveloperRole": false },
6092
+ reasoning: true,
6093
+ input: ["text"],
6094
+ cost: {
6095
+ input: 0.13,
6096
+ output: 0.85,
6097
+ cacheRead: 0,
6098
+ cacheWrite: 0,
6099
+ },
6100
+ contextWindow: 131072,
6101
+ maxTokens: 98304,
6102
+ },
6103
+ "zai-org/GLM-4.5V": {
6104
+ id: "zai-org/GLM-4.5V",
6105
+ name: "GLM-4.5V",
6106
+ api: "openai-completions",
6107
+ provider: "huggingface",
6108
+ baseUrl: "https://router.huggingface.co/v1",
6109
+ compat: { "supportsDeveloperRole": false },
6110
+ reasoning: true,
6111
+ input: ["text", "image"],
6112
+ cost: {
6113
+ input: 0.6,
6114
+ output: 1.8,
6115
+ cacheRead: 0,
6116
+ cacheWrite: 0,
6117
+ },
6118
+ contextWindow: 65536,
6119
+ maxTokens: 16384,
6120
+ },
6121
+ "zai-org/GLM-4.6": {
6122
+ id: "zai-org/GLM-4.6",
6123
+ name: "GLM-4.6",
6124
+ api: "openai-completions",
6125
+ provider: "huggingface",
6126
+ baseUrl: "https://router.huggingface.co/v1",
6127
+ compat: { "supportsDeveloperRole": false },
6128
+ reasoning: true,
6129
+ input: ["text"],
6130
+ cost: {
6131
+ input: 0.55,
6132
+ output: 2.2,
6133
+ cacheRead: 0,
6134
+ cacheWrite: 0,
6135
+ },
6136
+ contextWindow: 204800,
6137
+ maxTokens: 131072,
6138
+ },
5469
6139
  "zai-org/GLM-4.7": {
5470
6140
  id: "zai-org/GLM-4.7",
5471
6141
  name: "GLM-4.7",
@@ -5520,9 +6190,27 @@ export const MODELS = {
5520
6190
  contextWindow: 202752,
5521
6191
  maxTokens: 131072,
5522
6192
  },
5523
- "zai-org/GLM-5.1": {
5524
- id: "zai-org/GLM-5.1",
5525
- name: "GLM-5.1",
6193
+ "zai-org/GLM-5.1": {
6194
+ id: "zai-org/GLM-5.1",
6195
+ name: "GLM-5.1",
6196
+ api: "openai-completions",
6197
+ provider: "huggingface",
6198
+ baseUrl: "https://router.huggingface.co/v1",
6199
+ compat: { "supportsDeveloperRole": false },
6200
+ reasoning: true,
6201
+ input: ["text"],
6202
+ cost: {
6203
+ input: 1,
6204
+ output: 3.2,
6205
+ cacheRead: 0.2,
6206
+ cacheWrite: 0,
6207
+ },
6208
+ contextWindow: 202752,
6209
+ maxTokens: 131072,
6210
+ },
6211
+ "zai-org/GLM-5.2": {
6212
+ id: "zai-org/GLM-5.2",
6213
+ name: "GLM-5.2",
5526
6214
  api: "openai-completions",
5527
6215
  provider: "huggingface",
5528
6216
  baseUrl: "https://router.huggingface.co/v1",
@@ -5530,12 +6218,12 @@ export const MODELS = {
5530
6218
  reasoning: true,
5531
6219
  input: ["text"],
5532
6220
  cost: {
5533
- input: 1,
5534
- output: 3.2,
5535
- cacheRead: 0.2,
6221
+ input: 1.4,
6222
+ output: 4.4,
6223
+ cacheRead: 0,
5536
6224
  cacheWrite: 0,
5537
6225
  },
5538
- contextWindow: 202752,
6226
+ contextWindow: 262144,
5539
6227
  maxTokens: 131072,
5540
6228
  },
5541
6229
  },
@@ -5997,11 +6685,11 @@ export const MODELS = {
5997
6685
  api: "mistral-conversations",
5998
6686
  provider: "mistral",
5999
6687
  baseUrl: "https://api.mistral.ai",
6000
- reasoning: false,
6688
+ reasoning: true,
6001
6689
  input: ["text", "image"],
6002
6690
  cost: {
6003
- input: 0.4,
6004
- output: 2,
6691
+ input: 1.5,
6692
+ output: 7.5,
6005
6693
  cacheRead: 0,
6006
6694
  cacheWrite: 0,
6007
6695
  },
@@ -7110,6 +7798,95 @@ export const MODELS = {
7110
7798
  contextWindow: 1050000,
7111
7799
  maxTokens: 128000,
7112
7800
  },
7801
+ "gpt-5.6": {
7802
+ id: "gpt-5.6",
7803
+ name: "GPT-5.6",
7804
+ api: "openai-responses",
7805
+ provider: "openai",
7806
+ baseUrl: "https://api.openai.com/v1",
7807
+ reasoning: true,
7808
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
7809
+ input: ["text", "image"],
7810
+ cost: {
7811
+ input: 5,
7812
+ output: 30,
7813
+ cacheRead: 0.5,
7814
+ cacheWrite: 6.25,
7815
+ },
7816
+ contextWindow: 1050000,
7817
+ maxTokens: 128000,
7818
+ },
7819
+ "gpt-5.6-luna": {
7820
+ id: "gpt-5.6-luna",
7821
+ name: "GPT-5.6 Luna",
7822
+ api: "openai-responses",
7823
+ provider: "openai",
7824
+ baseUrl: "https://api.openai.com/v1",
7825
+ reasoning: true,
7826
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
7827
+ input: ["text", "image"],
7828
+ cost: {
7829
+ input: 1,
7830
+ output: 6,
7831
+ cacheRead: 0.1,
7832
+ cacheWrite: 1.25,
7833
+ },
7834
+ contextWindow: 1050000,
7835
+ maxTokens: 128000,
7836
+ },
7837
+ "gpt-5.6-sol": {
7838
+ id: "gpt-5.6-sol",
7839
+ name: "GPT-5.6 Sol",
7840
+ api: "openai-responses",
7841
+ provider: "openai",
7842
+ baseUrl: "https://api.openai.com/v1",
7843
+ reasoning: true,
7844
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
7845
+ input: ["text", "image"],
7846
+ cost: {
7847
+ input: 5,
7848
+ output: 30,
7849
+ cacheRead: 0.5,
7850
+ cacheWrite: 6.25,
7851
+ },
7852
+ contextWindow: 1050000,
7853
+ maxTokens: 128000,
7854
+ },
7855
+ "gpt-5.6-terra": {
7856
+ id: "gpt-5.6-terra",
7857
+ name: "GPT-5.6 Terra",
7858
+ api: "openai-responses",
7859
+ provider: "openai",
7860
+ baseUrl: "https://api.openai.com/v1",
7861
+ reasoning: true,
7862
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
7863
+ input: ["text", "image"],
7864
+ cost: {
7865
+ input: 2.5,
7866
+ output: 15,
7867
+ cacheRead: 0.25,
7868
+ cacheWrite: 3.125,
7869
+ },
7870
+ contextWindow: 1050000,
7871
+ maxTokens: 128000,
7872
+ },
7873
+ "gpt-realtime-2.1": {
7874
+ id: "gpt-realtime-2.1",
7875
+ name: "GPT-Realtime-2.1",
7876
+ api: "openai-responses",
7877
+ provider: "openai",
7878
+ baseUrl: "https://api.openai.com/v1",
7879
+ reasoning: true,
7880
+ input: ["text", "image"],
7881
+ cost: {
7882
+ input: 4,
7883
+ output: 24,
7884
+ cacheRead: 0.4,
7885
+ cacheWrite: 0,
7886
+ },
7887
+ contextWindow: 128000,
7888
+ maxTokens: 32000,
7889
+ },
7113
7890
  "o1": {
7114
7891
  id: "o1",
7115
7892
  name: "o1",
@@ -7356,6 +8133,78 @@ export const MODELS = {
7356
8133
  contextWindow: 272000,
7357
8134
  maxTokens: 128000,
7358
8135
  },
8136
+ "gpt-5.6": {
8137
+ id: "gpt-5.6",
8138
+ name: "GPT-5.6",
8139
+ api: "openai-codex-responses",
8140
+ provider: "openai-codex",
8141
+ baseUrl: "https://chatgpt.com/backend-api",
8142
+ reasoning: true,
8143
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
8144
+ input: ["text", "image"],
8145
+ cost: {
8146
+ input: 5,
8147
+ output: 30,
8148
+ cacheRead: 0.5,
8149
+ cacheWrite: 6.25,
8150
+ },
8151
+ contextWindow: 272000,
8152
+ maxTokens: 128000,
8153
+ },
8154
+ "gpt-5.6-luna": {
8155
+ id: "gpt-5.6-luna",
8156
+ name: "GPT-5.6 Luna",
8157
+ api: "openai-codex-responses",
8158
+ provider: "openai-codex",
8159
+ baseUrl: "https://chatgpt.com/backend-api",
8160
+ reasoning: true,
8161
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
8162
+ input: ["text", "image"],
8163
+ cost: {
8164
+ input: 1,
8165
+ output: 6,
8166
+ cacheRead: 0.1,
8167
+ cacheWrite: 1.25,
8168
+ },
8169
+ contextWindow: 272000,
8170
+ maxTokens: 128000,
8171
+ },
8172
+ "gpt-5.6-sol": {
8173
+ id: "gpt-5.6-sol",
8174
+ name: "GPT-5.6 Sol",
8175
+ api: "openai-codex-responses",
8176
+ provider: "openai-codex",
8177
+ baseUrl: "https://chatgpt.com/backend-api",
8178
+ reasoning: true,
8179
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
8180
+ input: ["text", "image"],
8181
+ cost: {
8182
+ input: 5,
8183
+ output: 30,
8184
+ cacheRead: 0.5,
8185
+ cacheWrite: 6.25,
8186
+ },
8187
+ contextWindow: 272000,
8188
+ maxTokens: 128000,
8189
+ },
8190
+ "gpt-5.6-terra": {
8191
+ id: "gpt-5.6-terra",
8192
+ name: "GPT-5.6 Terra",
8193
+ api: "openai-codex-responses",
8194
+ provider: "openai-codex",
8195
+ baseUrl: "https://chatgpt.com/backend-api",
8196
+ reasoning: true,
8197
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
8198
+ input: ["text", "image"],
8199
+ cost: {
8200
+ input: 2.5,
8201
+ output: 15,
8202
+ cacheRead: 0.25,
8203
+ cacheWrite: 3.125,
8204
+ },
8205
+ contextWindow: 272000,
8206
+ maxTokens: 128000,
8207
+ },
7359
8208
  },
7360
8209
  "opencode": {
7361
8210
  "big-pickle": {
@@ -7375,6 +8224,23 @@ export const MODELS = {
7375
8224
  contextWindow: 200000,
7376
8225
  maxTokens: 32000,
7377
8226
  },
8227
+ "claude-fable-5": {
8228
+ id: "claude-fable-5",
8229
+ name: "Claude Fable 5",
8230
+ api: "anthropic-messages",
8231
+ provider: "opencode",
8232
+ baseUrl: "https://opencode.ai/zen",
8233
+ reasoning: true,
8234
+ input: ["text", "image"],
8235
+ cost: {
8236
+ input: 10,
8237
+ output: 50,
8238
+ cacheRead: 1,
8239
+ cacheWrite: 12.5,
8240
+ },
8241
+ contextWindow: 1000000,
8242
+ maxTokens: 128000,
8243
+ },
7378
8244
  "claude-haiku-4-5": {
7379
8245
  id: "claude-haiku-4-5",
7380
8246
  name: "Claude Haiku 4.5",
@@ -7534,6 +8400,23 @@ export const MODELS = {
7534
8400
  contextWindow: 1000000,
7535
8401
  maxTokens: 64000,
7536
8402
  },
8403
+ "claude-sonnet-5": {
8404
+ id: "claude-sonnet-5",
8405
+ name: "Claude Sonnet 5",
8406
+ api: "anthropic-messages",
8407
+ provider: "opencode",
8408
+ baseUrl: "https://opencode.ai/zen",
8409
+ reasoning: true,
8410
+ input: ["text", "image"],
8411
+ cost: {
8412
+ input: 2,
8413
+ output: 10,
8414
+ cacheRead: 0.2,
8415
+ cacheWrite: 2.5,
8416
+ },
8417
+ contextWindow: 1000000,
8418
+ maxTokens: 128000,
8419
+ },
7537
8420
  "deepseek-v4-flash": {
7538
8421
  id: "deepseek-v4-flash",
7539
8422
  name: "DeepSeek V4 Flash",
@@ -7679,6 +8562,23 @@ export const MODELS = {
7679
8562
  contextWindow: 204800,
7680
8563
  maxTokens: 131072,
7681
8564
  },
8565
+ "glm-5.2": {
8566
+ id: "glm-5.2",
8567
+ name: "GLM-5.2",
8568
+ api: "openai-completions",
8569
+ provider: "opencode",
8570
+ baseUrl: "https://opencode.ai/zen/v1",
8571
+ reasoning: true,
8572
+ input: ["text"],
8573
+ cost: {
8574
+ input: 1.4,
8575
+ output: 4.4,
8576
+ cacheRead: 0.26,
8577
+ cacheWrite: 0,
8578
+ },
8579
+ contextWindow: 1000000,
8580
+ maxTokens: 131072,
8581
+ },
7682
8582
  "gpt-5": {
7683
8583
  id: "gpt-5",
7684
8584
  name: "GPT-5",
@@ -7959,13 +8859,84 @@ export const MODELS = {
7959
8859
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "minimal": null, "low": null },
7960
8860
  input: ["text", "image"],
7961
8861
  cost: {
7962
- input: 30,
7963
- output: 180,
7964
- cacheRead: 30,
8862
+ input: 30,
8863
+ output: 180,
8864
+ cacheRead: 30,
8865
+ cacheWrite: 0,
8866
+ },
8867
+ contextWindow: 1050000,
8868
+ maxTokens: 128000,
8869
+ },
8870
+ "gpt-5.6-luna": {
8871
+ id: "gpt-5.6-luna",
8872
+ name: "GPT-5.6 Luna",
8873
+ api: "openai-responses",
8874
+ provider: "opencode",
8875
+ baseUrl: "https://opencode.ai/zen/v1",
8876
+ reasoning: true,
8877
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8878
+ input: ["text", "image"],
8879
+ cost: {
8880
+ input: 1,
8881
+ output: 6,
8882
+ cacheRead: 0.1,
8883
+ cacheWrite: 1.25,
8884
+ },
8885
+ contextWindow: 1050000,
8886
+ maxTokens: 128000,
8887
+ },
8888
+ "gpt-5.6-sol": {
8889
+ id: "gpt-5.6-sol",
8890
+ name: "GPT-5.6 Sol",
8891
+ api: "openai-responses",
8892
+ provider: "opencode",
8893
+ baseUrl: "https://opencode.ai/zen/v1",
8894
+ reasoning: true,
8895
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8896
+ input: ["text", "image"],
8897
+ cost: {
8898
+ input: 5,
8899
+ output: 30,
8900
+ cacheRead: 0.5,
8901
+ cacheWrite: 6.25,
8902
+ },
8903
+ contextWindow: 1050000,
8904
+ maxTokens: 128000,
8905
+ },
8906
+ "gpt-5.6-terra": {
8907
+ id: "gpt-5.6-terra",
8908
+ name: "GPT-5.6 Terra",
8909
+ api: "openai-responses",
8910
+ provider: "opencode",
8911
+ baseUrl: "https://opencode.ai/zen/v1",
8912
+ reasoning: true,
8913
