@draht/ai 2026.4.5 → 2026.4.25-1

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 (62) hide show
  1. package/README.md +17 -8
  2. package/dist/cli.d.ts.map +1 -1
  3. package/dist/cli.js +1 -1
  4. package/dist/cli.js.map +1 -1
  5. package/dist/models.d.ts +7 -1
  6. package/dist/models.d.ts.map +1 -1
  7. package/dist/models.generated.d.ts +996 -519
  8. package/dist/models.generated.d.ts.map +1 -1
  9. package/dist/models.generated.js +1255 -765
  10. package/dist/models.generated.js.map +1 -1
  11. package/dist/models.js +20 -2
  12. package/dist/models.js.map +1 -1
  13. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  14. package/dist/providers/amazon-bedrock.js +46 -14
  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 +44 -19
  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 +2 -2
  21. package/dist/providers/azure-openai-responses.js.map +1 -1
  22. package/dist/providers/google-gemini-cli.d.ts.map +1 -1
  23. package/dist/providers/google-gemini-cli.js +2 -2
  24. package/dist/providers/google-gemini-cli.js.map +1 -1
  25. package/dist/providers/google-vertex.d.ts.map +1 -1
  26. package/dist/providers/google-vertex.js +1 -1
  27. package/dist/providers/google-vertex.js.map +1 -1
  28. package/dist/providers/google.d.ts.map +1 -1
  29. package/dist/providers/google.js +29 -4
  30. package/dist/providers/google.js.map +1 -1
  31. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  32. package/dist/providers/openai-codex-responses.js +2 -2
  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 +29 -16
  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 +8 -1
  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 +4 -4
  42. package/dist/providers/openai-responses.js.map +1 -1
  43. package/dist/providers/simple-options.d.ts +6 -1
  44. package/dist/providers/simple-options.d.ts.map +1 -1
  45. package/dist/providers/simple-options.js +10 -1
  46. package/dist/providers/simple-options.js.map +1 -1
  47. package/dist/types.d.ts +63 -6
  48. package/dist/types.d.ts.map +1 -1
  49. package/dist/types.js.map +1 -1
  50. package/dist/utils/oauth/index.d.ts +1 -0
  51. package/dist/utils/oauth/index.d.ts.map +1 -1
  52. package/dist/utils/oauth/index.js +4 -0
  53. package/dist/utils/oauth/index.js.map +1 -1
  54. package/dist/utils/oauth/opencode-go.d.ts +23 -0
  55. package/dist/utils/oauth/opencode-go.d.ts.map +1 -0
  56. package/dist/utils/oauth/opencode-go.js +46 -0
  57. package/dist/utils/oauth/opencode-go.js.map +1 -0
  58. package/dist/utils/overflow.d.ts +3 -2
  59. package/dist/utils/overflow.d.ts.map +1 -1
  60. package/dist/utils/overflow.js +24 -11
  61. package/dist/utils/overflow.js.map +1 -1
  62. package/package.json +5 -8
@@ -257,6 +257,23 @@ export const MODELS = {
257
257
  contextWindow: 1000000,
258
258
  maxTokens: 128000,
259
259
  },
260
+ "anthropic.claude-opus-4-7": {
261
+ id: "anthropic.claude-opus-4-7",
262
+ name: "Claude Opus 4.7",
263
+ api: "bedrock-converse-stream",
264
+ provider: "amazon-bedrock",
265
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
266
+ reasoning: true,
267
+ input: ["text", "image"],
268
+ cost: {
269
+ input: 5,
270
+ output: 25,
271
+ cacheRead: 0.5,
272
+ cacheWrite: 6.25,
273
+ },
274
+ contextWindow: 1000000,
275
+ maxTokens: 128000,
276
+ },
260
277
  "anthropic.claude-sonnet-4-20250514-v1:0": {
261
278
  id: "anthropic.claude-sonnet-4-20250514-v1:0",
262
279
  name: "Claude Sonnet 4",
@@ -308,6 +325,40 @@ export const MODELS = {
308
325
  contextWindow: 1000000,
309
326
  maxTokens: 64000,
310
327
  },
328
+ "au.anthropic.claude-opus-4-6-v1": {
329
+ id: "au.anthropic.claude-opus-4-6-v1",
330
+ name: "AU Anthropic Claude Opus 4.6",
331
+ api: "bedrock-converse-stream",
332
+ provider: "amazon-bedrock",
333
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
334
+ reasoning: true,
335
+ input: ["text", "image"],
336
+ cost: {
337
+ input: 16.5,
338
+ output: 82.5,
339
+ cacheRead: 0.5,
340
+ cacheWrite: 6.25,
341
+ },
342
+ contextWindow: 1000000,
343
+ maxTokens: 128000,
344
+ },
345
+ "au.anthropic.claude-sonnet-4-6": {
346
+ id: "au.anthropic.claude-sonnet-4-6",
347
+ name: "AU Anthropic Claude Sonnet 4.6",
348
+ api: "bedrock-converse-stream",
349
+ provider: "amazon-bedrock",
350
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
351
+ reasoning: true,
352
+ input: ["text", "image"],
353
+ cost: {
354
+ input: 3.3,
355
+ output: 16.5,
356
+ cacheRead: 0.33,
357
+ cacheWrite: 4.125,
358
+ },
359
+ contextWindow: 1000000,
360
+ maxTokens: 128000,
361
+ },
311
362
  "deepseek.r1-v1:0": {
312
363
  id: "deepseek.r1-v1:0",
313
364
  name: "DeepSeek-R1",
@@ -410,6 +461,23 @@ export const MODELS = {
410
461
  contextWindow: 1000000,
411
462
  maxTokens: 128000,
412
463
  },
464
+ "eu.anthropic.claude-opus-4-7": {
465
+ id: "eu.anthropic.claude-opus-4-7",
466
+ name: "Claude Opus 4.7 (EU)",
467
+ api: "bedrock-converse-stream",
468
+ provider: "amazon-bedrock",
469
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
470
+ reasoning: true,
471
+ input: ["text", "image"],
472
+ cost: {
473
+ input: 5,
474
+ output: 25,
475
+ cacheRead: 0.5,
476
+ cacheWrite: 6.25,
477
+ },
478
+ contextWindow: 1000000,
479
+ maxTokens: 128000,
480
+ },
413
481
  "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
414
482
  id: "eu.anthropic.claude-sonnet-4-20250514-v1:0",
415
483
  name: "Claude Sonnet 4 (EU)",
@@ -512,6 +580,23 @@ export const MODELS = {
512
580
  contextWindow: 1000000,
513
581
  maxTokens: 128000,
514
582
  },
583
+ "global.anthropic.claude-opus-4-7": {
584
+ id: "global.anthropic.claude-opus-4-7",
585
+ name: "Claude Opus 4.7 (Global)",
586
+ api: "bedrock-converse-stream",
587
+ provider: "amazon-bedrock",
588
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
589
+ reasoning: true,
590
+ input: ["text", "image"],
591
+ cost: {
592
+ input: 5,
593
+ output: 25,
594
+ cacheRead: 0.5,
595
+ cacheWrite: 6.25,
596
+ },
597
+ contextWindow: 1000000,
598
+ maxTokens: 128000,
599
+ },
515
600
  "global.anthropic.claude-sonnet-4-20250514-v1:0": {
516
601
  id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
517
602