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
8914
+ input: ["text", "image"],
8915
+ cost: {
8916
+ input: 2.5,
8917
+ output: 15,
8918
+ cacheRead: 0.25,
8919
+ cacheWrite: 3.125,
8920
+ },
8921
+ contextWindow: 1050000,
8922
+ maxTokens: 128000,
8923
+ },
8924
+ "grok-4.5": {
8925
+ id: "grok-4.5",
8926
+ name: "Grok 4.5",
8927
+ api: "openai-completions",
8928
+ provider: "opencode",
8929
+ baseUrl: "https://opencode.ai/zen/v1",
8930
+ reasoning: true,
8931
+ input: ["text", "image"],
8932
+ cost: {
8933
+ input: 2,
8934
+ output: 6,
8935
+ cacheRead: 0.5,
7965
8936
  cacheWrite: 0,
7966
8937
  },
7967
- contextWindow: 1050000,
7968
- maxTokens: 128000,
8938
+ contextWindow: 500000,
8939
+ maxTokens: 500000,
7969
8940
  },
7970
8941
  "grok-build-0.1": {
7971
8942
  id: "grok-build-0.1",
@@ -7986,6 +8957,23 @@ export const MODELS = {
7986
8957
  contextWindow: 256000,
7987
8958
  maxTokens: 256000,
7988
8959
  },
8960
+ "hy3-free": {
8961
+ id: "hy3-free",
8962
+ name: "Hy3 Free",
8963
+ api: "openai-completions",
8964
+ provider: "opencode",
8965
+ baseUrl: "https://opencode.ai/zen/v1",
8966
+ reasoning: true,
8967
+ input: ["text"],
8968
+ cost: {
8969
+ input: 0,
8970
+ output: 0,
8971
+ cacheRead: 0,
8972
+ cacheWrite: 0,
8973
+ },
8974
+ contextWindow: 190000,
8975
+ maxTokens: 64000,
8976
+ },
7989
8977
  "kimi-k2.5": {
7990
8978
  id: "kimi-k2.5",
7991
8979
  name: "Kimi K2.5",
@@ -8021,6 +9009,23 @@ export const MODELS = {
8021
9009
  contextWindow: 262144,
8022
9010
  maxTokens: 65536,
8023
9011
  },
9012
+ "kimi-k2.7-code": {
9013
+ id: "kimi-k2.7-code",
9014
+ name: "Kimi K2.7 Code",
9015
+ api: "openai-completions",
9016
+ provider: "opencode",
9017
+ baseUrl: "https://opencode.ai/zen/v1",
9018
+ reasoning: true,
9019
+ input: ["text", "image"],
9020
+ cost: {
9021
+ input: 0.95,
9022
+ output: 4,
9023
+ cacheRead: 0.19,
9024
+ cacheWrite: 0,
9025
+ },
9026
+ contextWindow: 262144,
9027
+ maxTokens: 262144,
9028
+ },
8024
9029
  "mimo-v2.5-free": {
8025
9030
  id: "mimo-v2.5-free",
8026
9031
  name: "MiMo V2.5 Free",
@@ -8040,7 +9045,7 @@ export const MODELS = {
8040
9045
  },
8041
9046
  "minimax-m2.5": {
8042
9047
  id: "minimax-m2.5",
8043
- name: "MiniMax M2.5",
9048
+ name: "MiniMax-M2.5",
8044
9049
  api: "openai-completions",
8045
9050
  provider: "opencode",
8046
9051
  baseUrl: "https://opencode.ai/zen/v1",
@@ -8057,7 +9062,7 @@ export const MODELS = {
8057
9062
  },
8058
9063
  "minimax-m2.7": {
8059
9064
  id: "minimax-m2.7",
8060
- name: "MiniMax M2.7",
9065
+ name: "MiniMax-M2.7",
8061
9066
  api: "openai-completions",
8062
9067
  provider: "opencode",
8063
9068
  baseUrl: "https://opencode.ai/zen/v1",
@@ -8072,6 +9077,23 @@ export const MODELS = {
8072
9077
  contextWindow: 204800,
8073
9078
  maxTokens: 131072,
8074
9079
  },
9080
+ "minimax-m3": {
9081
+ id: "minimax-m3",
9082
+ name: "MiniMax-M3",
9083
+ api: "openai-completions",
9084
+ provider: "opencode",
9085
+ baseUrl: "https://opencode.ai/zen/v1",
9086
+ reasoning: true,
9087
+ input: ["text", "image"],
9088
+ cost: {
9089
+ input: 0.3,
9090
+ output: 1.2,
9091
+ cacheRead: 0.06,
9092
+ cacheWrite: 0,
9093
+ },
9094
+ contextWindow: 512000,
9095
+ maxTokens: 128000,
9096
+ },
8075
9097
  "nemotron-3-ultra-free": {
8076
9098
  id: "nemotron-3-ultra-free",
8077
9099
  name: "Nemotron 3 Ultra Free",
@@ -8180,26 +9202,26 @@ export const MODELS = {
8180
9202
  contextWindow: 1000000,
8181
9203
  maxTokens: 384000,
8182
9204
  },
8183
- "glm-5": {
8184
- id: "glm-5",
8185
- name: "GLM-5",
9205
+ "glm-5.1": {
9206
+ id: "glm-5.1",
9207
+ name: "GLM-5.1",
8186
9208
  api: "openai-completions",
8187
9209
  provider: "opencode-go",
8188
9210
  baseUrl: "https://opencode.ai/zen/go/v1",
8189
9211
  reasoning: true,
8190
9212
  input: ["text"],
8191
9213
  cost: {
8192
- input: 1,
8193
- output: 3.2,
8194
- cacheRead: 0.2,
9214
+ input: 1.4,
9215
+ output: 4.4,
9216
+ cacheRead: 0.26,
8195
9217
  cacheWrite: 0,
8196
9218
  },
8197
9219
  contextWindow: 202752,
8198
9220
  maxTokens: 32768,
8199
9221
  },
8200
- "glm-5.1": {
8201
- id: "glm-5.1",
8202
- name: "GLM-5.1",
9222
+ "glm-5.2": {
9223
+ id: "glm-5.2",
9224
+ name: "GLM-5.2",
8203
9225
  api: "openai-completions",
8204
9226
  provider: "opencode-go",
8205
9227
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8211,8 +9233,8 @@ export const MODELS = {
8211
9233
  cacheRead: 0.26,
8212
9234
  cacheWrite: 0,
8213
9235
  },
8214
- contextWindow: 202752,
8215
- maxTokens: 32768,
9236
+ contextWindow: 1000000,
9237
+ maxTokens: 131072,
8216
9238
  },
8217
9239
  "kimi-k2.6": {
8218
9240
  id: "kimi-k2.6",
@@ -8286,7 +9308,7 @@ export const MODELS = {
8286
9308
  },
8287
9309
  "minimax-m2.7": {
8288
9310
  id: "minimax-m2.7",
8289
- name: "MiniMax M2.7",
9311
+ name: "MiniMax-M2.7",
8290
9312
  api: "openai-completions",
8291
9313
  provider: "opencode-go",
8292
9314
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8303,19 +9325,19 @@ export const MODELS = {
8303
9325
  },
8304
9326
  "minimax-m3": {
8305
9327
  id: "minimax-m3",
8306
- name: "MiniMax M3 (3x usage)",
9328
+ name: "MiniMax-M3",
8307
9329
  api: "anthropic-messages",
8308
9330
  provider: "opencode-go",
8309
9331
  baseUrl: "https://opencode.ai/zen/go",
8310
9332
  reasoning: true,
8311
9333
  input: ["text", "image"],
8312
9334
  cost: {
8313
- input: 0.1,
8314
- output: 0.4,
8315
- cacheRead: 0.02,
9335
+ input: 0.3,
9336
+ output: 1.2,
9337
+ cacheRead: 0.06,
8316
9338
  cacheWrite: 0,
8317
9339
  },
8318
- contextWindow: 512000,
9340
+ contextWindow: 1000000,
8319
9341
  maxTokens: 131072,
8320
9342
  },
8321
9343
  "qwen3.6-plus": {
@@ -8389,6 +9411,57 @@ export const MODELS = {
8389
9411
  contextWindow: 256000,
8390
9412
  maxTokens: 4096,
8391
9413
  },
9414
+ "aion-labs/aion-2.0": {
9415
+ id: "aion-labs/aion-2.0",
9416
+ name: "AionLabs: Aion-2.0",
9417
+ api: "openai-completions",
9418
+ provider: "openrouter",
9419
+ baseUrl: "https://openrouter.ai/api/v1",
9420
+ reasoning: true,
9421
+ input: ["text"],
9422
+ cost: {
9423
+ input: 0.7999999999999999,
9424
+ output: 1.5999999999999999,
9425
+ cacheRead: 0.19999999999999998,
9426
+ cacheWrite: 0,
9427
+ },
9428
+ contextWindow: 131072,
9429
+ maxTokens: 32768,
9430
+ },
9431
+ "aion-labs/aion-3.0": {
9432
+ id: "aion-labs/aion-3.0",
9433
+ name: "AionLabs: Aion-3.0",
9434
+ api: "openai-completions",
9435
+ provider: "openrouter",
9436
+ baseUrl: "https://openrouter.ai/api/v1",
9437
+ reasoning: true,
9438
+ input: ["text"],
9439
+ cost: {
9440
+ input: 3,
9441
+ output: 6,
9442
+ cacheRead: 0.75,
9443
+ cacheWrite: 0,
9444
+ },
9445
+ contextWindow: 131072,
9446
+ maxTokens: 32768,
9447
+ },
9448
+ "aion-labs/aion-3.0-mini": {
9449
+ id: "aion-labs/aion-3.0-mini",
9450
+ name: "AionLabs: Aion-3.0-Mini",
9451
+ api: "openai-completions",
9452
+ provider: "openrouter",
9453
+ baseUrl: "https://openrouter.ai/api/v1",
9454
+ reasoning: true,
9455
+ input: ["text"],
9456
+ cost: {
9457
+ input: 0.7,
9458
+ output: 1.4,
9459
+ cacheRead: 0.18,
9460
+ cacheWrite: 0,
9461
+ },
9462
+ contextWindow: 131072,
9463
+ maxTokens: 32768,
9464
+ },
8392
9465
  "amazon/nova-2-lite-v1": {
8393
9466
  id: "amazon/nova-2-lite-v1",
8394
9467
  name: "Amazon: Nova 2 Lite",
@@ -8491,23 +9564,6 @@ export const MODELS = {
8491
9564
  contextWindow: 200000,
8492
9565
  maxTokens: 4096,
8493
9566
  },
8494
- "anthropic/claude-3.5-haiku": {
8495
- id: "anthropic/claude-3.5-haiku",
8496
- name: "Anthropic: Claude 3.5 Haiku",
8497
- api: "openai-completions",
8498
- provider: "openrouter",
8499
- baseUrl: "https://openrouter.ai/api/v1",
8500
- reasoning: false,
8501
- input: ["text", "image"],
8502
- cost: {
8503
- input: 0.7999999999999999,
8504
- output: 4,
8505
- cacheRead: 0.08,
8506
- cacheWrite: 1,
8507
- },
8508
- contextWindow: 200000,
8509
- maxTokens: 8192,
8510
- },
8511
9567
  "anthropic/claude-fable-5": {
8512
9568
  id: "anthropic/claude-fable-5",
8513
9569
  name: "Anthropic: Claude Fable 5",
@@ -8611,24 +9667,6 @@ export const MODELS = {
8611
9667
  contextWindow: 1000000,
8612
9668
  maxTokens: 128000,
8613
9669
  },
8614
- "anthropic/claude-opus-4.6-fast": {
8615
- id: "anthropic/claude-opus-4.6-fast",
8616
- name: "Anthropic: Claude Opus 4.6 (Fast)",
8617
- api: "openai-completions",
8618
- provider: "openrouter",
8619
- baseUrl: "https://openrouter.ai/api/v1",
8620
- reasoning: true,
8621
- thinkingLevelMap: { "xhigh": "max" },
8622
- input: ["text", "image"],
8623
- cost: {
8624
- input: 30,
8625
- output: 150,
8626
- cacheRead: 3,
8627
- cacheWrite: 37.5,
8628
- },
8629
- contextWindow: 1000000,
8630
- maxTokens: 128000,
8631
- },
8632
9670
  "anthropic/claude-opus-4.7": {
8633
9671
  id: "anthropic/claude-opus-4.7",
8634
9672
  name: "Anthropic: Claude Opus 4.7",
@@ -8750,39 +9788,39 @@ export const MODELS = {
8750
9788
  contextWindow: 1000000,
8751
9789
  maxTokens: 128000,
8752
9790
  },
8753
- "arcee-ai/trinity-large-thinking": {
8754
- id: "arcee-ai/trinity-large-thinking",
8755
- name: "Arcee AI: Trinity Large Thinking",
9791
+ "anthropic/claude-sonnet-5": {
9792
+ id: "anthropic/claude-sonnet-5",
9793
+ name: "Anthropic: Claude Sonnet 5",
8756
9794
  api: "openai-completions",
8757
9795
  provider: "openrouter",
8758
9796
  baseUrl: "https://openrouter.ai/api/v1",
8759
9797
  reasoning: true,
8760
- input: ["text"],
9798
+ input: ["text", "image"],
8761
9799
  cost: {
8762
- input: 0.22,
8763
- output: 0.85,
8764
- cacheRead: 0.06,
8765
- cacheWrite: 0,
9800
+ input: 2,
9801
+ output: 10,
9802
+ cacheRead: 0.19999999999999998,
9803
+ cacheWrite: 2.5,
8766
9804
  },
8767
- contextWindow: 262144,
8768
- maxTokens: 262144,
9805
+ contextWindow: 1000000,
9806
+ maxTokens: 128000,
8769
9807
  },
8770
- "arcee-ai/trinity-mini": {
8771
- id: "arcee-ai/trinity-mini",
8772
- name: "Arcee AI: Trinity Mini",
9808
+ "arcee-ai/trinity-large-thinking": {
9809
+ id: "arcee-ai/trinity-large-thinking",
9810
+ name: "Arcee AI: Trinity Large Thinking",
8773
9811
  api: "openai-completions",
8774
9812
  provider: "openrouter",
8775
9813
  baseUrl: "https://openrouter.ai/api/v1",
8776
9814
  reasoning: true,
8777
9815
  input: ["text"],
8778
9816
  cost: {
8779
- input: 0.045,
8780
- output: 0.15,
8781
- cacheRead: 0,
9817
+ input: 0.25,
9818
+ output: 0.7999999999999999,
9819
+ cacheRead: 0.06,
8782
9820
  cacheWrite: 0,
8783
9821
  },
8784
- contextWindow: 131072,
8785
- maxTokens: 131072,
9822
+ contextWindow: 262144,
9823
+ maxTokens: 80000,
8786
9824
  },
8787
9825
  "arcee-ai/virtuoso-large": {
8788
9826
  id: "arcee-ai/virtuoso-large",
@@ -8920,6 +9958,23 @@ export const MODELS = {
8920
9958
  contextWindow: 128000,
8921
9959
  maxTokens: 4000,
8922
9960
  },
9961
+ "cohere/north-mini-code:free": {
9962
+ id: "cohere/north-mini-code:free",
9963
+ name: "Cohere: North Mini Code (free)",
9964
+ api: "openai-completions",
9965
+ provider: "openrouter",
9966
+ baseUrl: "https://openrouter.ai/api/v1",
9967
+ reasoning: true,
9968
+ input: ["text"],
9969
+ cost: {
9970
+ input: 0,
9971
+ output: 0,
9972
+ cacheRead: 0,
9973
+ cacheWrite: 0,
9974
+ },
9975
+ contextWindow: 256000,
9976
+ maxTokens: 64000,
9977
+ },
8923
9978
  "deepseek/deepseek-chat": {
8924
9979