  name: "Claude Sonnet 4 (Global)",
@@ -1209,6 +1294,23 @@ export const MODELS = {
1209
1294
  contextWindow: 131072,
1210
1295
  maxTokens: 65536,
1211
1296
  },
1297
+ "qwen.qwen3-coder-next": {
1298
+ id: "qwen.qwen3-coder-next",
1299
+ name: "Qwen3 Coder Next",
1300
+ api: "bedrock-converse-stream",
1301
+ provider: "amazon-bedrock",
1302
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1303
+ reasoning: true,
1304
+ input: ["text"],
1305
+ cost: {
1306
+ input: 0.22,
1307
+ output: 1.8,
1308
+ cacheRead: 0,
1309
+ cacheWrite: 0,
1310
+ },
1311
+ contextWindow: 131072,
1312
+ maxTokens: 65536,
1313
+ },
1212
1314
  "qwen.qwen3-next-80b-a3b": {
1213
1315
  id: "qwen.qwen3-next-80b-a3b",
1214
1316
  name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
@@ -1328,6 +1430,23 @@ export const MODELS = {
1328
1430
  contextWindow: 1000000,
1329
1431
  maxTokens: 128000,
1330
1432
  },
1433
+ "us.anthropic.claude-opus-4-7": {
1434
+ id: "us.anthropic.claude-opus-4-7",
1435
+ name: "Claude Opus 4.7 (US)",
1436
+ api: "bedrock-converse-stream",
1437
+ provider: "amazon-bedrock",
1438
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1439
+ reasoning: true,
1440
+ input: ["text", "image"],
1441
+ cost: {
1442
+ input: 5,
1443
+ output: 25,
1444
+ cacheRead: 0.5,
1445
+ cacheWrite: 6.25,
1446
+ },
1447
+ contextWindow: 1000000,
1448
+ maxTokens: 128000,
1449
+ },
1331
1450
  "us.anthropic.claude-sonnet-4-20250514-v1:0": {
1332
1451
  id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
1333
1452
  name: "Claude Sonnet 4 (US)",
@@ -1755,6 +1874,23 @@ export const MODELS = {
1755
1874
  contextWindow: 1000000,
1756
1875
  maxTokens: 128000,
1757
1876
  },
1877
+ "claude-opus-4-7": {
1878
+ id: "claude-opus-4-7",
1879
+ name: "Claude Opus 4.7",
1880
+ api: "anthropic-messages",
1881
+ provider: "anthropic",
1882
+ baseUrl: "https://api.anthropic.com",
1883
+ reasoning: true,
1884
+ input: ["text", "image"],
1885
+ cost: {
1886
+ input: 5,
1887
+ output: 25,
1888
+ cacheRead: 0.5,
1889
+ cacheWrite: 6.25,
1890
+ },
1891
+ contextWindow: 1000000,
1892
+ maxTokens: 128000,
1893
+ },
1758
1894
  "claude-sonnet-4-0": {
1759
1895
  id: "claude-sonnet-4-0",
1760
1896
  name: "Claude Sonnet 4 (latest)",
@@ -1842,23 +1978,6 @@ export const MODELS = {
1842
1978
  },
1843
1979
  },
1844
1980
  "azure-openai-responses": {
1845
- "codex-mini-latest": {
1846
- id: "codex-mini-latest",
1847
- name: "Codex Mini",
1848
- api: "azure-openai-responses",
1849
- provider: "azure-openai-responses",
1850
- baseUrl: "",
1851
- reasoning: true,
1852
- input: ["text"],
1853
- cost: {
1854
- input: 1.5,
1855
- output: 6,
1856
- cacheRead: 0.375,
1857
- cacheWrite: 0,
1858
- },
1859
- contextWindow: 200000,
1860
- maxTokens: 100000,
1861
- },
1862
1981
  "gpt-4": {
1863
1982
  id: "gpt-4",
1864
1983
  name: "GPT-4",
@@ -2403,6 +2522,23 @@ export const MODELS = {
2403
2522
  contextWindow: 1050000,
2404
2523
  maxTokens: 128000,
2405
2524
  },
2525
+ "gpt-5.5": {
2526
+ id: "gpt-5.5",
2527
+ name: "GPT-5.5",
2528
+ api: "azure-openai-responses",
2529
+ provider: "azure-openai-responses",
2530
+ baseUrl: "",
2531
+ reasoning: true,
2532
+ input: ["text", "image"],
2533
+ cost: {
2534
+ input: 5,
2535
+ output: 30,
2536
+ cacheRead: 0.5,
2537
+ cacheWrite: 0,
2538
+ },
2539
+ contextWindow: 1050000,
2540
+ maxTokens: 130000,
2541
+ },
2406
2542
  "o1": {
2407
2543
  id: "o1",
2408
2544
  name: "o1",
@@ -2665,6 +2801,24 @@ export const MODELS = {
2665
2801
  contextWindow: 1000000,
2666
2802
  maxTokens: 64000,
2667
2803
  },
2804
+ "claude-opus-4.7": {
2805
+ id: "claude-opus-4.7",
2806
+ name: "Claude Opus 4.7",
2807
+ api: "anthropic-messages",
2808
+ provider: "github-copilot",
2809
+ baseUrl: "https://api.individual.githubcopilot.com",
2810
+ 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" },
2811
+ reasoning: true,
2812
+ input: ["text", "image"],
2813
+ cost: {
2814
+ input: 0,
2815
+ output: 0,
2816
+ cacheRead: 0,
2817
+ cacheWrite: 0,
2818
+ },
2819
+ contextWindow: 144000,
2820
+ maxTokens: 64000,
2821
+ },
2668
2822
  "claude-sonnet-4": {
2669
2823
  id: "claude-sonnet-4",
2670
2824
  name: "Claude Sonnet 4",
@@ -3015,7 +3169,25 @@ export const MODELS = {
3015
3169
  },
3016
3170
  "gpt-5.4-mini": {
3017
3171
  id: "gpt-5.4-mini",
3018
- name: "GPT-5.4 mini",
3172
+ name: "GPT-5.4 Mini",
3173
+ api: "openai-responses",
3174
+ provider: "github-copilot",
3175
+ baseUrl: "https://api.individual.githubcopilot.com",
3176
+ 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" },
3177
+ reasoning: true,
3178
+ input: ["text", "image"],
3179
+ cost: {
3180
+ input: 0,
3181
+ output: 0,
3182
+ cacheRead: 0,
3183
+ cacheWrite: 0,
3184
+ },
3185
+ contextWindow: 400000,
3186
+ maxTokens: 128000,
3187
+ },
3188
+ "gpt-5.5": {
3189
+ id: "gpt-5.5",
3190
+ name: "GPT-5.5",
3019
3191
  api: "openai-responses",
3020
3192
  provider: "github-copilot",
3021
3193
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3148,7 +3320,7 @@ export const MODELS = {
3148
3320
  cost: {
3149
3321
  input: 0.3,
3150
3322
  output: 2.5,
3151
- cacheRead: 0.075,
3323
+ cacheRead: 0.03,
3152
3324
  cacheWrite: 0,
3153
3325
  },
3154
3326
  contextWindow: 1048576,
@@ -3267,7 +3439,7 @@ export const MODELS = {
3267
3439
  cost: {
3268
3440
  input: 1.25,
3269
3441
  output: 10,
3270
- cacheRead: 0.31,
3442
+ cacheRead: 0.125,
3271
3443
  cacheWrite: 0,
3272
3444
  },
3273
3445
  contextWindow: 1048576,
@@ -3477,8 +3649,8 @@ export const MODELS = {
3477
3649
  contextWindow: 131072,
3478
3650
  maxTokens: 8192,
3479
3651
  },
3480
- "gemma-4-26b": {
3481
- id: "gemma-4-26b",
3652
+ "gemma-4-26b-a4b-it": {
3653
+ id: "gemma-4-26b-a4b-it",
3482
3654
  name: "Gemma 4 26B",
3483
3655
  api: "google-generative-ai",
3484
3656
  provider: "google",
@@ -3494,8 +3666,8 @@ export const MODELS = {
3494
3666
  contextWindow: 256000,
3495
3667
  maxTokens: 8192,
3496
3668
  },
3497
- "gemma-4-31b": {
3498
- id: "gemma-4-31b",
3669
+ "gemma-4-31b-it": {
3670
+ id: "gemma-4-31b-it",
3499
3671
  name: "Gemma 4 31B",
3500
3672
  api: "google-generative-ai",
3501
3673
  provider: "google",
@@ -3976,6 +4148,23 @@ export const MODELS = {