  id: "deepseek/deepseek-chat",
8925
9980
  name: "DeepSeek: DeepSeek V3",
@@ -8946,8 +10001,8 @@ export const MODELS = {
8946
10001
  reasoning: false,
8947
10002
  input: ["text"],
8948
10003
  cost: {
8949
- input: 0.19999999999999998,
8950
- output: 0.77,
10004
+ input: 0.24,
10005
+ output: 0.8999999999999999,
8951
10006
  cacheRead: 0.135,
8952
10007
  cacheWrite: 0,
8953
10008
  },
@@ -9031,9 +10086,9 @@ export const MODELS = {
9031
10086
  reasoning: true,
9032
10087
  input: ["text"],
9033
10088
  cost: {
9034
- input: 0.2288,
9035
- output: 0.3432,
9036
- cacheRead: 0,
10089
+ input: 0.2145,
10090
+ output: 0.32175,
10091
+ cacheRead: 0.02145,
9037
10092
  cacheWrite: 0,
9038
10093
  },
9039
10094
  contextWindow: 131072,
@@ -9067,13 +10122,13 @@ export const MODELS = {
9067
10122
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
9068
10123
  input: ["text"],
9069
10124
  cost: {
9070
- input: 0.098,
9071
- output: 0.196,
9072
- cacheRead: 0.02,
10125
+ input: 0.077,
10126
+ output: 0.154,
10127
+ cacheRead: 0.015399999999999999,
9073
10128
  cacheWrite: 0,
9074
10129
  },
9075
10130
  contextWindow: 1048576,
9076
- maxTokens: 4096,
10131
+ maxTokens: 384000,
9077
10132
  },
9078
10133
  "deepseek/deepseek-v4-pro": {
9079
10134
  id: "deepseek/deepseek-v4-pro",
@@ -9094,23 +10149,6 @@ export const MODELS = {
9094
10149
  contextWindow: 1048576,
9095
10150
  maxTokens: 384000,
9096
10151
  },
9097
- "essentialai/rnj-1-instruct": {
9098
- id: "essentialai/rnj-1-instruct",
9099
- name: "EssentialAI: Rnj 1 Instruct",
9100
- api: "openai-completions",
9101
- provider: "openrouter",
9102
- baseUrl: "https://openrouter.ai/api/v1",
9103
- reasoning: false,
9104
- input: ["text"],
9105
- cost: {
9106
- input: 0.15,
9107
- output: 0.15,
9108
- cacheRead: 0,
9109
- cacheWrite: 0,
9110
- },
9111
- contextWindow: 32768,
9112
- maxTokens: 4096,
9113
- },
9114
10152
  "google/gemini-2.5-flash": {
9115
10153
  id: "google/gemini-2.5-flash",
9116
10154
  name: "Google: Gemini 2.5 Flash",
@@ -9145,23 +10183,6 @@ export const MODELS = {
9145
10183
  contextWindow: 1048576,
9146
10184
  maxTokens: 65535,
9147
10185
  },
9148
- "google/gemini-2.5-flash-lite-preview-09-2025": {
9149
- id: "google/gemini-2.5-flash-lite-preview-09-2025",
9150
- name: "Google: Gemini 2.5 Flash Lite Preview 09-2025",
9151
- api: "openai-completions",
9152
- provider: "openrouter",
9153
- baseUrl: "https://openrouter.ai/api/v1",
9154
- reasoning: true,
9155
- input: ["text", "image"],
9156
- cost: {
9157
- input: 0.09999999999999999,
9158
- output: 0.39999999999999997,
9159
- cacheRead: 0.01,
9160
- cacheWrite: 0.08333333333333334,
9161
- },
9162
- contextWindow: 1048576,
9163
- maxTokens: 65535,
9164
- },
9165
10186
  "google/gemini-2.5-pro": {
9166
10187
  id: "google/gemini-2.5-pro",
9167
10188
  name: "Google: Gemini 2.5 Pro",
@@ -9227,8 +10248,25 @@ export const MODELS = {
9227
10248
  cacheRead: 0.049999999999999996,
9228
10249
  cacheWrite: 0.08333333333333334,
9229
10250
  },
9230
- contextWindow: 1048576,
9231
- maxTokens: 65536,
10251
+ contextWindow: 1048576,
10252
+ maxTokens: 65535,
10253
+ },
10254
+ "google/gemini-3-pro-image": {
10255
+ id: "google/gemini-3-pro-image",
10256
+ name: "Google: Nano Banana Pro (Gemini 3 Pro Image)",
10257
+ api: "openai-completions",
10258
+ provider: "openrouter",
10259
+ baseUrl: "https://openrouter.ai/api/v1",
10260
+ reasoning: true,
10261
+ input: ["text", "image"],
10262
+ cost: {
10263
+ input: 2,
10264
+ output: 12,
10265
+ cacheRead: 0.19999999999999998,
10266
+ cacheWrite: 0.375,
10267
+ },
10268
+ contextWindow: 65536,
10269
+ maxTokens: 32768,
9232
10270
  },
9233
10271
  "google/gemini-3.1-flash-lite": {
9234
10272
  id: "google/gemini-3.1-flash-lite",
@@ -9415,7 +10453,7 @@ export const MODELS = {
9415
10453
  cacheWrite: 0,
9416
10454
  },
9417
10455
  contextWindow: 262144,
9418
- maxTokens: 32768,
10456
+ maxTokens: 8192,
9419
10457
  },
9420
10458
  "ibm-granite/granite-4.1-8b": {
9421
10459
  id: "ibm-granite/granite-4.1-8b",
@@ -9520,6 +10558,23 @@ export const MODELS = {
9520
10558
  contextWindow: 256000,
9521
10559
  maxTokens: 80000,
9522
10560
  },
10561
+ "liquid/lfm-2.5-1.2b-thinking:free": {
10562
+ id: "liquid/lfm-2.5-1.2b-thinking:free",
10563
+ name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
10564
+ api: "openai-completions",
10565
+ provider: "openrouter",
10566
+ baseUrl: "https://openrouter.ai/api/v1",
10567
+ reasoning: true,
10568
+ input: ["text"],
10569
+ cost: {
10570
+ input: 0,
10571
+ output: 0,
10572
+ cacheRead: 0,
10573
+ cacheWrite: 0,
10574
+ },
10575
+ contextWindow: 32768,
10576
+ maxTokens: 4096,
10577
+ },
9523
10578
  "meta-llama/llama-3.1-70b-instruct": {
9524
10579
  id: "meta-llama/llama-3.1-70b-instruct",
9525
10580
  name: "Meta: Llama 3.1 70B Instruct",
@@ -9597,8 +10652,8 @@ export const MODELS = {
9597
10652
  reasoning: false,
9598
10653
  input: ["text", "image"],
9599
10654
  cost: {
9600
- input: 0.15,
9601
- output: 0.6,
10655
+ input: 0.19999999999999998,
10656
+ output: 0.7999999999999999,
9602
10657
  cacheRead: 0,
9603
10658
  cacheWrite: 0,
9604
10659
  },
@@ -9649,12 +10704,12 @@ export const MODELS = {
9649
10704
  input: ["text"],
9650
10705
  cost: {
9651
10706
  input: 0.255,
9652
- output: 1,
9653
- cacheRead: 0.03,
10707
+ output: 1.02,
10708
+ cacheRead: 0,
9654
10709
  cacheWrite: 0,
9655
10710
  },
9656
10711
  contextWindow: 204800,
9657
- maxTokens: 196608,
10712
+ maxTokens: 131072,
9658
10713
  },
9659
10714
  "minimax/minimax-m2.1": {
9660
10715
  id: "minimax/minimax-m2.1",
@@ -9665,13 +10720,13 @@ export const MODELS = {
9665
10720
  reasoning: true,
9666
10721
  input: ["text"],
9667
10722
  cost: {
9668
- input: 0.29,
9669
- output: 0.95,
10723
+ input: 0.3,
10724
+ output: 1.2,
9670
10725
  cacheRead: 0.03,
9671
10726
  cacheWrite: 0,
9672
10727
  },
9673
10728
  contextWindow: 204800,
9674
- maxTokens: 196608,
10729
+ maxTokens: 131072,
9675
10730
  },
9676
10731
  "minimax/minimax-m2.5": {
9677
10732
  id: "minimax/minimax-m2.5",
@@ -9699,13 +10754,13 @@ export const MODELS = {
9699
10754
  reasoning: true,
9700
10755
  input: ["text"],
9701
10756
  cost: {
9702
- input: 0.25,
9703
- output: 1,
9704
- cacheRead: 0.049999999999999996,
10757
+ input: 0.24,
10758
+ output: 0.96,
10759
+ cacheRead: 0,
9705
10760
  cacheWrite: 0,
9706
10761
  },
9707
10762
  contextWindow: 204800,
9708
- maxTokens: 131072,
10763
+ maxTokens: 196608,
9709
10764
  },
9710
10765
  "minimax/minimax-m3": {
9711
10766
  id: "minimax/minimax-m3",
@@ -9722,7 +10777,7 @@ export const MODELS = {
9722
10777
  cacheWrite: 0,
9723
10778
  },
9724
10779
  contextWindow: 1048576,
9725
- maxTokens: 512000,
10780
+ maxTokens: 131072,
9726
10781
  },
9727
10782
  "mistralai/codestral-2508": {
9728
10783
  id: "mistralai/codestral-2508",
@@ -10028,7 +11083,7 @@ export const MODELS = {
10028
11083
  cacheWrite: 0,
10029
11084
  },
10030
11085
  contextWindow: 131072,
10031
- maxTokens: 32768,
11086
+ maxTokens: 100352,
10032
11087
  },
10033
11088
  "moonshotai/kimi-k2-0905": {
10034
11089
  id: "moonshotai/kimi-k2-0905",
@@ -10045,7 +11100,7 @@ export const MODELS = {
10045
11100
  cacheWrite: 0,
10046
11101
  },
10047
11102
  contextWindow: 262144,
10048
- maxTokens: 262144,
11103
+ maxTokens: 100352,
10049
11104
  },
10050
11105
  "moonshotai/kimi-k2-thinking": {
10051
11106
  id: "moonshotai/kimi-k2-thinking",
@@ -10058,11 +11113,11 @@ export const MODELS = {
10058
11113
  cost: {
10059
11114
  input: 0.6,
10060
11115
  output: 2.5,
10061
- cacheRead: 0,
11116
+ cacheRead: 0.15,
10062
11117
  cacheWrite: 0,
10063
11118
  },
10064
11119
  contextWindow: 262144,
10065
- maxTokens: 262144,
11120
+ maxTokens: 100352,
10066
11121
  },
10067
11122
  "moonshotai/kimi-k2.5": {
10068
11123
  id: "moonshotai/kimi-k2.5",
@@ -10091,13 +11146,13 @@ export const MODELS = {
10091
11146
  reasoning: true,
10092
11147
  input: ["text", "image"],
10093
11148
  cost: {
10094
- input: 0.6799999999999999,
11149
+ input: 0.66,
10095
11150
  output: 3.41,
10096
- cacheRead: 0.33999999999999997,
11151
+ cacheRead: 0.15,
10097
11152
  cacheWrite: 0,
10098
11153
  },
10099
11154
  contextWindow: 262144,
10100
- maxTokens: 262142,
11155
+ maxTokens: 262144,
10101
11156
  },
10102
11157
  "moonshotai/kimi-k2.7-code": {
10103
11158
  id: "moonshotai/kimi-k2.7-code",
@@ -10108,26 +11163,43 @@ export const MODELS = {
10108
11163
  reasoning: true,
10109
11164
  input: ["text", "image"],
10110
11165
  cost: {
10111
- input: 0.75,
10112
- output: 3.5,
10113
- cacheRead: 0.16,
11166
+ input: 0.719,
11167
+ output: 3.49,
11168
+ cacheRead: 0.149,
10114
11169
  cacheWrite: 0,
10115
11170
  },
10116
11171
  contextWindow: 262144,
10117
11172
  maxTokens: 262144,
10118
11173
  },
10119
- "nex-agi/nex-n2-pro:free": {
10120
- id: "nex-agi/nex-n2-pro:free",
10121
- name: "Nex AGI: Nex-N2-Pro (free)",
11174
+ "nex-agi/nex-n2-mini": {
11175
+ id: "nex-agi/nex-n2-mini",
11176
+ name: "Nex AGI: Nex-N2-Mini",
10122
11177
  api: "openai-completions",
10123
11178
  provider: "openrouter",
10124
11179
  baseUrl: "https://openrouter.ai/api/v1",
10125
11180
  reasoning: true,
10126
11181
  input: ["text", "image"],
10127
11182
  cost: {
10128
- input: 0,
10129
- output: 0,
10130
- cacheRead: 0,
11183
+ input: 0.024999999999999998,
11184
+ output: 0.09999999999999999,
11185
+ cacheRead: 0.0025,
11186
+ cacheWrite: 0,
11187
+ },
11188
+ contextWindow: 262144,
11189
+ maxTokens: 262144,
11190
+ },
11191
+ "nex-agi/nex-n2-pro": {
11192
+ id: "nex-agi/nex-n2-pro",
11193
+ name: "Nex AGI: Nex-N2-Pro",
11194
+ api: "openai-completions",
11195
+ provider: "openrouter",
11196
+ baseUrl: "https://openrouter.ai/api/v1",
11197
+ reasoning: true,
11198
+ input: ["text", "image"],
11199
+ cost: {
11200
+ input: 0.25,
11201
+ output: 1,
11202
+ cacheRead: 0.024999999999999998,
10131
11203
  cacheWrite: 0,
10132
11204
  },
10133
11205
  contextWindow: 262144,
@@ -10210,7 +11282,7 @@ export const MODELS = {
10210
11282
  reasoning: true,
10211
11283
  input: ["text"],
10212
11284
  cost: {
10213
- input: 0.09,
11285
+ input: 0.08,
10214
11286
  output: 0.44999999999999996,
10215
11287
  cacheRead: 0,
10216
11288
  cacheWrite: 0,
@@ -10245,8 +11317,8 @@ export const MODELS = {
10245
11317
  input: ["text"],
10246
11318
  cost: {
10247
11319
  input: 0.5,
10248
- output: 2.5,
10249
- cacheRead: 0.15,
11320
+ output: 2.2,
11321
+ cacheRead: 0.09999999999999999,
10250
11322
  cacheWrite: 0,
10251
11323
  },
10252
11324
  contextWindow: 1000000,
@@ -10944,6 +12016,114 @@ export const MODELS = {
10944
12016
  contextWindow: 1050000,
10945
12017
  maxTokens: 128000,
10946
12018
  },
12019
+ "openai/gpt-5.6-luna": {
12020
+ id: "openai/gpt-5.6-luna",
12021
+ name: "OpenAI: GPT-5.6 Luna",
12022
+ api: "openai-completions",
12023
+ provider: "openrouter",
12024
+ baseUrl: "https://openrouter.ai/api/v1",
12025
+ reasoning: true,
12026
+ thinkingLevelMap: { "xhigh": "xhigh" },
12027
+ input: ["text", "image"],
12028
+ cost: {
12029
+ input: 1,
12030
+ output: 6,
12031
+ cacheRead: 0.09999999999999999,
12032
+ cacheWrite: 1.25,
12033
+ },
12034
+ contextWindow: 1050000,
12035
+ maxTokens: 128000,
12036
+ },
12037
+ "openai/gpt-5.6-luna-pro": {
12038
+ id: "openai/gpt-5.6-luna-pro",
12039
+ name: "OpenAI: GPT-5.6 Luna Pro",
12040
+ api: "openai-completions",
12041
+ provider: "openrouter",
12042
+ baseUrl: "https://openrouter.ai/api/v1",
12043
+ reasoning: true,
12044
+ thinkingLevelMap: { "xhigh": "xhigh", "off": null, "minimal": null, "low": null },
12045
+ input: ["text", "image"],
12046
+ cost: {
12047
+ input: 1,
12048
+ output: 6,
12049
+ cacheRead: 0.09999999999999999,
12050
+ cacheWrite: 1.25,
12051
+ },
12052
+ contextWindow: 1050000,
12053
+ maxTokens: 128000,
12054
+ },
12055
+ "openai/gpt-5.6-sol": {
12056
+ id: "openai/gpt-5.6-sol",
12057
+ name: "OpenAI: GPT-5.6 Sol",
12058
+ api: "openai-completions",
12059
+ provider: "openrouter",
12060
+ baseUrl: "https://openrouter.ai/api/v1",
12061
+ reasoning: true,
12062
+ thinkingLevelMap: { "xhigh": "xhigh" },
12063
+ input: ["text", "image"],
12064
+ cost: {
12065
+ input: 5,
12066
+ output: 30,
12067
+ cacheRead: 0.5,
12068
+ cacheWrite: 6.25,
12069
+ },
12070
+ contextWindow: 1050000,
12071
+ maxTokens: 128000,
12072
+ },
12073
+ "openai/gpt-5.6-sol-pro": {
12074
+ id: "openai/gpt-5.6-sol-pro",
12075
+ name: "OpenAI: GPT-5.6 Sol Pro",
12076
+ api: "openai-completions",
12077
+ provider: "openrouter",
12078
+ baseUrl: "https://openrouter.ai/api/v1",
12079
+ reasoning: true,
12080
+ thinkingLevelMap: { "xhigh": "xhigh", "off": null, "minimal": null, "low": null },
12081
+ input: ["text", "image"],
12082
+ cost: {
12083
+ input: 5,
12084
+ output: 30,
12085
+ cacheRead: 0.5,
12086
+ cacheWrite: 6.25,
12087
+ },
12088
+ contextWindow: 1050000,
12089
+ maxTokens: 128000,
12090
+ },
12091
+ "openai/gpt-5.6-terra": {
12092
+ id: "openai/gpt-5.6-terra",
12093
+ name: "OpenAI: GPT-5.6 Terra",
12094
+ api: "openai-completions",
12095
+ provider: "openrouter",
12096
+ baseUrl: "https://openrouter.ai/api/v1",
12097
+ reasoning: true,
12098
+ thinkingLevelMap: { "xhigh": "xhigh" },
12099
+ input: ["text", "image"],
12100
+ cost: {
12101
+ input: 2.5,
12102
+ output: 15,
12103
+ cacheRead: 0.25,
12104
+ cacheWrite: 3.125,
12105
+ },
12106
+ contextWindow: 1050000,
12107
+ maxTokens: 128000,
12108
+ },
12109
+ "openai/gpt-5.6-terra-pro": {
12110
+ id: "openai/gpt-5.6-terra-pro",
12111
+ name: "OpenAI: GPT-5.6 Terra Pro",
12112
+ api: "openai-completions",
12113
+ provider: "openrouter",
12114
+ baseUrl: "https://openrouter.ai/api/v1",
12115
+ reasoning: true,
12116
+ thinkingLevelMap: { "xhigh": "xhigh", "off": null, "minimal": null, "low": null },
12117
+ input: ["text", "image"],
12118
+ cost: {
12119
+ input: 2.5,
12120
+ output: 15,
12121
+ cacheRead: 0.25,
12122
+ cacheWrite: 3.125,
12123
+ },
12124
+ contextWindow: 1050000,
12125
+ maxTokens: 128000,
12126
+ },
10947
12127
  "openai/gpt-audio": {
10948
12128
  id: "openai/gpt-audio",
10949
12129
  name: "OpenAI: GPT Audio",
@@ -11004,7 +12184,7 @@ export const MODELS = {
11004
12184
  reasoning: true,
11005
12185
  input: ["text"],
11006
12186
  cost: {
11007
- input: 0.039,
12187
+ input: 0.036,
11008
12188
  output: 0.18,
11009
12189
  cacheRead: 0,
11010
12190
  cacheWrite: 0,
@@ -11061,7 +12241,7 @@ export const MODELS = {
11061
12241
  cacheWrite: 0,
11062
12242
  },
11063
12243
  contextWindow: 131072,
11064
- maxTokens: 8192,
12244
+ maxTokens: 32768,
11065
12245
  },
11066
12246
  "openai/gpt-oss-safeguard-20b": {
11067
12247
  id: "openai/gpt-oss-safeguard-20b",
@@ -11074,7 +12254,7 @@ export const MODELS = {
11074
12254
  cost: {
11075
12255
  input: 0.075,
11076
12256
  output: 0.3,
11077
- cacheRead: 0.037,
12257
+ cacheRead: 0.0375,
11078
12258
  cacheWrite: 0,
11079
12259
  },
11080
12260
  contextWindow: 131072,
@@ -11267,22 +12447,22 @@ export const MODELS = {
11267
12447
  contextWindow: 200000,
11268
12448
  maxTokens: 4096,
11269
12449
  },
11270
- "openrouter/owl-alpha": {
11271
- id: "openrouter/owl-alpha",
11272
- name: "Owl Alpha",
12450
+ "poolside/laguna-m.1": {
12451
+ id: "poolside/laguna-m.1",
12452
+ name: "Poolside: Laguna M.1",
11273
12453
  api: "openai-completions",
11274
12454
  provider: "openrouter",
11275
12455
  baseUrl: "https://openrouter.ai/api/v1",
11276
- reasoning: false,
12456
+ reasoning: true,
11277
12457
  input: ["text"],
11278
12458
  cost: {
11279
- input: 0,
11280
- output: 0,
11281
- cacheRead: 0,
12459
+ input: 0.19999999999999998,
12460
+ output: 0.39999999999999997,
12461
+ cacheRead: 0.09999999999999999,
11282
12462
  cacheWrite: 0,
11283
12463
  },
11284
- contextWindow: 1048756,
11285
- maxTokens: 262144,
12464
+ contextWindow: 262144,
12465
+ maxTokens: 32768,
11286
12466
  },
11287
12467
  "poolside/laguna-m.1:free": {
11288
12468
  id: "poolside/laguna-m.1:free",
@@ -11301,39 +12481,39 @@ export const MODELS = {
11301
12481
  contextWindow: 262144,
11302
12482
  maxTokens: 32768,
11303
12483
  },
11304
- "poolside/laguna-xs.2:free": {
11305
- id: "poolside/laguna-xs.2:free",
11306
- name: "Poolside: Laguna XS.2 (free)",
12484
+ "poolside/laguna-xs-2.1": {
12485
+ id: "poolside/laguna-xs-2.1",
12486
+ name: "Poolside: Laguna XS 2.1",
11307
12487
  api: "openai-completions",
11308
12488
  provider: "openrouter",
11309
12489
  baseUrl: "https://openrouter.ai/api/v1",
11310
12490
  reasoning: true,
11311
12491
  input: ["text"],
11312
12492
  cost: {
11313
- input: 0,
11314
- output: 0,
11315
- cacheRead: 0,
12493
+ input: 0.06,
12494
+ output: 0.12,
12495
+ cacheRead: 0.03,
11316
12496
  cacheWrite: 0,
11317
12497
  },
11318
12498
  contextWindow: 262144,
11319
12499
  maxTokens: 32768,
11320
12500
  },
11321
- "prime-intellect/intellect-3": {
11322
- id: "prime-intellect/intellect-3",
11323
- name: "Prime Intellect: INTELLECT-3",
12501
+ "poolside/laguna-xs-2.1:free": {
12502
+ id: "poolside/laguna-xs-2.1:free",
12503
+ name: "Poolside: Laguna XS 2.1 (free)",
11324
12504
  api: "openai-completions",
11325
12505
  provider: "openrouter",
11326
12506
  baseUrl: "https://openrouter.ai/api/v1",
11327
12507
  reasoning: true,
11328
12508
  input: ["text"],
11329
12509
  cost: {
11330
- input: 0.19999999999999998,
11331
- output: 1.1,
12510
+ input: 0,
12511
+ output: 0,
11332
12512
  cacheRead: 0,
11333
12513
  cacheWrite: 0,
11334
12514
  },
11335
- contextWindow: 131072,
11336
- maxTokens: 131072,
12515
+ contextWindow: 262144,
12516
+ maxTokens: 32768,
11337
12517
  },
11338
12518
  "qwen/qwen-2.5-72b-instruct": {
11339
12519
  id: "qwen/qwen-2.5-72b-instruct",
@@ -11352,6 +12532,23 @@ export const MODELS = {
11352
12532
  contextWindow: 131072,
11353
12533
  maxTokens: 16384,
11354
12534
  },
12535
+ "qwen/qwen-2.5-7b-instruct": {
12536
+ id: "qwen/qwen-2.5-7b-instruct",
12537
+ name: "Qwen: Qwen2.5 7B Instruct",
12538
+ api: "openai-completions",
12539
+ provider: "openrouter",
12540
+ baseUrl: "https://openrouter.ai/api/v1",
12541
+ reasoning: false,
12542
+ input: ["text"],
12543
+ cost: {
12544
+ input: 0.04,
12545
+ output: 0.09999999999999999,
12546
+ cacheRead: 0,
12547
+ cacheWrite: 0,
12548
+ },
12549
+ contextWindow: 131072,
12550
+ maxTokens: 32768,
12551
+ },
11355
12552
  "qwen/qwen-plus": {
11356
12553
  id: "qwen/qwen-plus",
11357
12554
  name: "Qwen: Qwen-Plus",
@@ -11447,7 +12644,7 @@ export const MODELS = {
11447
12644
  input: ["text"],
11448
12645
  cost: {
11449
12646
  input: 0.09,
11450
- output: 0.09999999999999999,
12647
+ output: 0.55,
11451
12648
  cacheRead: 0,
11452
12649
  cacheWrite: 0,
11453
12650
  },
@@ -11463,13 +12660,13 @@ export const MODELS = {
11463
12660
  reasoning: true,
11464
12661
  input: ["text"],
11465
12662
  cost: {
11466
- input: 0.09999999999999999,
11467
- output: 0.09999999999999999,
11468
- cacheRead: 0.09999999999999999,
12663
+ input: 0.14950000000000002,
12664
+ output: 1.495,
12665
+ cacheRead: 0,
11469
12666
  cacheWrite: 0,
11470
12667
  },
11471
12668
  contextWindow: 262144,
11472
- maxTokens: 262144,
12669
+ maxTokens: 4096,
11473
12670
  },
11474
12671
  "qwen/qwen3-30b-a3b": {
11475
12672
  id: "qwen/qwen3-30b-a3b",
@@ -11514,13 +12711,13 @@ export const MODELS = {
11514
12711
  reasoning: true,
11515
12712
  input: ["text"],
11516
12713
  cost: {
11517
- input: 0.08,
11518
- output: 0.39999999999999997,
11519
- cacheRead: 0.08,
12714
+ input: 0.13,
12715
+ output: 1.56,
12716
+ cacheRead: 0,
11520
12717
  cacheWrite: 0,
11521
12718
  },
11522
12719
  contextWindow: 131072,
11523
- maxTokens: 131072,
12720
+ maxTokens: 32768,
11524
12721
  },
11525
12722
  "qwen/qwen3-32b": {
11526
12723
  id: "qwen/qwen3-32b",
@@ -11548,9 +12745,9 @@ export const MODELS = {
11548
12745
  reasoning: true,
11549
12746
  input: ["text"],
11550
12747
  cost: {
11551
- input: 0.049999999999999996,
11552
- output: 0.39999999999999997,
11553
- cacheRead: 0.049999999999999996,
12748
+ input: 0.117,
12749
+ output: 0.45499999999999996,
12750
+ cacheRead: 0,
11554
12751
  cacheWrite: 0,
11555
12752
  },
11556
12753
  contextWindow: 131072,
@@ -11837,8 +13034,8 @@ export const MODELS = {
11837
13034
  reasoning: false,
11838
13035
  input: ["text", "image"],
11839
13036
  cost: {
11840
- input: 0.08,
11841
- output: 0.5,
13037
+ input: 0.117,
13038
+ output: 0.45499999999999996,
11842
13039
  cacheRead: 0,
11843
13040
  cacheWrite: 0,
11844
13041
  },
@@ -11924,7 +13121,7 @@ export const MODELS = {
11924
13121
  cost: {
11925
13122
  input: 0.385,
11926
13123
  output: 2.4499999999999997,
11927
- cacheRead: 0,
13124
+ cacheRead: 0.111,
11928
13125
  cacheWrite: 0,
11929
13126
  },
11930
13127
  contextWindow: 256000,
@@ -12007,9 +13204,9 @@ export const MODELS = {
12007
13204
  reasoning: true,
12008
13205
  input: ["text", "image"],
12009
13206
  cost: {
12010
- input: 0.28850000000000003,
12011
- output: 3.17,
12012
- cacheRead: 0,
13207
+ input: 0.28500000000000003,
13208
+ output: 2.4,
13209
+ cacheRead: 0.15,
12013
13210
  cacheWrite: 0,
12014
13211
  },
12015
13212
  contextWindow: 262144,
@@ -12024,9 +13221,9 @@ export const MODELS = {
12024
13221
  reasoning: true,
12025
13222
  input: ["text", "image"],
12026
13223
  cost: {
12027
- input: 0.15,
13224
+ input: 0.14,
12028
13225
  output: 1,
12029
- cacheRead: 0.049999999999999996,
13226
+ cacheRead: 0,
12030
13227
  cacheWrite: 0,
12031
13228
  },
12032
13229
  contextWindow: 262144,
@@ -12148,7 +13345,24 @@ export const MODELS = {
12148
13345
  cacheRead: 0,
12149
13346
  cacheWrite: 0,
12150
13347
  },
12151
- contextWindow: 256000,
13348
+ contextWindow: 256000,
13349
+ maxTokens: 128000,
13350
+ },
13351
+ "sakana/fugu-ultra": {
13352
+ id: "sakana/fugu-ultra",
13353
+ name: "Sakana: Fugu Ultra",
13354
+ api: "openai-completions",
13355
+ provider: "openrouter",
13356
+ baseUrl: "https://openrouter.ai/api/v1",
13357
+ reasoning: true,
13358
+ input: ["text", "image"],
13359
+ cost: {
13360
+ input: 5,
13361
+ output: 30,
13362
+ cacheRead: 0.5,
13363
+ cacheWrite: 0,
13364
+ },
13365
+ contextWindow: 1000000,
12152
13366
  maxTokens: 128000,
12153
13367
  },
12154
13368
  "sao10k/l3.1-euryale-70b": {
@@ -12177,13 +13391,13 @@ export const MODELS = {
12177
13391
  reasoning: true,
12178
13392
  input: ["text"],
12179
13393
  cost: {
12180
- input: 0.09,
13394
+ input: 0.09999999999999999,
12181
13395
  output: 0.3,
12182
- cacheRead: 0.02,
13396
+ cacheRead: 0,
12183
13397
  cacheWrite: 0,