3976
4148
  contextWindow: 1048576,
3977
4149
  maxTokens: 65536,
3978
4150
  },
4151
+ "gemini-3.1-pro-preview-customtools": {
4152
+ id: "gemini-3.1-pro-preview-customtools",
4153
+ name: "Gemini 3.1 Pro Preview Custom Tools (Vertex)",
4154
+ api: "google-vertex",
4155
+ provider: "google-vertex",
4156
+ baseUrl: "https://{location}-aiplatform.googleapis.com",
4157
+ reasoning: true,
4158
+ input: ["text", "image"],
4159
+ cost: {
4160
+ input: 2,
4161
+ output: 12,
4162
+ cacheRead: 0.2,
4163
+ cacheWrite: 0,
4164
+ },
4165
+ contextWindow: 1048576,
4166
+ maxTokens: 65536,
4167
+ },
3979
4168
  },
3980
4169
  "groq": {
3981
4170
  "deepseek-r1-distill-llama-70b": {
@@ -4322,6 +4511,24 @@ export const MODELS = {
4322
4511
  contextWindow: 204800,
4323
4512
  maxTokens: 131072,
4324
4513
  },
4514
+ "MiniMaxAI/MiniMax-M2.7": {
4515
+ id: "MiniMaxAI/MiniMax-M2.7",
4516
+ name: "MiniMax-M2.7",
4517
+ api: "openai-completions",
4518
+ provider: "huggingface",
4519
+ baseUrl: "https://router.huggingface.co/v1",
4520
+ compat: { "supportsDeveloperRole": false },
4521
+ reasoning: true,
4522
+ input: ["text"],
4523
+ cost: {
4524
+ input: 0.3,
4525
+ output: 1.2,
4526
+ cacheRead: 0.06,
4527
+ cacheWrite: 0,
4528
+ },
4529
+ contextWindow: 204800,
4530
+ maxTokens: 131072,
4531
+ },
4325
4532
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4326
4533
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4327
4534
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4556,6 +4763,24 @@ export const MODELS = {
4556
4763
  contextWindow: 262144,
4557
4764
  maxTokens: 262144,
4558
4765
  },
4766
+ "moonshotai/Kimi-K2.6": {
4767
+ id: "moonshotai/Kimi-K2.6",
4768
+ name: "Kimi-K2.6",
4769
+ api: "openai-completions",
4770
+ provider: "huggingface",
4771
+ baseUrl: "https://router.huggingface.co/v1",
4772
+ compat: { "supportsDeveloperRole": false },
4773
+ reasoning: true,
4774
+ input: ["text", "image"],
4775
+ cost: {
4776
+ input: 0.95,
4777
+ output: 4,
4778
+ cacheRead: 0.16,
4779
+ cacheWrite: 0,
4780
+ },
4781
+ contextWindow: 262144,
4782
+ maxTokens: 262144,
4783
+ },
4559
4784
  "zai-org/GLM-4.7": {
4560
4785
  id: "zai-org/GLM-4.7",
4561
4786
  name: "GLM-4.7",
@@ -4610,6 +4835,24 @@ export const MODELS = {
4610
4835
  contextWindow: 202752,
4611
4836
  maxTokens: 131072,
4612
4837
  },
4838
+ "zai-org/GLM-5.1": {
4839
+ id: "zai-org/GLM-5.1",
4840
+ name: "GLM-5.1",
4841
+ api: "openai-completions",
4842
+ provider: "huggingface",
4843
+ baseUrl: "https://router.huggingface.co/v1",
4844
+ compat: { "supportsDeveloperRole": false },
4845
+ reasoning: true,
4846
+ input: ["text"],
4847
+ cost: {
4848
+ input: 1,
4849
+ output: 3.2,
4850
+ cacheRead: 0.2,
4851
+ cacheWrite: 0,
4852
+ },
4853
+ contextWindow: 202752,
4854
+ maxTokens: 131072,
4855
+ },
4613
4856
  },
4614
4857
  "kimi-coding": {
4615
4858
  "k2p5": {
@@ -4629,6 +4872,23 @@ export const MODELS = {
4629
4872
  contextWindow: 262144,
4630
4873
  maxTokens: 32768,
4631
4874
  },
4875
+ "k2p6": {
4876
+ id: "k2p6",
4877
+ name: "Kimi K2.6",
4878
+ api: "anthropic-messages",
4879
+ provider: "kimi-coding",
4880
+ baseUrl: "https://api.kimi.com/coding",
4881
+ reasoning: true,
4882
+ input: ["text", "image"],
4883
+ cost: {
4884
+ input: 0,
4885
+ output: 0,
4886
+ cacheRead: 0,
4887
+ cacheWrite: 0,
4888
+ },
4889
+ contextWindow: 262144,
4890
+ maxTokens: 32768,
4891
+ },
4632
4892
  "kimi-k2-thinking": {
4633
4893
  id: "kimi-k2-thinking",
4634
4894
  name: "Kimi K2 Thinking",
@@ -4648,9 +4908,9 @@ export const MODELS = {
4648
4908
  },
4649
4909
  },
4650
4910
  "minimax": {
4651
- "MiniMax-M2": {
4652
- id: "MiniMax-M2",
4653
- name: "MiniMax-M2",
4911
+ "MiniMax-M2.7": {
4912
+ id: "MiniMax-M2.7",
4913
+ name: "MiniMax-M2.7",
4654
4914
  api: "anthropic-messages",
4655
4915
  provider: "minimax",
4656
4916
  baseUrl: "https://api.minimax.io/anthropic",
@@ -4659,100 +4919,15 @@ export const MODELS = {
4659
4919
  cost: {
4660
4920
  input: 0.3,
4661
4921
  output: 1.2,
4662
- cacheRead: 0,
4663
- cacheWrite: 0,
4922
+ cacheRead: 0.06,
4923
+ cacheWrite: 0.375,
4664
4924
  },
4665
4925
  contextWindow: 204800,
4666
4926
  maxTokens: 131072,
4667
4927
  },
4668
- "MiniMax-M2.1": {
4669
- id: "MiniMax-M2.1",
4670
- name: "MiniMax-M2.1",
4671
- api: "anthropic-messages",
4672
- provider: "minimax",
4673
- baseUrl: "https://api.minimax.io/anthropic",
4674
- reasoning: true,
4675
- input: ["text"],
4676
- cost: {
4677
- input: 0.3,
4678
- output: 1.2,
4679
- cacheRead: 0,
4680
- cacheWrite: 0,
4681
- },
4682
- contextWindow: 204800,
4683
- maxTokens: 131072,
4684
- },
4685
- "MiniMax-M2.1-highspeed": {
4686
- id: "MiniMax-M2.1-highspeed",
4687
- name: "MiniMax-M2.1-highspeed",
4688
- api: "anthropic-messages",
4689
- provider: "minimax",
4690
- baseUrl: "https://api.minimax.io/anthropic",
4691
- reasoning: true,
4692
- input: ["text"],
4693
- cost: {
4694
- input: 0.6,
4695
- output: 2.4,
4696
- cacheRead: 0,
4697
- cacheWrite: 0,
4698
- },
4699
- contextWindow: 204800,
4700
- maxTokens: 131072,
4701
- },
4702
- "MiniMax-M2.5": {
4703
- id: "MiniMax-M2.5",
4704
- name: "MiniMax-M2.5",
4705
- api: "anthropic-messages",
4706
- provider: "minimax",
4707
- baseUrl: "https://api.minimax.io/anthropic",
4708
- reasoning: true,
4709
- input: ["text"],
4710
- cost: {
4711
- input: 0.3,
4712
- output: 1.2,
4713
- cacheRead: 0.03,
4714
- cacheWrite: 0.375,
4715
- },
4716
- contextWindow: 204800,
4717
- maxTokens: 131072,
4718
- },
4719
- "MiniMax-M2.5-highspeed": {
4720
- id: "MiniMax-M2.5-highspeed",
4721
- name: "MiniMax-M2.5-highspeed",
4722
- api: "anthropic-messages",
4723
- provider: "minimax",
4724
- baseUrl: "https://api.minimax.io/anthropic",
4725
- reasoning: true,
4726
- input: ["text"],
4727
- cost: {
4728
- input: 0.6,
4729
- output: 2.4,
4730
- cacheRead: 0.06,
4731
- cacheWrite: 0.375,
4732
- },
4733
- contextWindow: 204800,
4734
- maxTokens: 131072,
4735
- },
4736
- "MiniMax-M2.7": {
4737
- id: "MiniMax-M2.7",
4738
- name: "MiniMax-M2.7",
4739
- api: "anthropic-messages",
4740
- provider: "minimax",
4741
- baseUrl: "https://api.minimax.io/anthropic",
4742