12184
13398
  },
12185
13399
  contextWindow: 262144,
12186
- maxTokens: 16384,
13400
+ maxTokens: 65536,
12187
13401
  },
12188
13402
  "stepfun/step-3.7-flash": {
12189
13403
  id: "stepfun/step-3.7-flash",
@@ -12202,6 +13416,23 @@ export const MODELS = {
12202
13416
  contextWindow: 256000,
12203
13417
  maxTokens: 256000,
12204
13418
  },
13419
+ "tencent/hy3": {
13420
+ id: "tencent/hy3",
13421
+ name: "Tencent: Hy3",
13422
+ api: "openai-completions",
13423
+ provider: "openrouter",
13424
+ baseUrl: "https://openrouter.ai/api/v1",
13425
+ reasoning: true,
13426
+ input: ["text"],
13427
+ cost: {
13428
+ input: 0.14,
13429
+ output: 0.58,
13430
+ cacheRead: 0.035,
13431
+ cacheWrite: 0,
13432
+ },
13433
+ contextWindow: 262144,
13434
+ maxTokens: 4096,
13435
+ },
12205
13436
  "tencent/hy3-preview": {
12206
13437
  id: "tencent/hy3-preview",
12207
13438
  name: "Tencent: Hy3 preview",
@@ -12211,30 +13442,30 @@ export const MODELS = {
12211
13442
  reasoning: true,
12212
13443
  input: ["text"],
12213
13444
  cost: {
12214
- input: 0.06599999999999999,
12215
- output: 0.26,
12216
- cacheRead: 0.029,
13445
+ input: 0.063,
13446
+ output: 0.21,
13447
+ cacheRead: 0.020999999999999998,
12217
13448
  cacheWrite: 0,
12218
13449
  },
12219
13450
  contextWindow: 262144,
12220
- maxTokens: 262144,
13451
+ maxTokens: 4096,
12221
13452
  },
12222
- "thedrummer/rocinante-12b": {
12223
- id: "thedrummer/rocinante-12b",
12224
- name: "TheDrummer: Rocinante 12B",
13453
+ "tencent/hy3:free": {
13454
+ id: "tencent/hy3:free",
13455
+ name: "Tencent: Hy3 (free)",
12225
13456
  api: "openai-completions",
12226
13457
  provider: "openrouter",
12227
13458
  baseUrl: "https://openrouter.ai/api/v1",
12228
- reasoning: false,
13459
+ reasoning: true,
12229
13460
  input: ["text"],
12230
13461
  cost: {
12231
- input: 0.16999999999999998,
12232
- output: 0.43,
13462
+ input: 0,
13463
+ output: 0,
12233
13464
  cacheRead: 0,
12234
13465
  cacheWrite: 0,
12235
13466
  },
12236
- contextWindow: 32768,
12237
- maxTokens: 32768,
13467
+ contextWindow: 262144,
13468
+ maxTokens: 262144,
12238
13469
  },
12239
13470
  "thedrummer/unslopnemo-12b": {
12240
13471
  id: "thedrummer/unslopnemo-12b",
@@ -12304,39 +13535,39 @@ export const MODELS = {
12304
13535
  contextWindow: 1000000,
12305
13536
  maxTokens: 4096,
12306
13537
  },
12307
- "x-ai/grok-build-0.1": {
12308
- id: "x-ai/grok-build-0.1",
12309
- name: "xAI: Grok Build 0.1",
13538
+ "x-ai/grok-4.5": {
13539
+ id: "x-ai/grok-4.5",
13540
+ name: "xAI: Grok 4.5",
12310
13541
  api: "openai-completions",
12311
13542
  provider: "openrouter",
12312
13543
  baseUrl: "https://openrouter.ai/api/v1",
12313
13544
  reasoning: true,
12314
13545
  input: ["text", "image"],
12315
13546
  cost: {
12316
- input: 1,
12317
- output: 2,
12318
- cacheRead: 0.19999999999999998,
13547
+ input: 2,
13548
+ output: 6,
13549
+ cacheRead: 0.5,
12319
13550
  cacheWrite: 0,
12320
13551
  },
12321
- contextWindow: 256000,
13552
+ contextWindow: 500000,
12322
13553
  maxTokens: 4096,
12323
13554
  },
12324
- "xiaomi/mimo-v2-flash": {
12325
- id: "xiaomi/mimo-v2-flash",
12326
- name: "Xiaomi: MiMo-V2-Flash",
13555
+ "x-ai/grok-build-0.1": {
13556
+ id: "x-ai/grok-build-0.1",
13557
+ name: "xAI: Grok Build 0.1",
12327
13558
  api: "openai-completions",
12328
13559
  provider: "openrouter",
12329
13560
  baseUrl: "https://openrouter.ai/api/v1",
12330
13561
  reasoning: true,
12331
- input: ["text"],
13562
+ input: ["text", "image"],
12332
13563
  cost: {
12333
- input: 0.09999999999999999,
12334
- output: 0.3,
12335
- cacheRead: 0.01,
13564
+ input: 1,
13565
+ output: 2,
13566
+ cacheRead: 0.19999999999999998,
12336
13567
  cacheWrite: 0,
12337
13568
  },
12338
- contextWindow: 262144,
12339
- maxTokens: 65536,
13569
+ contextWindow: 256000,
13570
+ maxTokens: 4096,
12340
13571
  },
12341
13572
  "xiaomi/mimo-v2.5": {
12342
13573
  id: "xiaomi/mimo-v2.5",
@@ -12347,13 +13578,13 @@ export const MODELS = {
12347
13578
  reasoning: true,
12348
13579
  input: ["text", "image"],
12349
13580
  cost: {
12350
- input: 0.14,
13581
+ input: 0.105,
12351
13582
  output: 0.28,
12352
- cacheRead: 0.0028,
13583
+ cacheRead: 0.028,
12353
13584
  cacheWrite: 0,
12354
13585
  },
12355
13586
  contextWindow: 1048576,
12356
- maxTokens: 131072,
13587
+ maxTokens: 4096,
12357
13588
  },
12358
13589
  "xiaomi/mimo-v2.5-pro": {
12359
13590
  id: "xiaomi/mimo-v2.5-pro",
@@ -12433,12 +13664,12 @@ export const MODELS = {
12433
13664
  input: ["text"],
12434
13665
  cost: {
12435
13666
  input: 0.43,
12436
- output: 1.74,
13667
+ output: 1.75,
12437
13668
  cacheRead: 0.08,
12438
13669
  cacheWrite: 0,
12439
13670
  },
12440
- contextWindow: 202752,
12441
- maxTokens: 131072,
13671
+ contextWindow: 200000,
13672
+ maxTokens: 16384,
12442
13673
  },
12443
13674
  "z-ai/glm-4.6v": {
12444
13675
  id: "z-ai/glm-4.6v",
@@ -12534,13 +13765,47 @@ export const MODELS = {
12534
13765
  reasoning: true,
12535
13766
  input: ["text"],
12536
13767
  cost: {
12537
- input: 0.98,
12538
- output: 3.08,
12539
- cacheRead: 0.182,
13768
+ input: 0.966,
13769
+ output: 3.036,
13770
+ cacheRead: 0.1794,
12540
13771
  cacheWrite: 0,
12541
13772
  },
12542
13773
  contextWindow: 202752,
12543
- maxTokens: 4096,
13774
+ maxTokens: 128000,
13775
+ },
13776
+ "z-ai/glm-5.2": {
13777
+ id: "z-ai/glm-5.2",
13778
+ name: "Z.ai: GLM 5.2",
13779
+ api: "openai-completions",
13780
+ provider: "openrouter",
13781
+ baseUrl: "https://openrouter.ai/api/v1",
13782
+ reasoning: true,
13783
+ input: ["text"],
13784
+ cost: {
13785
+ input: 0.9299999999999999,
13786
+ output: 3,
13787
+ cacheRead: 0.18,
13788
+ cacheWrite: 0,
13789
+ },
13790
+ contextWindow: 1048576,
13791
+ maxTokens: 32768,
13792
+ },
13793
+ "z-ai/glm-5v-turbo": {
13794
+ id: "z-ai/glm-5v-turbo",
13795
+ name: "Z.ai: GLM 5V Turbo",
13796
+ api: "openai-completions",
13797
+ provider: "openrouter",
13798
+ baseUrl: "https://openrouter.ai/api/v1",
13799
+ reasoning: true,
13800
+ input: ["text", "image"],
13801
+ cost: {
13802
+ input: 1.2,
13803
+ output: 4,
13804
+ cacheRead: 0.24,
13805
+ cacheWrite: 0,
13806
+ },
13807
+ contextWindow: 202752,
13808
+ maxTokens: 131072,
12544
13809
  },
12545
13810
  "~anthropic/claude-fable-latest": {
12546
13811
  id: "~anthropic/claude-fable-latest",
@@ -12602,10 +13867,10 @@ export const MODELS = {
12602
13867
  reasoning: true,
12603
13868
  input: ["text", "image"],
12604
13869
  cost: {
12605
- input: 3,
12606
- output: 15,
12607
- cacheRead: 0.3,
12608
- cacheWrite: 3.75,
13870
+ input: 2,
13871
+ output: 10,
13872
+ cacheRead: 0.19999999999999998,
13873
+ cacheWrite: 2.5,
12609
13874
  },
12610
13875
  contextWindow: 1000000,
12611
13876
  maxTokens: 128000,
@@ -12653,13 +13918,13 @@ export const MODELS = {
12653
13918
  reasoning: true,
12654
13919
  input: ["text", "image"],
12655
13920
  cost: {
12656
- input: 0.6799999999999999,
13921
+ input: 0.66,
12657
13922
  output: 3.41,
12658
- cacheRead: 0.33999999999999997,
13923
+ cacheRead: 0.15,
12659
13924
  cacheWrite: 0,
12660
13925
  },
12661
13926
  contextWindow: 262144,
12662
- maxTokens: 262142,
13927
+ maxTokens: 262144,
12663
13928
  },
12664
13929
  "~openai/gpt-latest": {
12665
13930
  id: "~openai/gpt-latest",
@@ -12673,7 +13938,7 @@ export const MODELS = {
12673
13938
  input: 5,
12674
13939
  output: 30,
12675
13940
  cacheRead: 0.5,
12676
- cacheWrite: 0,
13941
+ cacheWrite: 6.25,
12677
13942
  },
12678
13943
  contextWindow: 1050000,
12679
13944
  maxTokens: 128000,
@@ -12695,6 +13960,23 @@ export const MODELS = {
12695
13960
  contextWindow: 400000,
12696
13961
  maxTokens: 128000,
12697
13962
  },
13963
+ "~x-ai/grok-latest": {
13964
+ id: "~x-ai/grok-latest",
13965
+ name: "xAI: Grok Latest",
13966
+ api: "openai-completions",
13967
+ provider: "openrouter",
13968
+ baseUrl: "https://openrouter.ai/api/v1",
13969
+ reasoning: true,
13970
+ input: ["text", "image"],
13971
+ cost: {
13972
+ input: 2,
13973
+ output: 6,
13974
+ cacheRead: 0.5,
13975
+ cacheWrite: 0,
13976
+ },
13977
+ contextWindow: 500000,
13978
+ maxTokens: 4096,
13979
+ },
12698
13980
  },
12699
13981
  "together": {
12700
13982
  "MiniMaxAI/MiniMax-M2.7": {
@@ -12912,8 +14194,8 @@ export const MODELS = {
12912
14194
  reasoning: false,
12913
14195
  input: ["text"],
12914
14196
  cost: {
12915
- input: 0.88,
12916
- output: 0.88,
14197
+ input: 1.04,
14198
+ output: 1.04,
12917
14199
  cacheRead: 0,
12918
14200
  cacheWrite: 0,
12919
14201
  },
@@ -13053,6 +14335,25 @@ export const MODELS = {
13053
14335
  contextWindow: 202752,
13054
14336
  maxTokens: 131072,
13055
14337
  },
14338
+ "zai-org/GLM-5.2": {
14339
+ id: "zai-org/GLM-5.2",
14340
+ name: "GLM-5.2",
14341
+ api: "openai-completions",
14342
+ provider: "together",
14343
+ baseUrl: "https://api.together.ai/v1",
14344
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
14345
+ reasoning: true,
14346
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
14347
+ input: ["text"],
14348
+ cost: {
14349
+ input: 1.4,
14350
+ output: 4.4,
14351
+ cacheRead: 0.26,
14352
+ cacheWrite: 0,
14353
+ },
14354
+ contextWindow: 262144,
14355
+ maxTokens: 164000,
14356
+ },
13056
14357
  },
13057
14358
  "vercel-ai-gateway": {
13058
14359
  "alibaba/qwen-3-14b": {
@@ -13163,7 +14464,7 @@ export const MODELS = {
13163
14464
  api: "anthropic-messages",
13164
14465
  provider: "vercel-ai-gateway",
13165
14466
  baseUrl: "https://ai-gateway.vercel.sh",
13166
- reasoning: true,
14467
+ reasoning: false,
13167
14468
  input: ["text"],
13168
14469
  cost: {
13169
14470
  input: 1.5,
@@ -13180,7 +14481,7 @@ export const MODELS = {
13180
14481
  api: "anthropic-messages",
13181
14482
  provider: "vercel-ai-gateway",
13182
14483
  baseUrl: "https://ai-gateway.vercel.sh",
13183
- reasoning: true,
14484
+ reasoning: false,
13184
14485
  input: ["text"],
13185
14486
  cost: {
13186
14487
  input: 0.15,
@@ -13197,7 +14498,7 @@ export const MODELS = {
13197
14498
  api: "anthropic-messages",
13198
14499
  provider: "vercel-ai-gateway",
13199
14500
  baseUrl: "https://ai-gateway.vercel.sh",
13200
- reasoning: true,
14501
+ reasoning: false,
13201
14502
  input: ["text"],
13202
14503
  cost: {
13203
14504
  input: 0.5,
@@ -13310,6 +14611,40 @@ export const MODELS = {
13310
14611
  contextWindow: 131072,
13311
14612
  maxTokens: 32768,
13312
14613
  },
14614
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
14615
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
14616
+ name: "Qwen3 VL 235B A22B Instruct",
14617
+ api: "anthropic-messages",
14618
+ provider: "vercel-ai-gateway",
14619
+ baseUrl: "https://ai-gateway.vercel.sh",
14620
+ reasoning: false,
14621
+ input: ["text", "image"],
14622
+ cost: {
14623
+ input: 0.39999999999999997,
14624
+ output: 1.5999999999999999,
14625
+ cacheRead: 0,
14626
+ cacheWrite: 0,
14627
+ },
14628
+ contextWindow: 131072,
14629
+ maxTokens: 129024,
14630
+ },
14631
+ "alibaba/qwen3-vl-instruct": {
14632
+ id: "alibaba/qwen3-vl-instruct",
14633
+ name: "Qwen3 VL 235B A22B Instruct",
14634
+ api: "anthropic-messages",
14635
+ provider: "vercel-ai-gateway",
14636
+ baseUrl: "https://ai-gateway.vercel.sh",
14637
+ reasoning: false,
14638
+ input: ["text", "image"],
14639
+ cost: {
14640
+ input: 0.39999999999999997,
14641
+ output: 1.5999999999999999,
14642
+ cacheRead: 0,
14643
+ cacheWrite: 0,
14644
+ },
14645
+ contextWindow: 131072,
14646
+ maxTokens: 129024,
14647
+ },
13313
14648
  "alibaba/qwen3-vl-thinking": {
13314
14649
  id: "alibaba/qwen3-vl-thinking",
13315
14650
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13429,6 +14764,74 @@ export const MODELS = {
13429
14764
  contextWindow: 1000000,
13430
14765
  maxTokens: 64000,
13431
14766
  },
14767
+ "amazon/nova-2-lite": {
14768
+ id: "amazon/nova-2-lite",
14769
+ name: "Nova 2 Lite",
14770
+ api: "anthropic-messages",
14771
+ provider: "vercel-ai-gateway",
14772
+ baseUrl: "https://ai-gateway.vercel.sh",
14773
+ reasoning: true,
14774
+ input: ["text", "image"],
14775
+ cost: {
14776
+ input: 0.3,
14777
+ output: 2.5,
14778
+ cacheRead: 0.075,
14779
+ cacheWrite: 0,
14780
+ },
14781
+ contextWindow: 1000000,
14782
+ maxTokens: 1000000,
14783
+ },
14784
+ "amazon/nova-lite": {
14785
+ id: "amazon/nova-lite",
14786
+ name: "Nova Lite",
14787
+ api: "anthropic-messages",
14788
+ provider: "vercel-ai-gateway",
14789
+ baseUrl: "https://ai-gateway.vercel.sh",
14790
+ reasoning: false,
14791
+ input: ["text", "image"],
14792
+ cost: {
14793
+ input: 0.06,
14794
+ output: 0.24,
14795
+ cacheRead: 0,
14796
+ cacheWrite: 0,
14797
+ },
14798
+ contextWindow: 300000,
14799
+ maxTokens: 8192,
14800
+ },
14801
+ "amazon/nova-micro": {
14802
+ id: "amazon/nova-micro",
14803
+ name: "Nova Micro",
14804
+ api: "anthropic-messages",
14805
+ provider: "vercel-ai-gateway",
14806
+ baseUrl: "https://ai-gateway.vercel.sh",
14807
+ reasoning: false,
14808
+ input: ["text"],
14809
+ cost: {
14810
+ input: 0.035,
14811
+ output: 0.14,
14812
+ cacheRead: 0,
14813
+ cacheWrite: 0,
14814
+ },
14815
+ contextWindow: 128000,
14816
+ maxTokens: 8192,
14817
+ },
14818
+ "amazon/nova-pro": {
14819
+ id: "amazon/nova-pro",
14820
+ name: "Nova Pro",
14821
+ api: "anthropic-messages",
14822
+ provider: "vercel-ai-gateway",
14823
+ baseUrl: "https://ai-gateway.vercel.sh",
14824
+ reasoning: false,
14825
+ input: ["text", "image"],
14826
+ cost: {
14827
+ input: 0.7999999999999999,
14828
+ output: 3.1999999999999997,
14829
+ cacheRead: 0,
14830
+ cacheWrite: 0,
14831
+ },
14832
+ contextWindow: 300000,
14833
+ maxTokens: 8192,
14834
+ },
13432
14835
  "anthropic/claude-3-haiku": {
13433
14836
  id: "anthropic/claude-3-haiku",
13434
14837
  name: "Claude 3 Haiku",
@@ -13463,6 +14866,23 @@ export const MODELS = {
13463
14866
  contextWindow: 200000,
13464
14867
  maxTokens: 8192,
13465
14868
  },
14869
+ "anthropic/claude-fable-5": {
14870
+ id: "anthropic/claude-fable-5",
14871
+ name: "Claude Fable 5",
14872
+ api: "anthropic-messages",
14873
+ provider: "vercel-ai-gateway",
14874
+ baseUrl: "https://ai-gateway.vercel.sh",
14875
+ reasoning: true,
14876
+ input: ["text", "image"],
14877
+ cost: {
14878
+ input: 10,
14879
+ output: 50,
14880
+ cacheRead: 1,
14881
+ cacheWrite: 12.5,
14882
+ },
14883
+ contextWindow: 1000000,
14884
+ maxTokens: 128000,
14885
+ },
13466
14886
  "anthropic/claude-haiku-4.5": {
13467
14887
  id: "anthropic/claude-haiku-4.5",
13468
14888
  name: "Claude Haiku 4.5",
@@ -13495,7 +14915,7 @@ export const MODELS = {
13495
14915
  cacheWrite: 18.75,
13496
14916
  },
13497
14917
  contextWindow: 200000,
13498
- maxTokens: 32000,
14918
+ maxTokens: 8192,
13499
14919
  },
13500
14920
  "anthropic/claude-opus-4.1": {
13501
14921
  id: "anthropic/claude-opus-4.1",
@@ -13602,7 +15022,7 @@ export const MODELS = {
13602
15022
  cacheWrite: 3.75,
13603
15023
  },
13604
15024
  contextWindow: 1000000,
13605
- maxTokens: 64000,
15025
+ maxTokens: 8192,
13606
15026
  },
13607
15027
  "anthropic/claude-sonnet-4.5": {
13608
15028
  id: "anthropic/claude-sonnet-4.5",
@@ -13639,6 +15059,23 @@ export const MODELS = {
13639
15059
  contextWindow: 1000000,
13640
15060
  maxTokens: 128000,
13641
15061
  },
15062
+ "anthropic/claude-sonnet-5": {
15063
+ id: "anthropic/claude-sonnet-5",
15064
+ name: "Claude Sonnet 5",
15065
+ api: "anthropic-messages",
15066
+ provider: "vercel-ai-gateway",
15067
+ baseUrl: "https://ai-gateway.vercel.sh",
15068
+ reasoning: true,
15069
+ input: ["text", "image"],
15070
+ cost: {
15071
+ input: 2,
15072
+ output: 10,
15073
+ cacheRead: 0.19999999999999998,
15074
+ cacheWrite: 2.5,
15075
+ },
15076
+ contextWindow: 1000000,
15077
+ maxTokens: 128000,
15078
+ },
13642
15079
  "arcee-ai/trinity-large-preview": {
13643
15080
  id: "arcee-ai/trinity-large-preview",
13644
15081
  name: "Trinity Large Preview",
@@ -13673,6 +15110,23 @@ export const MODELS = {
13673
15110
  contextWindow: 262100,
13674
15111
  maxTokens: 80000,
13675
15112
  },
15113
+ "arcee-ai/trinity-mini": {
15114
+ id: "arcee-ai/trinity-mini",
15115
+ name: "Trinity Mini",
15116
+ api: "anthropic-messages",
15117
+ provider: "vercel-ai-gateway",
15118
+ baseUrl: "https://ai-gateway.vercel.sh",
15119
+ reasoning: false,
15120
+ input: ["text"],
15121
+ cost: {
15122
+ input: 0.045,
15123
+ output: 0.15,
15124
+ cacheRead: 0,
15125
+ cacheWrite: 0,
15126
+ },
15127
+ contextWindow: 131072,
15128
+ maxTokens: 131072,
15129
+ },
13676
15130
  "bytedance/seed-1.6": {
13677
15131
  id: "bytedance/seed-1.6",
13678
15132
  name: "Seed 1.6",
@@ -13680,7 +15134,7 @@ export const MODELS = {
13680
15134
  provider: "vercel-ai-gateway",
13681
15135
  baseUrl: "https://ai-gateway.vercel.sh",
13682
15136
  reasoning: true,
13683
- input: ["text"],
15137
+ input: ["text", "image"],
13684
15138
  cost: {
13685
15139
  input: 0.25,
13686
15140
  output: 2,
@@ -13690,6 +15144,23 @@ export const MODELS = {
13690
15144
  contextWindow: 256000,
13691
15145
  maxTokens: 32000,
13692
15146
  },
15147
+ "bytedance/seed-1.8": {
15148
+ id: "bytedance/seed-1.8",
15149
+ name: "Bytedance Seed 1.8",
15150
+ api: "anthropic-messages",
15151
+ provider: "vercel-ai-gateway",
15152
+ baseUrl: "https://ai-gateway.vercel.sh",
15153
+ reasoning: true,
15154
+ input: ["text", "image"],
15155
+ cost: {
15156
+ input: 0.25,
15157
+ output: 2,
15158
+ cacheRead: 0.049999999999999996,
15159
+ cacheWrite: 0,
15160
+ },
15161
+ contextWindow: 256000,
15162
+ maxTokens: 64000,
15163
+ },
13693
15164
  "cohere/command-a": {
13694
15165
  id: "cohere/command-a",
13695
15166
  name: "Command A",
@@ -13750,13 +15221,13 @@ export const MODELS = {
13750
15221
  reasoning: true,
13751
15222
  input: ["text"],
13752
15223
  cost: {
13753
- input: 0.56,
13754
- output: 1.68,
13755
- cacheRead: 0.28,
15224
+ input: 0.21,
15225
+ output: 0.7899999999999999,
15226
+ cacheRead: 0.13,
13756
15227
  cacheWrite: 0,
13757
15228
  },
13758
15229
  contextWindow: 163840,
13759
- maxTokens: 8192,
15230
+ maxTokens: 128000,
13760
15231
  },
13761
15232
  "deepseek/deepseek-v3.1-terminus": {
13762
15233
  id: "deepseek/deepseek-v3.1-terminus",
@@ -13781,8 +15252,8 @@ export const MODELS = {
13781
15252
  api: "anthropic-messages",
13782
15253
  provider: "vercel-ai-gateway",
13783
15254
  baseUrl: "https://ai-gateway.vercel.sh",
13784
- reasoning: true,
13785
- input: ["text", "image"],
15255
+ reasoning: false,
15256
+ input: ["text"],
13786
15257
  cost: {
13787
15258
  input: 0.28,
13788
15259
  output: 0.42,
@@ -13799,7 +15270,7 @@ export const MODELS = {
13799
15270
  provider: "vercel-ai-gateway",
13800
15271
  baseUrl: "https://ai-gateway.vercel.sh",
13801
15272
  reasoning: true,
13802
- input: ["text", "image"],
15273
+ input: ["text"],
13803
15274
  cost: {
13804
15275
  input: 0.62,
13805
15276
  output: 1.85,
@@ -13816,11 +15287,11 @@ export const MODELS = {
13816
15287
  provider: "vercel-ai-gateway",
13817
15288
  baseUrl: "https://ai-gateway.vercel.sh",
13818
15289
  reasoning: true,
13819
- input: ["text", "image"],
15290
+ input: ["text"],
13820
15291
  cost: {
13821
15292
  input: 0.14,
13822
15293
  output: 0.28,
13823
- cacheRead: 0.0028,
15294
+ cacheRead: 0.028,
13824
15295
  cacheWrite: 0,
13825
15296
  },
13826
15297
  contextWindow: 1000000,
@@ -14019,7 +15490,7 @@ export const MODELS = {
14019
15490
  api: "anthropic-messages",
14020
15491
  provider: "vercel-ai-gateway",
14021
15492
  baseUrl: "https://ai-gateway.vercel.sh",
14022
- reasoning: false,
15493
+ reasoning: true,
14023
15494
  input: ["text", "image"],
14024
15495
  cost: {
14025
15496
  input: 0.14,
@@ -14064,6 +15535,57 @@ export const MODELS = {
14064
15535
  contextWindow: 32000,
14065
15536
  maxTokens: 16384,
14066
15537
  },
15538
+ "interfaze/interfaze-beta": {
15539
+ id: "interfaze/interfaze-beta",
15540
+ name: "Interfaze Beta",
15541
+ api: "anthropic-messages",
15542
+ provider: "vercel-ai-gateway",
15543
+ baseUrl: "https://ai-gateway.vercel.sh",
15544
+ reasoning: true,
15545
+ input: ["text", "image"],
15546
+ cost: {
15547
+ input: 1.5,
15548
+ output: 3.5,
15549
+ cacheRead: 0,
15550
+ cacheWrite: 0,
15551
+ },
15552
+ contextWindow: 1000000,
15553
+ maxTokens: 32000,
15554
+ },
15555
+ "kwaipilot/kat-coder-air-v2.5": {
15556
+ id: "kwaipilot/kat-coder-air-v2.5",
15557
+ name: "Kat Coder Air V2.5",
15558
+ api: "anthropic-messages",
15559
+ provider: "vercel-ai-gateway",
15560
+ baseUrl: "https://ai-gateway.vercel.sh",
15561
+ reasoning: true,
15562
+ input: ["text"],
15563
+ cost: {
15564
+ input: 0.15,
15565
+ output: 0.6,
15566
+ cacheRead: 0.03,
15567
+ cacheWrite: 0,
15568
+ },
15569
+ contextWindow: 256000,
15570
+ maxTokens: 80000,
15571
+ },
15572
+ "kwaipilot/kat-coder-pro-v1": {
15573
+ id: "kwaipilot/kat-coder-pro-v1",
15574
+ name: "KAT-Coder-Pro V1",
15575
+ api: "anthropic-messages",
15576
+ provider: "vercel-ai-gateway",
15577
+ baseUrl: "https://ai-gateway.vercel.sh",
15578
+ reasoning: false,
15579
+ input: ["text"],
15580
+ cost: {
15581
+ input: 0.3,
15582
+ output: 1.2,
15583
+ cacheRead: 0.06,
15584
+ cacheWrite: 0,
15585
+ },
15586
+ contextWindow: 256000,
15587
+ maxTokens: 32000,
15588
+ },
14067
15589
  "kwaipilot/kat-coder-pro-v2": {
14068
15590
  id: "kwaipilot/kat-coder-pro-v2",
14069
15591
  name: "Kat Coder Pro V2",
@@ -14081,6 +15603,23 @@ export const MODELS = {
14081
15603
  contextWindow: 256000,
14082
15604
  maxTokens: 256000,
14083
15605
  },
15606
+ "kwaipilot/kat-coder-pro-v2.5": {
15607
+ id: "kwaipilot/kat-coder-pro-v2.5",
15608
+ name: "Kat Coder Pro V2.5",
15609
+ api: "anthropic-messages",
15610
+ provider: "vercel-ai-gateway",
15611
+ baseUrl: "https://ai-gateway.vercel.sh",
15612
+ reasoning: true,
15613
+ input: ["text"],
15614
+ cost: {
15615
+ input: 0.74,
15616
+ output: 2.96,
15617
+ cacheRead: 0.15,
15618
+ cacheWrite: 0,
15619
+ },
15620
+ contextWindow: 256000,
15621
+ maxTokens: 80000,
15622
+ },
14084
15623
  "meituan/longcat-flash-chat": {
14085
15624
  id: "meituan/longcat-flash-chat",
14086
15625
  name: "LongCat Flash Chat",
@@ -14098,6 +15637,23 @@ export const MODELS = {
14098
15637
  contextWindow: 128000,
14099
15638
  maxTokens: 100000,
14100
15639
  },
15640
+ "meituan/longcat-flash-thinking-2601": {
15641
+ id: "meituan/longcat-flash-thinking-2601",
15642
+ name: "LongCat Flash Thinking 2601",
15643
+ api: "anthropic-messages",
15644
+ provider: "vercel-ai-gateway",
15645
+ baseUrl: "https://ai-gateway.vercel.sh",
15646
+ reasoning: true,
15647