- reasoning: true,
4743
- input: ["text"],
4744
- cost: {
4745
- input: 0.3,
4746
- output: 1.2,
4747
- cacheRead: 0.06,
4748
- cacheWrite: 0.375,
4749
- },
4750
- contextWindow: 204800,
4751
- maxTokens: 131072,
4752
- },
4753
- "MiniMax-M2.7-highspeed": {
4754
- id: "MiniMax-M2.7-highspeed",
4755
- name: "MiniMax-M2.7-highspeed",
4928
+ "MiniMax-M2.7-highspeed": {
4929
+ id: "MiniMax-M2.7-highspeed",
4930
+ name: "MiniMax-M2.7-highspeed",
4756
4931
  api: "anthropic-messages",
4757
4932
  provider: "minimax",
4758
4933
  baseUrl: "https://api.minimax.io/anthropic",
@@ -4769,91 +4944,6 @@ export const MODELS = {
4769
4944
  },
4770
4945
  },
4771
4946
  "minimax-cn": {
4772
- "MiniMax-M2": {
4773
- id: "MiniMax-M2",
4774
- name: "MiniMax-M2",
4775
- api: "anthropic-messages",
4776
- provider: "minimax-cn",
4777
- baseUrl: "https://api.minimaxi.com/anthropic",
4778
- reasoning: true,
4779
- input: ["text"],
4780
- cost: {
4781
- input: 0.3,
4782
- output: 1.2,
4783
- cacheRead: 0,
4784
- cacheWrite: 0,
4785
- },
4786
- contextWindow: 204800,
4787
- maxTokens: 131072,
4788
- },
4789
- "MiniMax-M2.1": {
4790
- id: "MiniMax-M2.1",
4791
- name: "MiniMax-M2.1",
4792
- api: "anthropic-messages",
4793
- provider: "minimax-cn",
4794
- baseUrl: "https://api.minimaxi.com/anthropic",
4795
- reasoning: true,
4796
- input: ["text"],
4797
- cost: {
4798
- input: 0.3,
4799
- output: 1.2,
4800
- cacheRead: 0,
4801
- cacheWrite: 0,
4802
- },
4803
- contextWindow: 204800,
4804
- maxTokens: 131072,
4805
- },
4806
- "MiniMax-M2.1-highspeed": {
4807
- id: "MiniMax-M2.1-highspeed",
4808
- name: "MiniMax-M2.1-highspeed",
4809
- api: "anthropic-messages",
4810
- provider: "minimax-cn",
4811
- baseUrl: "https://api.minimaxi.com/anthropic",
4812
- reasoning: true,
4813
- input: ["text"],
4814
- cost: {
4815
- input: 0.6,
4816
- output: 2.4,
4817
- cacheRead: 0,
4818
- cacheWrite: 0,
4819
- },
4820
- contextWindow: 204800,
4821
- maxTokens: 131072,
4822
- },
4823
- "MiniMax-M2.5": {
4824
- id: "MiniMax-M2.5",
4825
- name: "MiniMax-M2.5",
4826
- api: "anthropic-messages",
4827
- provider: "minimax-cn",
4828
- baseUrl: "https://api.minimaxi.com/anthropic",
4829
- reasoning: true,
4830
- input: ["text"],
4831
- cost: {
4832
- input: 0.3,
4833
- output: 1.2,
4834
- cacheRead: 0.03,
4835
- cacheWrite: 0.375,
4836
- },
4837
- contextWindow: 204800,
4838
- maxTokens: 131072,
4839
- },
4840
- "MiniMax-M2.5-highspeed": {
4841
- id: "MiniMax-M2.5-highspeed",
4842
- name: "MiniMax-M2.5-highspeed",
4843
- api: "anthropic-messages",
4844
- provider: "minimax-cn",
4845
- baseUrl: "https://api.minimaxi.com/anthropic",
4846
- reasoning: true,
4847
- input: ["text"],
4848
- cost: {
4849
- input: 0.6,
4850
- output: 2.4,
4851
- cacheRead: 0.06,
4852
- cacheWrite: 0.375,
4853
- },
4854
- contextWindow: 204800,
4855
- maxTokens: 131072,
4856
- },
4857
4947
  "MiniMax-M2.7": {
4858
4948
  id: "MiniMax-M2.7",
4859
4949
  name: "MiniMax-M2.7",
@@ -4893,9 +4983,9 @@ export const MODELS = {
4893
4983
  "codestral-latest": {
4894
4984
  id: "codestral-latest",
4895
4985
  name: "Codestral (latest)",
4896
- api: "mistral-conversations",
4986
+ api: "openai-completions",
4897
4987
  provider: "mistral",
4898
- baseUrl: "https://api.mistral.ai",
4988
+ baseUrl: "https://api.mistral.ai/v1",
4899
4989
  reasoning: false,
4900
4990
  input: ["text"],
4901
4991
  cost: {
@@ -4910,9 +5000,9 @@ export const MODELS = {
4910
5000
  "devstral-2512": {
4911
5001
  id: "devstral-2512",
4912
5002
  name: "Devstral 2",
4913
- api: "mistral-conversations",
5003
+ api: "openai-completions",
4914
5004
  provider: "mistral",
4915
- baseUrl: "https://api.mistral.ai",
5005
+ baseUrl: "https://api.mistral.ai/v1",
4916
5006
  reasoning: false,
4917
5007
  input: ["text"],
4918
5008
  cost: {
@@ -4927,9 +5017,9 @@ export const MODELS = {
4927
5017
  "devstral-medium-2507": {
4928
5018
  id: "devstral-medium-2507",
4929
5019
  name: "Devstral Medium",
4930
- api: "mistral-conversations",
5020
+ api: "openai-completions",
4931
5021
  provider: "mistral",
4932
- baseUrl: "https://api.mistral.ai",
5022
+ baseUrl: "https://api.mistral.ai/v1",
4933
5023
  reasoning: false,
4934
5024
  input: ["text"],
4935
5025
  cost: {
@@ -4944,9 +5034,9 @@ export const MODELS = {
4944
5034
  "devstral-medium-latest": {
4945
5035
  id: "devstral-medium-latest",
4946
5036
  name: "Devstral 2 (latest)",
4947
- api: "mistral-conversations",
5037
+ api: "openai-completions",
4948
5038
  provider: "mistral",
4949
- baseUrl: "https://api.mistral.ai",
5039
+ baseUrl: "https://api.mistral.ai/v1",
4950
5040
  reasoning: false,
4951
5041
  input: ["text"],
4952
5042
  cost: {
@@ -4961,9 +5051,9 @@ export const MODELS = {
4961
5051
  "devstral-small-2505": {
4962
5052
  id: "devstral-small-2505",
4963
5053
  name: "Devstral Small 2505",
4964
- api: "mistral-conversations",
5054
+ api: "openai-completions",
4965
5055
  provider: "mistral",
4966
- baseUrl: "https://api.mistral.ai",
5056
+ baseUrl: "https://api.mistral.ai/v1",
4967
5057
  reasoning: false,
4968
5058
  input: ["text"],
4969
5059
  cost: {
@@ -4978,9 +5068,9 @@ export const MODELS = {
4978
5068
  "devstral-small-2507": {
4979
5069
  id: "devstral-small-2507",
4980
5070
  name: "Devstral Small",
4981
- api: "mistral-conversations",
5071
+ api: "openai-completions",
4982
5072
  provider: "mistral",
4983
- baseUrl: "https://api.mistral.ai",
5073
+ baseUrl: "https://api.mistral.ai/v1",
4984
5074
  reasoning: false,
4985
5075
  input: ["text"],
4986
5076
  cost: {
@@ -4995,9 +5085,9 @@ export const MODELS = {
4995
5085
  "labs-devstral-small-2512": {
4996
5086
  id: "labs-devstral-small-2512",
4997
5087
  name: "Devstral Small 2",
4998
- api: "mistral-conversations",
5088
+ api: "openai-completions",
4999
5089
  provider: "mistral",
5000
- baseUrl: "https://api.mistral.ai",
5090
+ baseUrl: "https://api.mistral.ai/v1",
5001
5091
  reasoning: false,
5002
5092
  input: ["text", "image"],
5003
5093
  cost: {
@@ -5012,9 +5102,9 @@ export const MODELS = {
5012
5102
  "magistral-medium-latest": {
5013
5103
  id: "magistral-medium-latest",
5014
5104
  name: "Magistral Medium (latest)",
5015
- api: "mistral-conversations",
5105
+ api: "openai-completions",
5016
5106
  provider: "mistral",
5017
- baseUrl: "https://api.mistral.ai",
5107
+ baseUrl: "https://api.mistral.ai/v1",
5018
5108
  reasoning: true,
5019
5109
  input: ["text"],
5020
5110
  cost: {
@@ -5029,9 +5119,9 @@ export const MODELS = {
5029
5119
  "magistral-small": {
5030
5120
  id: "magistral-small",
5031
5121
  name: "Magistral Small",
5032
- api: "mistral-conversations",
5122
+ api: "openai-completions",
5033
5123
  provider: "mistral",
5034
- baseUrl: "https://api.mistral.ai",
5124
+ baseUrl: "https://api.mistral.ai/v1",
5035
5125
  reasoning: true,
5036
5126
  input: ["text"],
5037
5127
  cost: {
@@ -5046,9 +5136,9 @@ export const MODELS = {
5046
5136
  "ministral-3b-latest": {
5047
5137
  id: "ministral-3b-latest",
5048
5138
  name: "Ministral 3B (latest)",
5049
- api: "mistral-conversations",
5139
+ api: "openai-completions",
5050
5140
  provider: "mistral",
5051
- baseUrl: "https://api.mistral.ai",
5141
+ baseUrl: "https://api.mistral.ai/v1",
5052
5142
  reasoning: false,
5053
5143
  input: ["text"],
5054
5144
  cost: {
@@ -5063,9 +5153,9 @@ export const MODELS = {
5063
5153
  "ministral-8b-latest": {
5064
5154
  id: "ministral-8b-latest",
5065
5155
  name: "Ministral 8B (latest)",
5066
- api: "mistral-conversations",
5156
+ api: "openai-completions",
5067
5157
  provider: "mistral",
5068
- baseUrl: "https://api.mistral.ai",
5158
+ baseUrl: "https://api.mistral.ai/v1",
5069
5159
  reasoning: false,
5070
5160
  input: ["text"],
5071
5161
  cost: {
@@ -5080,9 +5170,9 @@ export const MODELS = {
5080
5170
  "mistral-large-2411": {
5081
5171
  id: "mistral-large-2411",
5082
5172
  name: "Mistral Large 2.1",
5083
- api: "mistral-conversations",
5173
+ api: "openai-completions",
5084
5174
  provider: "mistral",
5085
- baseUrl: "https://api.mistral.ai",
5175
+ baseUrl: "https://api.mistral.ai/v1",
5086
5176
  reasoning: false,
5087
5177
  input: ["text"],
5088
5178
  cost: {
@@ -5097,9 +5187,9 @@ export const MODELS = {
5097
5187
  "mistral-large-2512": {
5098
5188
  id: "mistral-large-2512",
5099
5189
  name: "Mistral Large 3",
5100
- api: "mistral-conversations",
5190
+ api: "openai-completions",
5101
5191
  provider: "mistral",
5102
- baseUrl: "https://api.mistral.ai",
5192
+ baseUrl: "https://api.mistral.ai/v1",
5103
5193
  reasoning: false,
5104
5194
  input: ["text", "image"],
5105
5195
  cost: {
@@ -5114,9 +5204,9 @@ export const MODELS = {
5114
5204
  "mistral-large-latest": {
5115
5205
  id: "mistral-large-latest",
5116
5206
  name: "Mistral Large (latest)",
5117
- api: "mistral-conversations",
5207
+ api: "openai-completions",
5118
5208
  provider: "mistral",
5119
- baseUrl: "https://api.mistral.ai",
5209
+ baseUrl: "https://api.mistral.ai/v1",
5120
5210
  reasoning: false,
5121
5211
  input: ["text", "image"],
5122
5212
  cost: {
@@ -5131,9 +5221,9 @@ export const MODELS = {
5131
5221
  "mistral-medium-2505": {
5132
5222
  id: "mistral-medium-2505",
5133
5223
  name: "Mistral Medium 3",
5134
- api: "mistral-conversations",
5224
+ api: "openai-completions",
5135
5225
  provider: "mistral",
5136
- baseUrl: "https://api.mistral.ai",
5226
+ baseUrl: "https://api.mistral.ai/v1",
5137
5227
  reasoning: false,
5138
5228
  input: ["text", "image"],
5139
5229
  cost: {
@@ -5148,9 +5238,9 @@ export const MODELS = {
5148
5238
  "mistral-medium-2508": {
5149
5239
  id: "mistral-medium-2508",
5150
5240
  name: "Mistral Medium 3.1",
5151
- api: "mistral-conversations",
5241
+ api: "openai-completions",
5152
5242
  provider: "mistral",
5153
- baseUrl: "https://api.mistral.ai",
5243
+ baseUrl: "https://api.mistral.ai/v1",
5154
5244
  reasoning: false,
5155
5245
  input: ["text", "image"],
5156
5246
  cost: {
@@ -5165,9 +5255,9 @@ export const MODELS = {
5165
5255
  "mistral-medium-latest": {
5166
5256
  id: "mistral-medium-latest",
5167
5257
  name: "Mistral Medium (latest)",
5168
- api: "mistral-conversations",
5258
+ api: "openai-completions",
5169
5259
  provider: "mistral",
5170
- baseUrl: "https://api.mistral.ai",
5260
+ baseUrl: "https://api.mistral.ai/v1",
5171
5261
  reasoning: false,
5172
5262
  input: ["text", "image"],
5173
5263
  cost: {
@@ -5182,9 +5272,9 @@ export const MODELS = {
5182
5272
  "mistral-nemo": {
5183
5273
  id: "mistral-nemo",
5184
5274
  name: "Mistral Nemo",
5185
- api: "mistral-conversations",
5275
+ api: "openai-completions",
5186
5276
  provider: "mistral",
5187
- baseUrl: "https://api.mistral.ai",
5277
+ baseUrl: "https://api.mistral.ai/v1",
5188
5278
  reasoning: false,
5189
5279
  input: ["text"],
5190
5280
  cost: {
@@ -5199,9 +5289,9 @@ export const MODELS = {
5199
5289
  "mistral-small-2506": {
5200
5290
  id: "mistral-small-2506",
5201
5291
  name: "Mistral Small 3.2",
5202
- api: "mistral-conversations",
5292
+ api: "openai-completions",
5203
5293
  provider: "mistral",
5204
- baseUrl: "https://api.mistral.ai",
5294
+ baseUrl: "https://api.mistral.ai/v1",
5205
5295
  reasoning: false,
5206
5296
  input: ["text", "image"],
5207
5297
  cost: {
@@ -5216,9 +5306,9 @@ export const MODELS = {
5216
5306
  "mistral-small-2603": {
5217
5307
  id: "mistral-small-2603",
5218
5308
  name: "Mistral Small 4",
5219
- api: "mistral-conversations",
5309
+ api: "openai-completions",
5220
5310
  provider: "mistral",
5221
- baseUrl: "https://api.mistral.ai",
5311
+ baseUrl: "https://api.mistral.ai/v1",
5222
5312
  reasoning: true,
5223
5313
  input: ["text", "image"],
5224
5314
  cost: {
@@ -5233,9 +5323,9 @@ export const MODELS = {
5233
5323
  "mistral-small-latest": {
5234
5324
  id: "mistral-small-latest",
5235
5325
  name: "Mistral Small (latest)",
5236
- api: "mistral-conversations",
5326
+ api: "openai-completions",
5237
5327
  provider: "mistral",
5238
- baseUrl: "https://api.mistral.ai",
5328
+ baseUrl: "https://api.mistral.ai/v1",
5239
5329
  reasoning: true,
5240
5330
  input: ["text", "image"],
5241
5331
  cost: {
@@ -5250,9 +5340,9 @@ export const MODELS = {
5250
5340
  "open-mistral-7b": {
5251
5341
  id: "open-mistral-7b",
5252
5342
  name: "Mistral 7B",
5253
- api: "mistral-conversations",
5343
+ api: "openai-completions",
5254
5344
  provider: "mistral",