+ input: ["text"],
15648
+ cost: {
15649
+ input: 0,
15650
+ output: 0,
15651
+ cacheRead: 0,
15652
+ cacheWrite: 0,
15653
+ },
15654
+ contextWindow: 32768,
15655
+ maxTokens: 32768,
15656
+ },
14101
15657
  "meta/llama-3.1-70b": {
14102
15658
  id: "meta/llama-3.1-70b",
14103
15659
  name: "Llama 3.1 70B Instruct",
@@ -14217,6 +15773,23 @@ export const MODELS = {
14217
15773
  contextWindow: 128000,
14218
15774
  maxTokens: 8192,
14219
15775
  },
15776
+ "meta/muse-spark-1.1": {
15777
+ id: "meta/muse-spark-1.1",
15778
+ name: "Muse Spark 1.1",
15779
+ api: "anthropic-messages",
15780
+ provider: "vercel-ai-gateway",
15781
+ baseUrl: "https://ai-gateway.vercel.sh",
15782
+ reasoning: true,
15783
+ input: ["text", "image"],
15784
+ cost: {
15785
+ input: 1.25,
15786
+ output: 4.25,
15787
+ cacheRead: 0.15,
15788
+ cacheWrite: 0,
15789
+ },
15790
+ contextWindow: 1048576,
15791
+ maxTokens: 1048576,
15792
+ },
14220
15793
  "minimax/minimax-m2": {
14221
15794
  id: "minimax/minimax-m2",
14222
15795
  name: "MiniMax M2",
@@ -14411,7 +15984,7 @@ export const MODELS = {
14411
15984
  provider: "vercel-ai-gateway",
14412
15985
  baseUrl: "https://ai-gateway.vercel.sh",
14413
15986
  reasoning: false,
14414
- input: ["text"],
15987
+ input: ["text", "image"],
14415
15988
  cost: {
14416
15989
  input: 0.09999999999999999,
14417
15990
  output: 0.3,
@@ -14421,6 +15994,57 @@ export const MODELS = {
14421
15994
  contextWindow: 256000,
14422
15995
  maxTokens: 256000,
14423
15996
  },
15997
+ "mistral/magistral-medium": {
15998
+ id: "mistral/magistral-medium",
15999
+ name: "Magistral Medium 2509",
16000
+ api: "anthropic-messages",
16001
+ provider: "vercel-ai-gateway",
16002
+ baseUrl: "https://ai-gateway.vercel.sh",
16003
+ reasoning: true,
16004
+ input: ["text", "image"],
16005
+ cost: {
16006
+ input: 2,
16007
+ output: 5,
16008
+ cacheRead: 0,
16009
+ cacheWrite: 0,
16010
+ },
16011
+ contextWindow: 128000,
16012
+ maxTokens: 64000,
16013
+ },
16014
+ "mistral/magistral-small": {
16015
+ id: "mistral/magistral-small",
16016
+ name: "Magistral Small 2509",
16017
+ api: "anthropic-messages",
16018
+ provider: "vercel-ai-gateway",
16019
+ baseUrl: "https://ai-gateway.vercel.sh",
16020
+ reasoning: true,
16021
+ input: ["text", "image"],
16022
+ cost: {
16023
+ input: 0.5,
16024
+ output: 1.5,
16025
+ cacheRead: 0,
16026
+ cacheWrite: 0,
16027
+ },
16028
+ contextWindow: 128000,
16029
+ maxTokens: 64000,
16030
+ },
16031
+ "mistral/ministral-14b": {
16032
+ id: "mistral/ministral-14b",
16033
+ name: "Ministral 14B",
16034
+ api: "anthropic-messages",
16035
+ provider: "vercel-ai-gateway",
16036
+ baseUrl: "https://ai-gateway.vercel.sh",
16037
+ reasoning: false,
16038
+ input: ["text", "image"],
16039
+ cost: {
16040
+ input: 0.19999999999999998,
16041
+ output: 0.19999999999999998,
16042
+ cacheRead: 0,
16043
+ cacheWrite: 0,
16044
+ },
16045
+ contextWindow: 256000,
16046
+ maxTokens: 256000,
16047
+ },
14424
16048
  "mistral/ministral-3b": {
14425
16049
  id: "mistral/ministral-3b",
14426
16050
  name: "Ministral 3B",
@@ -14455,6 +16079,23 @@ export const MODELS = {
14455
16079
  contextWindow: 128000,
14456
16080
  maxTokens: 4000,
14457
16081
  },
16082
+ "mistral/mistral-large-3": {
16083
+ id: "mistral/mistral-large-3",
16084
+ name: "Mistral Large 3",
16085
+ api: "anthropic-messages",
16086
+ provider: "vercel-ai-gateway",
16087
+ baseUrl: "https://ai-gateway.vercel.sh",
16088
+ reasoning: false,
16089
+ input: ["text", "image"],
16090
+ cost: {
16091
+ input: 0.5,
16092
+ output: 1.5,
16093
+ cacheRead: 0,
16094
+ cacheWrite: 0,
16095
+ },
16096
+ contextWindow: 256000,
16097
+ maxTokens: 256000,
16098
+ },
14458
16099
  "mistral/mistral-medium": {
14459
16100
  id: "mistral/mistral-medium",
14460
16101
  name: "Mistral Medium 3.1",
@@ -14479,7 +16120,7 @@ export const MODELS = {
14479
16120
  provider: "vercel-ai-gateway",
14480
16121
  baseUrl: "https://ai-gateway.vercel.sh",
14481
16122
  reasoning: true,
14482
- input: ["text"],
16123
+ input: ["text", "image"],
14483
16124
  cost: {
14484
16125
  input: 1.5,
14485
16126
  output: 7.5,
@@ -14498,13 +16139,13 @@ export const MODELS = {
14498
16139
  reasoning: false,
14499
16140
  input: ["text"],
14500
16141
  cost: {
14501
- input: 0.02,
14502
- output: 0.04,
16142
+ input: 0.15,
16143
+ output: 0.15,
14503
16144
  cacheRead: 0,
14504
16145
  cacheWrite: 0,
14505
16146
  },
14506
- contextWindow: 131072,
14507
- maxTokens: 131072,
16147
+ contextWindow: 128000,
16148
+ maxTokens: 128000,
14508
16149
  },
14509
16150
  "mistral/mistral-small": {
14510
16151
  id: "mistral/mistral-small",
@@ -14583,13 +16224,13 @@ export const MODELS = {
14583
16224
  reasoning: true,
14584
16225
  input: ["text"],
14585
16226
  cost: {
14586
- input: 0.6,
14587
- output: 2.5,
14588
- cacheRead: 0.15,
16227
+ input: 0.47,
16228
+ output: 2,
16229
+ cacheRead: 0.14100000000000001,
14589
16230
  cacheWrite: 0,
14590
16231
  },
14591
- contextWindow: 262114,
14592
- maxTokens: 262114,
16232
+ contextWindow: 216144,
16233
+ maxTokens: 216144,
14593
16234
  },
14594
16235
  "moonshotai/kimi-k2.5": {
14595
16236
  id: "moonshotai/kimi-k2.5",
@@ -14657,7 +16298,24 @@ export const MODELS = {
14657
16298
  cacheWrite: 0,
14658
16299
  },
14659
16300
  contextWindow: 262144,
14660
- maxTokens: 4096,
16301
+ maxTokens: 32768,
16302
+ },
16303
+ "nvidia/nemotron-3-nano-30b-a3b": {
16304
+ id: "nvidia/nemotron-3-nano-30b-a3b",
16305
+ name: "Nemotron 3 Nano 30B A3B",
16306
+ api: "anthropic-messages",
16307
+ provider: "vercel-ai-gateway",
16308
+ baseUrl: "https://ai-gateway.vercel.sh",
16309
+ reasoning: true,
16310
+ input: ["text"],
16311
+ cost: {
16312
+ input: 0.049999999999999996,
16313
+ output: 0.24,
16314
+ cacheRead: 0,
16315
+ cacheWrite: 0,
16316
+ },
16317
+ contextWindow: 262144,
16318
+ maxTokens: 262144,
14661
16319
  },
14662
16320
  "nvidia/nemotron-3-super-120b-a12b": {
14663
16321
  id: "nvidia/nemotron-3-super-120b-a12b",
@@ -14727,6 +16385,23 @@ export const MODELS = {
14727
16385
  contextWindow: 131072,
14728
16386
  maxTokens: 131072,
14729
16387
  },
16388
+ "openai/gpt-3.5-turbo": {
16389
+ id: "openai/gpt-3.5-turbo",
16390
+ name: "GPT-3.5 Turbo",
16391
+ api: "anthropic-messages",
16392
+ provider: "vercel-ai-gateway",
16393
+ baseUrl: "https://ai-gateway.vercel.sh",
16394
+ reasoning: false,
16395
+ input: ["text"],
16396
+ cost: {
16397
+ input: 0.5,
16398
+ output: 1.5,
16399
+ cacheRead: 0,
16400
+ cacheWrite: 0,
16401
+ },
16402
+ contextWindow: 16385,
16403
+ maxTokens: 4096,
16404
+ },
14730
16405
  "openai/gpt-4-turbo": {
14731
16406
  id: "openai/gpt-4-turbo",
14732
16407
  name: "GPT-4 Turbo",
@@ -14852,7 +16527,7 @@ export const MODELS = {
14852
16527
  api: "anthropic-messages",
14853
16528
  provider: "vercel-ai-gateway",
14854
16529
  baseUrl: "https://ai-gateway.vercel.sh",
14855
- reasoning: true,
16530
+ reasoning: false,
14856
16531
  input: ["text", "image"],
14857
16532
  cost: {
14858
16533
  input: 1.25,
@@ -14988,7 +16663,7 @@ export const MODELS = {
14988
16663
  api: "anthropic-messages",
14989
16664
  provider: "vercel-ai-gateway",
14990
16665
  baseUrl: "https://ai-gateway.vercel.sh",
14991
- reasoning: true,
16666
+ reasoning: false,
14992
16667
  input: ["text", "image"],
14993
16668
  cost: {
14994
16669
  input: 1.25,
@@ -15040,7 +16715,7 @@ export const MODELS = {
15040
16715
  api: "anthropic-messages",
15041
16716
  provider: "vercel-ai-gateway",
15042
16717
  baseUrl: "https://ai-gateway.vercel.sh",
15043
- reasoning: true,
16718
+ reasoning: false,
15044
16719
  thinkingLevelMap: { "xhigh": "xhigh" },
15045
16720
  input: ["text", "image"],
15046
16721
  cost: {
@@ -15094,7 +16769,7 @@ export const MODELS = {
15094
16769
  api: "anthropic-messages",
15095
16770
  provider: "vercel-ai-gateway",
15096
16771
  baseUrl: "https://ai-gateway.vercel.sh",
15097
- reasoning: true,
16772
+ reasoning: false,
15098
16773
  thinkingLevelMap: { "xhigh": "xhigh" },
15099
16774
  input: ["text", "image"],
15100
16775
  cost: {
@@ -15232,6 +16907,60 @@ export const MODELS = {
15232
16907
  contextWindow: 1000000,
15233
16908
  maxTokens: 128000,
15234
16909
  },
16910
+ "openai/gpt-5.6-luna": {
16911
+ id: "openai/gpt-5.6-luna",
16912
+ name: "GPT 5.6 Luna",
16913
+ api: "anthropic-messages",
16914
+ provider: "vercel-ai-gateway",
16915
+ baseUrl: "https://ai-gateway.vercel.sh",
16916
+ reasoning: true,
16917
+ thinkingLevelMap: { "xhigh": "xhigh" },
16918
+ input: ["text", "image"],
16919
+ cost: {
16920
+ input: 1,
16921
+ output: 6,
16922
+ cacheRead: 0.09999999999999999,
16923
+ cacheWrite: 1.25,
16924
+ },
16925
+ contextWindow: 1050000,
16926
+ maxTokens: 128000,
16927
+ },
16928
+ "openai/gpt-5.6-sol": {
16929
+ id: "openai/gpt-5.6-sol",
16930
+ name: "GPT 5.6 Sol",
16931
+ api: "anthropic-messages",
16932
+ provider: "vercel-ai-gateway",
16933
+ baseUrl: "https://ai-gateway.vercel.sh",
16934
+ reasoning: true,
16935
+ thinkingLevelMap: { "xhigh": "xhigh" },
16936
+ input: ["text", "image"],
16937
+ cost: {
16938
+ input: 5,
16939
+ output: 30,
16940
+ cacheRead: 0.5,
16941
+ cacheWrite: 6.25,
16942
+ },
16943
+ contextWindow: 1050000,
16944
+ maxTokens: 128000,
16945
+ },
16946
+ "openai/gpt-5.6-terra": {
16947
+ id: "openai/gpt-5.6-terra",
16948
+ name: "GPT 5.6 Terra",
16949
+ api: "anthropic-messages",
16950
+ provider: "vercel-ai-gateway",
16951
+ baseUrl: "https://ai-gateway.vercel.sh",
16952
+ reasoning: true,
16953
+ thinkingLevelMap: { "xhigh": "xhigh" },
16954
+ input: ["text", "image"],
16955
+ cost: {
16956
+ input: 2.5,
16957
+ output: 15,
16958
+ cacheRead: 0.25,
16959
+ cacheWrite: 3.125,
16960
+ },
16961
+ contextWindow: 1050000,
16962
+ maxTokens: 128000,
16963
+ },
15235
16964
  "openai/gpt-oss-120b": {
15236
16965
  id: "openai/gpt-oss-120b",
15237
16966
  name: "GPT OSS 120B",
@@ -15241,13 +16970,13 @@ export const MODELS = {
15241
16970
  reasoning: true,
15242
16971
  input: ["text"],
15243
16972
  cost: {
15244
- input: 0.35,
15245
- output: 0.75,
15246
- cacheRead: 0.25,
16973
+ input: 0.09999999999999999,
16974
+ output: 0.5,
16975
+ cacheRead: 0,
15247
16976
  cacheWrite: 0,
15248
16977
  },
15249
16978
  contextWindow: 131072,
15250
- maxTokens: 131000,
16979
+ maxTokens: 131072,
15251
16980
  },
15252
16981
  "openai/gpt-oss-20b": {
15253
16982
  id: "openai/gpt-oss-20b",
@@ -15385,39 +17114,22 @@ export const MODELS = {
15385
17114
  contextWindow: 200000,
15386
17115
  maxTokens: 100000,
15387
17116
  },
15388
- "perplexity/sonar": {
15389
- id: "perplexity/sonar",
15390
- name: "Sonar",
15391
- api: "anthropic-messages",
15392
- provider: "vercel-ai-gateway",
15393
- baseUrl: "https://ai-gateway.vercel.sh",
15394
- reasoning: false,
15395
- input: ["text", "image"],
15396
- cost: {
15397
- input: 0,
15398
- output: 0,
15399
- cacheRead: 0,
15400
- cacheWrite: 0,
15401
- },
15402
- contextWindow: 127000,
15403
- maxTokens: 8000,
15404
- },
15405
- "perplexity/sonar-pro": {
15406
- id: "perplexity/sonar-pro",
15407
- name: "Sonar Pro",
17117
+ "sakana/fugu-ultra": {
17118
+ id: "sakana/fugu-ultra",
17119
+ name: "Fugu Ultra",
15408
17120
  api: "anthropic-messages",
15409
17121
  provider: "vercel-ai-gateway",