5255
- baseUrl: "https://api.mistral.ai",
5345
+ baseUrl: "https://api.mistral.ai/v1",
5256
5346
  reasoning: false,
5257
5347
  input: ["text"],
5258
5348
  cost: {
@@ -5267,9 +5357,9 @@ export const MODELS = {
5267
5357
  "open-mixtral-8x22b": {
5268
5358
  id: "open-mixtral-8x22b",
5269
5359
  name: "Mixtral 8x22B",
5270
- api: "mistral-conversations",
5360
+ api: "openai-completions",
5271
5361
  provider: "mistral",
5272
- baseUrl: "https://api.mistral.ai",
5362
+ baseUrl: "https://api.mistral.ai/v1",
5273
5363
  reasoning: false,
5274
5364
  input: ["text"],
5275
5365
  cost: {
@@ -5284,9 +5374,9 @@ export const MODELS = {
5284
5374
  "open-mixtral-8x7b": {
5285
5375
  id: "open-mixtral-8x7b",
5286
5376
  name: "Mixtral 8x7B",
5287
- api: "mistral-conversations",
5377
+ api: "openai-completions",
5288
5378
  provider: "mistral",
5289
- baseUrl: "https://api.mistral.ai",
5379
+ baseUrl: "https://api.mistral.ai/v1",
5290
5380
  reasoning: false,
5291
5381
  input: ["text"],
5292
5382
  cost: {
@@ -5301,9 +5391,9 @@ export const MODELS = {
5301
5391
  "pixtral-12b": {
5302
5392
  id: "pixtral-12b",
5303
5393
  name: "Pixtral 12B",
5304
- api: "mistral-conversations",
5394
+ api: "openai-completions",
5305
5395
  provider: "mistral",
5306
- baseUrl: "https://api.mistral.ai",
5396
+ baseUrl: "https://api.mistral.ai/v1",
5307
5397
  reasoning: false,
5308
5398
  input: ["text", "image"],
5309
5399
  cost: {
@@ -5318,9 +5408,9 @@ export const MODELS = {
5318
5408
  "pixtral-large-latest": {
5319
5409
  id: "pixtral-large-latest",
5320
5410
  name: "Pixtral Large (latest)",
5321
- api: "mistral-conversations",
5411
+ api: "openai-completions",
5322
5412
  provider: "mistral",
5323
- baseUrl: "https://api.mistral.ai",
5413
+ baseUrl: "https://api.mistral.ai/v1",
5324
5414
  reasoning: false,
5325
5415
  input: ["text", "image"],
5326
5416
  cost: {
@@ -5334,23 +5424,6 @@ export const MODELS = {
5334
5424
  },
5335
5425
  },
5336
5426
  "openai": {
5337
- "codex-mini-latest": {
5338
- id: "codex-mini-latest",
5339
- name: "Codex Mini",
5340
- api: "openai-responses",
5341
- provider: "openai",
5342
- baseUrl: "https://api.openai.com/v1",
5343
- reasoning: true,
5344
- input: ["text"],
5345
- cost: {
5346
- input: 1.5,
5347
- output: 6,
5348
- cacheRead: 0.375,
5349
- cacheWrite: 0,
5350
- },
5351
- contextWindow: 200000,
5352
- maxTokens: 100000,
5353
- },
5354
5427
  "gpt-4": {
5355
5428
  id: "gpt-4",
5356
5429
  name: "GPT-4",
@@ -5895,6 +5968,23 @@ export const MODELS = {
5895
5968
  contextWindow: 1050000,
5896
5969
  maxTokens: 128000,
5897
5970
  },
5971
+ "gpt-5.5": {
5972
+ id: "gpt-5.5",
5973
+ name: "GPT-5.5",
5974
+ api: "openai-responses",
5975
+ provider: "openai",
5976
+ baseUrl: "https://api.openai.com/v1",
5977
+ reasoning: true,
5978
+ input: ["text", "image"],
5979
+ cost: {
5980
+ input: 5,
5981
+ output: 30,
5982
+ cacheRead: 0.5,
5983
+ cacheWrite: 0,
5984
+ },
5985
+ contextWindow: 1050000,
5986
+ maxTokens: 130000,
5987
+ },
5898
5988
  "o1": {
5899
5989
  id: "o1",
5900
5990
  name: "o1",
@@ -6290,6 +6380,23 @@ export const MODELS = {
6290
6380
  contextWindow: 1000000,
6291
6381
  maxTokens: 128000,
6292
6382
  },
6383
+ "claude-opus-4-7": {
6384
+ id: "claude-opus-4-7",
6385
+ name: "Claude Opus 4.7",
6386
+ api: "anthropic-messages",
6387
+ provider: "opencode",
6388
+ baseUrl: "https://opencode.ai/zen",
6389
+ reasoning: true,
6390
+ input: ["text", "image"],
6391
+ cost: {
6392
+ input: 5,
6393
+ output: 25,
6394
+ cacheRead: 0.5,
6395
+ cacheWrite: 6.25,
6396
+ },
6397
+ contextWindow: 1000000,
6398
+ maxTokens: 128000,
6399
+ },
6293
6400
  "claude-sonnet-4": {
6294
6401
  id: "claude-sonnet-4",
6295
6402
  name: "Claude Sonnet 4",
@@ -6392,6 +6499,23 @@ export const MODELS = {
6392
6499
  contextWindow: 204800,
6393
6500
  maxTokens: 131072,
6394
6501
  },
6502
+ "glm-5.1": {
6503
+ id: "glm-5.1",
6504
+ name: "GLM-5.1",
6505
+ api: "openai-completions",
6506
+ provider: "opencode",
6507
+ baseUrl: "https://opencode.ai/zen/v1",
6508
+ reasoning: true,
6509
+ input: ["text"],
6510
+ cost: {
6511
+ input: 1.4,
6512
+ output: 4.4,
6513
+ cacheRead: 0.26,
6514
+ cacheWrite: 0,
6515
+ },
6516
+ contextWindow: 204800,
6517
+ maxTokens: 131072,
6518
+ },
6395
6519
  "gpt-5": {
6396
6520
  id: "gpt-5",
6397
6521
  name: "GPT-5",
@@ -6630,39 +6754,124 @@ export const MODELS = {
6630
6754
  contextWindow: 1050000,
6631
6755
  maxTokens: 128000,
6632
6756
  },
6633
- "kimi-k2.5": {
6634
- id: "kimi-k2.5",
6635
- name: "Kimi K2.5",
6636
- api: "openai-completions",
6757
+ "gpt-5.5": {
6758
+ id: "gpt-5.5",
6759
+ name: "GPT-5.5",
6760
+ api: "openai-responses",
6637
6761
  provider: "opencode",
6638
6762
  baseUrl: "https://opencode.ai/zen/v1",
6639
6763
  reasoning: true,
6640
6764
  input: ["text", "image"],
6641
6765
  cost: {
6642
- input: 0.6,
6643
- output: 3,
6644
- cacheRead: 0.08,
6766
+ input: 5,
6767
+ output: 30,
6768
+ cacheRead: 0.5,
6645
6769
  cacheWrite: 0,
6646
6770
  },
6647
- contextWindow: 262144,
6648
- maxTokens: 65536,
6771
+ contextWindow: 1050000,
6772
+ maxTokens: 130000,
6649
6773
  },
6650
- "minimax-m2.5": {
6651
- id: "minimax-m2.5",
6652
- name: "MiniMax M2.5",
6653
- api: "openai-completions",
6774
+ "gpt-5.5-pro": {
6775
+ id: "gpt-5.5-pro",
6776
+ name: "GPT-5.5 Pro",
6777
+ api: "openai-responses",
6654
6778
  provider: "opencode",
6655
6779
  baseUrl: "https://opencode.ai/zen/v1",
6656
6780
  reasoning: true,
6657
- input: ["text"],
6781
+ input: ["text", "image"],
6658
6782
  cost: {
6659
- input: 0.3,
6660
- output: 1.2,
6661
- cacheRead: 0.06,
6783
+ input: 30,
6784
+ output: 180,
6785
+ cacheRead: 30,
6662
6786
  cacheWrite: 0,
6663
6787
  },
6664
- contextWindow: 204800,
6665
- maxTokens: 131072,
6788
+ contextWindow: 1050000,
6789
+ maxTokens: 128000,
6790
+ },
6791
+ "hy3-preview-free": {
6792
+ id: "hy3-preview-free",
6793
+ name: "Hy3 preview Free",
6794
+ api: "openai-completions",
6795
+ provider: "opencode",
6796
+ baseUrl: "https://opencode.ai/zen/v1",
6797
+ reasoning: true,
6798
+ input: ["text"],
6799
+ cost: {
6800
+ input: 0,
6801
+ output: 0,
6802
+ cacheRead: 0,
6803
+ cacheWrite: 0,
6804
+ },
6805
+ contextWindow: 256000,
6806
+ maxTokens: 64000,
6807
+ },
6808
+ "kimi-k2.5": {
6809
+ id: "kimi-k2.5",
6810
+ name: "Kimi K2.5",
6811
+ api: "openai-completions",
6812
+ provider: "opencode",
6813
+ baseUrl: "https://opencode.ai/zen/v1",
6814
+ reasoning: true,
6815
+ input: ["text", "image"],
6816
+ cost: {
6817
+ input: 0.6,
6818
+ output: 3,
6819
+ cacheRead: 0.08,
6820
+ cacheWrite: 0,
6821
+ },
6822
+ contextWindow: 262144,
6823
+ maxTokens: 65536,
6824
+ },
6825
+ "kimi-k2.6": {
6826
+ id: "kimi-k2.6",
6827
+ name: "Kimi K2.6",
6828
+ api: "openai-completions",
6829
+ provider: "opencode",
6830
+ baseUrl: "https://opencode.ai/zen/v1",
6831
+ reasoning: true,
6832
+ input: ["text", "image"],
6833
+ cost: {
6834
+ input: 0.95,
6835
+ output: 4,
6836
+ cacheRead: 0.16,
6837
+ cacheWrite: 0,
6838
+ },
6839
+ contextWindow: 262144,
6840
+ maxTokens: 65536,
6841
+ },
6842
+ "ling-2.6-flash-free": {
6843
+ id: "ling-2.6-flash-free",
6844
+ name: "Ling 2.6 Flash Free",
6845
+ api: "openai-completions",
6846
+ provider: "opencode",
6847
+ baseUrl: "https://opencode.ai/zen/v1",
6848
+ reasoning: false,
6849
+ input: ["text"],
6850
+ cost: {
6851
+ input: 0,
6852
+ output: 0,
6853
+ cacheRead: 0,
6854
+ cacheWrite: 0,
6855
+ },
6856
+ contextWindow: 262100,
6857
+ maxTokens: 32800,
6858
+ },
6859
+ "minimax-m2.5": {
6860
+ id: "minimax-m2.5",
6861
+ name: "MiniMax M2.5",
6862
+ api: "openai-completions",
6863
+ provider: "opencode",
6864
+ baseUrl: "https://opencode.ai/zen/v1",
6865
+ reasoning: true,
6866
+ input: ["text"],
6867
+ cost: {
6868
+ input: 0.3,
6869
+ output: 1.2,
6870
+ cacheRead: 0.06,
6871
+ cacheWrite: 0,
6872
+ },
6873
+ contextWindow: 204800,
6874
+ maxTokens: 131072,
6666
6875
  },
6667
6876
  "minimax-m2.5-free": {
6668
6877
  id: "minimax-m2.5-free",
@@ -6681,26 +6890,26 @@ export const MODELS = {
6681
6890
  contextWindow: 204800,
6682
6891
  maxTokens: 131072,
6683
6892
  },
6684
- "nemotron-3-super-free": {
6685
- id: "nemotron-3-super-free",
6686
- name: "Nemotron 3 Super Free",
6893
+ "minimax-m2.7": {
6894
+ id: "minimax-m2.7",
6895
+ name: "MiniMax M2.7",
6687
6896
  api: "openai-completions",
6688
6897
  provider: "opencode",
6689
6898
  baseUrl: "https://opencode.ai/zen/v1",
6690
6899
  reasoning: true,
6691
6900
  input: ["text"],
6692
6901
  cost: {
6693
- input: 0,
6694
- output: 0,
6695
- cacheRead: 0,
6902
+ input: 0.3,
6903
+ output: 1.2,
6904
+ cacheRead: 0.06,
6696
6905
  cacheWrite: 0,
6697
6906
  },
6698
6907
  contextWindow: 204800,
6699
- maxTokens: 128000,
6908
+ maxTokens: 131072,
6700
6909
  },
6701
- "qwen3.6-plus-free": {
6702
- id: "qwen3.6-plus-free",
6703
- name: "Qwen3.6 Plus Free",
6910
+ "nemotron-3-super-free": {
6911
+ id: "nemotron-3-super-free",
6912
+ name: "Nemotron 3 Super Free",
6704
6913
  api: "openai-completions",
6705
6914
  provider: "opencode",
6706
6915
  baseUrl: "https://opencode.ai/zen/v1",
@@ -6712,11 +6921,79 @@ export const MODELS = {
6712
6921
  cacheRead: 0,
6713
6922
  cacheWrite: 0,
6714
6923
  },
6715
- contextWindow: 1048576,
6716
- maxTokens: 64000,
6924
+ contextWindow: 204800,
6925
+ maxTokens: 128000,
6926
+ },
6927
+ "qwen3.5-plus": {
6928
+ id: "qwen3.5-plus",
6929
+ name: "Qwen3.5 Plus",
6930
+ api: "anthropic-messages",
6931
+ provider: "opencode",
6932
+ baseUrl: "https://opencode.ai/zen",
6933
+ reasoning: true,
6934
+ input: ["text", "image"],
6935
+ cost: {
6936
+ input: 0.2,
6937
+ output: 1.2,
6938
+ cacheRead: 0.02,
6939
+ cacheWrite: 0.25,
6940
+ },
6941
+ contextWindow: 262144,
6942
+ maxTokens: 65536,
6943
+ },
6944
+ "qwen3.6-plus": {
6945
+ id: "qwen3.6-plus",
6946
+ name: "Qwen3.6 Plus",
6947
+ api: "anthropic-messages",
6948
+ provider: "opencode",
6949
+ baseUrl: "https://opencode.ai/zen",
6950
+ reasoning: true,
6951
+ input: ["text", "image"],
6952
+ cost: {
6953
+ input: 0.5,
6954
+ output: 3,
6955
+ cacheRead: 0.05,
6956
+ cacheWrite: 0.625,
6957
+ },
6958
+ contextWindow: 262144,
6959
+ maxTokens: 65536,
6717
6960
  },
6718
6961
  },
6719
6962
  "opencode-go": {
6963
+ "deepseek-v4-flash": {
6964
+ id: "deepseek-v4-flash",
6965
+ name: "DeepSeek V4 Flash",
6966
+ api: "openai-completions",
6967
+ provider: "opencode-go",
6968
+ baseUrl: "https://opencode.ai/zen/go/v1",
6969
+ reasoning: true,
6970
+ input: ["text"],
6971
+ cost: {
6972
+ input: 0.14,
6973
+ output: 0.28,
6974
+ cacheRead: 0.028,
6975
+ cacheWrite: 0,
6976
+ },
6977
+ contextWindow: 1000000,
6978
+ maxTokens: 384000,
6979
+ },
6980
+ "deepseek-v4-pro": {
6981
+ id: "deepseek-v4-pro",
6982
+ name: "DeepSeek V4 Pro",
6983
+ api: "openai-completions",
6984
+ provider: "opencode-go",
6985
+ baseUrl: "https://opencode.ai/zen/go/v1",
6986
+ reasoning: true,
6987
+ input: ["text"],
6988
+ cost: {
6989
+ input: 1.74,
6990
+ output: 3.48,
6991
+ cacheRead: 0.145,
6992
+ cacheWrite: 0,
6993
+ },
6994
+ contextWindow: 1000000,
6995
+ maxTokens: 384000,
6996
+ },
6720
6997
  "glm-5": {
6721
6998
  id: "glm-5",
6722
6999
  name: "GLM-5",
@@ -6734,6 +7011,23 @@ export const MODELS = {
6734
7011
  contextWindow: 204800,
6735
7012
  maxTokens: 131072,
6736
7013
  },
7014
+ "glm-5.1": {
7015
+ id: "glm-5.1",
7016
+ name: "GLM-5.1",
7017
+ api: "openai-completions",
7018
+ provider: "opencode-go",
7019
+ baseUrl: "https://opencode.ai/zen/go/v1",
7020
+ reasoning: true,
7021
+ input: ["text"],
7022
+ cost: {
7023
+ input: 1.4,
7024
+ output: 4.4,
7025
+ cacheRead: 0.26,
7026
+ cacheWrite: 0,
7027
+ },
7028
+ contextWindow: 204800,
7029
+ maxTokens: 131072,
7030
+ },
6737
7031
  "kimi-k2.5": {
6738
7032
  id: "kimi-k2.5",
6739
7033
  name: "Kimi K2.5",
@@ -6751,6 +7045,23 @@ export const MODELS = {
6751
7045
  contextWindow: 262144,
6752
7046
  maxTokens: 65536,
6753
7047
  },
7048
+ "kimi-k2.6": {
7049
+ id: "kimi-k2.6",
7050
+ name: "Kimi K2.6 (3x limits)",
7051
+ api: "openai-completions",
7052
+ provider: "opencode-go",
7053
+ baseUrl: "https://opencode.ai/zen/go/v1",
7054
+ reasoning: true,
7055
+ input: ["text", "image"],
7056
+ cost: {
7057
+ input: 0.32,
7058
+ output: 1.34,
7059
+ cacheRead: 0.054,
7060
+ cacheWrite: 0,
7061
+ },
7062