15410
17122
  baseUrl: "https://ai-gateway.vercel.sh",
15411
- reasoning: false,
17123
+ reasoning: true,
15412
17124
  input: ["text", "image"],
15413
17125
  cost: {
15414
- input: 0,
15415
- output: 0,
15416
- cacheRead: 0,
17126
+ input: 5,
17127
+ output: 30,
17128
+ cacheRead: 0.5,
15417
17129
  cacheWrite: 0,
15418
17130
  },
15419
- contextWindow: 200000,
15420
- maxTokens: 8000,
17131
+ contextWindow: 1000000,
17132
+ maxTokens: 1000000,
15421
17133
  },
15422
17134
  "stepfun/step-3.5-flash": {
15423
17135
  id: "stepfun/step-3.5-flash",
@@ -15430,8 +17142,8 @@ export const MODELS = {
15430
17142
  cost: {
15431
17143
  input: 0.09,
15432
17144
  output: 0.3,
15433
- cacheRead: 0,
15434
- cacheWrite: 0.02,
17145
+ cacheRead: 0.02,
17146
+ cacheWrite: 0,
15435
17147
  },
15436
17148
  contextWindow: 262114,
15437
17149
  maxTokens: 262114,
@@ -15606,6 +17318,23 @@ export const MODELS = {
15606
17318
  contextWindow: 1000000,
15607
17319
  maxTokens: 1000000,
15608
17320
  },
17321
+ "xai/grok-4.5": {
17322
+ id: "xai/grok-4.5",
17323
+ name: "Grok 4.5",
17324
+ api: "anthropic-messages",
17325
+ provider: "vercel-ai-gateway",
17326
+ baseUrl: "https://ai-gateway.vercel.sh",
17327
+ reasoning: true,
17328
+ input: ["text", "image"],
17329
+ cost: {
17330
+ input: 2,
17331
+ output: 6,
17332
+ cacheRead: 0.5,
17333
+ cacheWrite: 0,
17334
+ },
17335
+ contextWindow: 500000,
17336
+ maxTokens: 500000,
17337
+ },
15609
17338
  "xai/grok-build-0.1": {
15610
17339
  id: "xai/grok-build-0.1",
15611
17340
  name: "Grok Build 0.1",
@@ -15681,7 +17410,7 @@ export const MODELS = {
15681
17410
  provider: "vercel-ai-gateway",
15682
17411
  baseUrl: "https://ai-gateway.vercel.sh",
15683
17412
  reasoning: true,
15684
- input: ["text", "image"],
17413
+ input: ["text"],
15685
17414
  cost: {
15686
17415
  input: 0.435,
15687
17416
  output: 0.87,
@@ -15693,7 +17422,7 @@ export const MODELS = {
15693
17422
  },
15694
17423
  "zai/glm-4.5": {
15695
17424
  id: "zai/glm-4.5",
15696
- name: "GLM-4.5",
17425
+ name: "GLM 4.5",
15697
17426
  api: "anthropic-messages",
15698
17427
  provider: "vercel-ai-gateway",
15699
17428
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -15802,13 +17531,13 @@ export const MODELS = {
15802
17531
  reasoning: true,
15803
17532
  input: ["text"],
15804
17533
  cost: {
15805
- input: 2.25,
15806
- output: 2.75,
15807
- cacheRead: 2.25,
17534
+ input: 0.6,
17535
+ output: 2.2,
17536
+ cacheRead: 0.12,
15808
17537
  cacheWrite: 0,
15809
17538
  },
15810
- contextWindow: 131000,
15811
- maxTokens: 40000,
17539
+ contextWindow: 200000,
17540
+ maxTokens: 120000,
15812
17541
  },
15813
17542
  "zai/glm-4.7-flash": {
15814
17543
  id: "zai/glm-4.7-flash",
@@ -15853,8 +17582,8 @@ export const MODELS = {
15853
17582
  reasoning: true,
15854
17583
  input: ["text"],
15855
17584
  cost: {
15856
- input: 1,
15857
- output: 3.1999999999999997,
17585
+ input: 0.95,
17586
+ output: 3.15,
15858
17587
  cacheRead: 0.19999999999999998,
15859
17588
  cacheWrite: 0,
15860
17589
  },
@@ -15885,15 +17614,49 @@ export const MODELS = {
15885
17614
  provider: "vercel-ai-gateway",
15886
17615
  baseUrl: "https://ai-gateway.vercel.sh",
15887
17616
  reasoning: true,
15888
- input: ["text", "image"],
17617
+ input: ["text"],
17618
+ cost: {
17619
+ input: 1.3,
17620
+ output: 4.300000000000001,
17621
+ cacheRead: 0.26,
17622
+ cacheWrite: 0,
17623
+ },
17624
+ contextWindow: 202000,
17625
+ maxTokens: 202000,
17626
+ },
17627
+ "zai/glm-5.2": {
17628
+ id: "zai/glm-5.2",
17629
+ name: "GLM 5.2",
17630
+ api: "anthropic-messages",
17631
+ provider: "vercel-ai-gateway",
17632
+ baseUrl: "https://ai-gateway.vercel.sh",
17633
+ reasoning: true,
17634
+ input: ["text"],
15889
17635
  cost: {
15890
17636
  input: 1.4,
15891
17637
  output: 4.4,
15892
17638
  cacheRead: 0.26,
15893
17639
  cacheWrite: 0,
15894
17640
  },
15895
- contextWindow: 202800,
15896
- maxTokens: 64000,
17641
+ contextWindow: 1040000,
17642
+ maxTokens: 128000,
17643
+ },
17644
+ "zai/glm-5.2-fast": {
17645
+ id: "zai/glm-5.2-fast",
17646
+ name: "GLM 5.2 Fast",
17647
+ api: "anthropic-messages",
17648
+ provider: "vercel-ai-gateway",
17649
+ baseUrl: "https://ai-gateway.vercel.sh",
17650
+ reasoning: true,
17651
+ input: ["text"],
17652
+ cost: {
17653
+ input: 2.0999999999999996,
17654
+ output: 6.6000000000000005,
17655
+ cacheRead: 0.21,
17656
+ cacheWrite: 0,
17657
+ },
17658
+ contextWindow: 1000000,
17659
+ maxTokens: 128000,
15897
17660
  },
15898
17661
  "zai/glm-5v-turbo": {
15899
17662
  id: "zai/glm-5v-turbo",
@@ -15999,6 +17762,23 @@ export const MODELS = {
15999
17762
  contextWindow: 1000000,
16000
17763
  maxTokens: 30000,
16001
17764
  },
17765
+ "grok-4.5": {
17766
+ id: "grok-4.5",
17767
+ name: "Grok 4.5",
17768
+ api: "openai-completions",
17769
+ provider: "xai",
17770
+ baseUrl: "https://api.x.ai/v1",
17771
+ reasoning: true,
17772
+ input: ["text", "image"],
17773
+ cost: {
17774
+ input: 2,
17775
+ output: 6,
17776
+ cacheRead: 0.5,
17777
+ cacheWrite: 0,
17778
+ },
17779
+ contextWindow: 500000,
17780
+ maxTokens: 500000,
17781
+ },
16002
17782
  "grok-build-0.1": {
16003
17783
  id: "grok-build-0.1",
16004
17784
  name: "Grok Build 0.1",
@@ -16045,9 +17825,9 @@ export const MODELS = {
16045
17825
  reasoning: true,
16046
17826
  input: ["text"],
16047
17827
  cost: {
16048
- input: 0.1,
16049
- output: 0.3,
16050
- cacheRead: 0.01,
17828
+ input: 0.14,
17829
+ output: 0.28,
17830
+ cacheRead: 0.0028,
16051
17831
  cacheWrite: 0,
16052
17832
  },
16053
17833
  contextWindow: 262144,
@@ -16063,9 +17843,9 @@ export const MODELS = {
16063
17843
  reasoning: true,
16064
17844
  input: ["text", "image"],
16065
17845
  cost: {
16066
- input: 0.4,
16067
- output: 2,
16068
- cacheRead: 0.08,
17846
+ input: 0.14,
17847
+ output: 0.28,
17848
+ cacheRead: 0.0028,
16069
17849
  cacheWrite: 0,
16070
17850
  },
16071
17851
  contextWindow: 262144,
@@ -16081,9 +17861,9 @@ export const MODELS = {
16081
17861
  reasoning: true,
16082
17862
  input: ["text"],
16083
17863
  cost: {
16084
- input: 1,
16085
- output: 3,
16086
- cacheRead: 0.2,
17864
+ input: 0.435,
17865
+ output: 0.87,
17866
+ cacheRead: 0.0036,
16087
17867
  cacheWrite: 0,
16088
17868
  },
16089
17869
  contextWindow: 1048576,
@@ -16099,9 +17879,9 @@ export const MODELS = {
16099
17879
  reasoning: true,
16100
17880
  input: ["text", "image"],
16101
17881
  cost: {
16102
- input: 0.4,
16103
- output: 2,
16104
- cacheRead: 0.08,
17882
+ input: 0.14,
17883
+ output: 0.28,
17884
+ cacheRead: 0.0028,
16105
17885
  cacheWrite: 0,
16106
17886
  },
16107
17887
  contextWindow: 1048576,
@@ -16117,9 +17897,9 @@ export const MODELS = {
16117
17897
  reasoning: true,
16118
17898
  input: ["text"],
16119
17899
  cost: {
16120
- input: 1,
16121
- output: 3,
16122
- cacheRead: 0.2,
17900
+ input: 0.435,
17901
+ output: 0.87,
17902
+ cacheRead: 0.0036,
16123
17903
  cacheWrite: 0,
16124
17904
  },
16125
17905
  contextWindow: 1048576,
@@ -16155,9 +17935,9 @@ export const MODELS = {
16155
17935
  reasoning: true,
16156
17936
  input: ["text", "image"],
16157
17937
  cost: {
16158
- input: 0.4,
16159
- output: 2,
16160
- cacheRead: 0.08,
17938
+ input: 0.14,
17939
+ output: 0.28,
17940
+ cacheRead: 0.0028,
16161
17941
  cacheWrite: 0,
16162
17942
  },
16163
17943
  contextWindow: 262144,
@@ -16173,9 +17953,9 @@ export const MODELS = {
16173
17953
  reasoning: true,
16174
17954
  input: ["text"],
16175
17955
  cost: {
16176
- input: 1,
16177
- output: 3,
16178
- cacheRead: 0.2,
17956
+ input: 0.435,
17957
+ output: 0.87,
17958
+ cacheRead: 0.0036,
16179
17959
  cacheWrite: 0,
16180
17960
  },
16181
17961
  contextWindow: 1048576,
@@ -16191,9 +17971,9 @@ export const MODELS = {
16191
17971
  reasoning: true,
16192
17972
  input: ["text", "image"],
16193
17973
  cost: {
16194
- input: 0.4,
16195
- output: 2,
16196
- cacheRead: 0.08,
17974
+ input: 0.14,
17975
+ output: 0.28,
17976
+ cacheRead: 0.0028,
16197
17977
  cacheWrite: 0,
16198
17978
  },
16199
17979
  contextWindow: 1048576,
@@ -16209,9 +17989,9 @@ export const MODELS = {
16209
17989
  reasoning: true,
16210
17990
  input: ["text"],
16211
17991
  cost: {
16212
- input: 1,
16213
- output: 3,
16214
- cacheRead: 0.2,
17992
+ input: 0.435,
17993
+ output: 0.87,
17994
+ cacheRead: 0.0036,
16215
17995
  cacheWrite: 0,
16216
17996
  },
16217
17997
  contextWindow: 1048576,
@@ -16247,9 +18027,9 @@ export const MODELS = {
16247
18027
  reasoning: true,
16248
18028
  input: ["text", "image"],
16249
18029
  cost: {
16250
- input: 0.4,
16251
- output: 2,
16252
- cacheRead: 0.08,
18030
+ input: 0.14,
18031
+ output: 0.28,
18032
+ cacheRead: 0.0028,
16253
18033
  cacheWrite: 0,
16254
18034
  },
16255
18035
  contextWindow: 262144,
@@ -16265,9 +18045,9 @@ export const MODELS = {
16265
18045
  reasoning: true,
16266
18046
  input: ["text"],
16267
18047
  cost: {
16268
- input: 1,
16269
- output: 3,
16270
- cacheRead: 0.2,
18048
+ input: 0.435,
18049
+ output: 0.87,
18050
+ cacheRead: 0.0036,
16271
18051
  cacheWrite: 0,
16272
18052
  },
16273
18053
  contextWindow: 1048576,
@@ -16283,9 +18063,9 @@ export const MODELS = {
16283
18063
  reasoning: true,
16284
18064
  input: ["text", "image"],
16285
18065
  cost: {
16286
- input: 0.4,
16287
- output: 2,
16288
- cacheRead: 0.08,
18066
+ input: 0.14,
18067
+ output: 0.28,
18068
+ cacheRead: 0.0028,
16289
18069
  cacheWrite: 0,
16290
18070
  },
16291
18071
  contextWindow: 1048576,
@@ -16301,9 +18081,9 @@ export const MODELS = {
16301
18081
  reasoning: true,
16302
18082
  input: ["text"],
16303
18083
  cost: {
16304
- input: 1,
16305
- output: 3,
16306
- cacheRead: 0.2,
18084
+ input: 0.435,
18085
+ output: 0.87,
18086
+ cacheRead: 0.0036,
16307
18087
  cacheWrite: 0,
16308
18088
  },
16309
18089
  contextWindow: 1048576,
@@ -16339,9 +18119,9 @@ export const MODELS = {
16339
18119
  reasoning: true,
16340
18120
  input: ["text", "image"],
16341
18121
  cost: {
16342
- input: 0.4,
16343
- output: 2,
16344
- cacheRead: 0.08,
18122
+ input: 0.14,
18123
+ output: 0.28,
18124
+ cacheRead: 0.0028,
16345
18125
  cacheWrite: 0,
16346
18126
  },
16347
18127
  contextWindow: 262144,
@@ -16357,9 +18137,9 @@ export const MODELS = {
16357
18137
  reasoning: true,
16358
18138
  input: ["text"],
16359
18139
  cost: {
16360
- input: 1,
16361
- output: 3,
16362
- cacheRead: 0.2,
18140
+ input: 0.435,
18141
+ output: 0.87,
18142
+ cacheRead: 0.0036,
16363
18143
  cacheWrite: 0,
16364
18144
  },
16365
18145
  contextWindow: 1048576,
@@ -16375,9 +18155,9 @@ export const MODELS = {
16375
18155
  reasoning: true,
16376
18156
  input: ["text", "image"],
16377
18157
  cost: {
16378
- input: 0.4,
16379
- output: 2,
16380
- cacheRead: 0.08,
18158
+ input: 0.14,
18159
+ output: 0.28,
18160
+ cacheRead: 0.0028,
16381
18161
  cacheWrite: 0,
16382
18162
  },
16383
18163
  contextWindow: 1048576,
@@ -16393,9 +18173,9 @@ export const MODELS = {
16393
18173
  reasoning: true,
16394
18174
  input: ["text"],
16395
18175
  cost: {
16396
- input: 1,
16397
- output: 3,
16398
- cacheRead: 0.2,
18176
+ input: 0.435,
18177
+ output: 0.87,
18178
+ cacheRead: 0.0036,
16399
18179
  cacheWrite: 0,
16400
18180
  },
16401
18181
  contextWindow: 1048576,