+ contextWindow: 262144,
7063
+ maxTokens: 65536,
7064
+ },
6754
7065
  "mimo-v2-omni": {
6755
7066
  id: "mimo-v2-omni",
6756
7067
  name: "MiMo V2 Omni",
@@ -6766,7 +7077,7 @@ export const MODELS = {
6766
7077
  cacheWrite: 0,
6767
7078
  },
6768
7079
  contextWindow: 262144,
6769
- maxTokens: 64000,
7080
+ maxTokens: 128000,
6770
7081
  },
6771
7082
  "mimo-v2-pro": {
6772
7083
  id: "mimo-v2-pro",
@@ -6783,7 +7094,41 @@ export const MODELS = {
6783
7094
  cacheWrite: 0,
6784
7095
  },
6785
7096
  contextWindow: 1048576,
6786
- maxTokens: 64000,
7097
+ maxTokens: 128000,
7098
+ },
7099
+ "mimo-v2.5": {
7100
+ id: "mimo-v2.5",
7101
+ name: "MiMo V2.5",
7102
+ api: "openai-completions",
7103
+ provider: "opencode-go",
7104
+ baseUrl: "https://opencode.ai/zen/go/v1",
7105
+ reasoning: true,
7106
+ input: ["text", "image"],
7107
+ cost: {
7108
+ input: 0.4,
7109
+ output: 2,
7110
+ cacheRead: 0.08,
7111
+ cacheWrite: 0,
7112
+ },
7113
+ contextWindow: 262144,
7114
+ maxTokens: 128000,
7115
+ },
7116
+ "mimo-v2.5-pro": {
7117
+ id: "mimo-v2.5-pro",
7118
+ name: "MiMo V2.5 Pro",
7119
+ api: "openai-completions",
7120
+ provider: "opencode-go",
7121
+ baseUrl: "https://opencode.ai/zen/go/v1",
7122
+ reasoning: true,
7123
+ input: ["text"],
7124
+ cost: {
7125
+ input: 1,
7126
+ output: 3,
7127
+ cacheRead: 0.2,
7128
+ cacheWrite: 0,
7129
+ },
7130
+ contextWindow: 1048576,
7131
+ maxTokens: 128000,
6787
7132
  },
6788
7133
  "minimax-m2.5": {
6789
7134
  id: "minimax-m2.5",
@@ -6800,7 +7145,7 @@ export const MODELS = {
6800
7145
  cacheWrite: 0,
6801
7146
  },
6802
7147
  contextWindow: 204800,
6803
- maxTokens: 131072,
7148
+ maxTokens: 65536,
6804
7149
  },
6805
7150
  "minimax-m2.7": {
6806
7151
  id: "minimax-m2.7",
@@ -6819,6 +7164,40 @@ export const MODELS = {
6819
7164
  contextWindow: 204800,
6820
7165
  maxTokens: 131072,
6821
7166
  },
7167
+ "qwen3.5-plus": {
7168
+ id: "qwen3.5-plus",
7169
+ name: "Qwen3.5 Plus",
7170
+ api: "anthropic-messages",
7171
+ provider: "opencode-go",
7172
+ baseUrl: "https://opencode.ai/zen/go",
7173
+ reasoning: true,
7174
+ input: ["text", "image"],
7175
+ cost: {
7176
+ input: 0.2,
7177
+ output: 1.2,
7178
+ cacheRead: 0.02,
7179
+ cacheWrite: 0.25,
7180
+ },
7181
+ contextWindow: 262144,
7182
+ maxTokens: 65536,
7183
+ },
7184
+ "qwen3.6-plus": {
7185
+ id: "qwen3.6-plus",
7186
+ name: "Qwen3.6 Plus",
7187
+ api: "anthropic-messages",
7188
+ provider: "opencode-go",
7189
+ baseUrl: "https://opencode.ai/zen/go",
7190
+ reasoning: true,
7191
+ input: ["text", "image"],
7192
+ cost: {
7193
+ input: 0.5,
7194
+ output: 3,
7195
+ cacheRead: 0.05,
7196
+ cacheWrite: 0.625,
7197
+ },
7198
+ contextWindow: 262144,
7199
+ maxTokens: 65536,
7200
+ },
6822
7201
  },
6823
7202
  "openrouter": {
6824
7203
  "ai21/jamba-large-1.7": {
@@ -6870,7 +7249,7 @@ export const MODELS = {
6870
7249
  cacheWrite: 0,
6871
7250
  },
6872
7251
  contextWindow: 65536,
6873
- maxTokens: 4096,
7252
+ maxTokens: 16384,
6874
7253
  },
6875
7254
  "amazon/nova-2-lite-v1": {
6876
7255
  id: "amazon/nova-2-lite-v1",
@@ -7006,7 +7385,7 @@ export const MODELS = {
7006
7385
  cacheWrite: 3.75,
7007
7386
  },
7008
7387
  contextWindow: 200000,
7009
- maxTokens: 64000,
7388
+ maxTokens: 128000,
7010
7389
  },
7011
7390
  "anthropic/claude-3.7-sonnet:thinking": {
7012
7391
  id: "anthropic/claude-3.7-sonnet:thinking",
@@ -7110,6 +7489,40 @@ export const MODELS = {
7110
7489
  contextWindow: 1000000,
7111
7490
  maxTokens: 128000,
7112
7491
  },
7492
+ "anthropic/claude-opus-4.6-fast": {
7493
+ id: "anthropic/claude-opus-4.6-fast",
7494
+ name: "Anthropic: Claude Opus 4.6 (Fast)",
7495
+ api: "openai-completions",
7496
+ provider: "openrouter",
7497
+ baseUrl: "https://openrouter.ai/api/v1",
7498
+ reasoning: true,
7499
+ input: ["text", "image"],
7500
+ cost: {
7501
+ input: 30,
7502
+ output: 150,
7503
+ cacheRead: 3,
7504
+ cacheWrite: 37.5,
7505
+ },
7506
+ contextWindow: 1000000,
7507
+ maxTokens: 128000,
7508
+ },
7509
+ "anthropic/claude-opus-4.7": {
7510
+ id: "anthropic/claude-opus-4.7",
7511
+ name: "Anthropic: Claude Opus 4.7",
7512
+ api: "openai-completions",
7513
+ provider: "openrouter",
7514
+ baseUrl: "https://openrouter.ai/api/v1",
7515
+ reasoning: true,
7516
+ input: ["text", "image"],
7517
+ cost: {
7518
+ input: 5,
7519
+ output: 25,
7520
+ cacheRead: 0.5,
7521
+ cacheWrite: 6.25,
7522
+ },
7523
+ contextWindow: 1000000,
7524
+ maxTokens: 128000,
7525
+ },
7113
7526
  "anthropic/claude-sonnet-4": {
7114
7527
  id: "anthropic/claude-sonnet-4",
7115
7528
  name: "Anthropic: Claude Sonnet 4",
@@ -7124,7 +7537,7 @@ export const MODELS = {
7124
7537
  cacheRead: 0.3,
7125
7538
  cacheWrite: 3.75,
7126
7539
  },
7127
- contextWindow: 200000,
7540
+ contextWindow: 1000000,
7128
7541
  maxTokens: 64000,
7129
7542
  },
7130
7543
  "anthropic/claude-sonnet-4.5": {
@@ -7161,17 +7574,17 @@ export const MODELS = {
7161
7574
  contextWindow: 1000000,
7162
7575
  maxTokens: 128000,
7163
7576
  },
7164
- "arcee-ai/trinity-large-preview:free": {
7165
- id: "arcee-ai/trinity-large-preview:free",
7166
- name: "Arcee AI: Trinity Large Preview (free)",
7577
+ "arcee-ai/trinity-large-preview": {
7578
+ id: "arcee-ai/trinity-large-preview",
7579
+ name: "Arcee AI: Trinity Large Preview",
7167
7580
  api: "openai-completions",
7168
7581
  provider: "openrouter",
7169
7582
  baseUrl: "https://openrouter.ai/api/v1",
7170
7583
  reasoning: false,
7171
7584
  input: ["text"],
7172
7585
  cost: {
7173
- input: 0,
7174
- output: 0,
7586
+ input: 0.15,
7587
+ output: 0.44999999999999996,
7175
7588
  cacheRead: 0,
7176
7589
  cacheWrite: 0,
7177
7590
  },
@@ -7189,45 +7602,28 @@ export const MODELS = {
7189
7602
  cost: {
7190
7603
  input: 0.22,
7191
7604
  output: 0.85,
7192
- cacheRead: 0,
7605
+ cacheRead: 0.06,
7193
7606
  cacheWrite: 0,
7194
7607
  },
7195
7608
  contextWindow: 262144,
7196
7609
  maxTokens: 262144,
7197
7610
  },
7198
- "arcee-ai/trinity-mini": {
7199
- id: "arcee-ai/trinity-mini",
7200
- name: "Arcee AI: Trinity Mini",
7201
- api: "openai-completions",
7202
- provider: "openrouter",
7203
- baseUrl: "https://openrouter.ai/api/v1",
7204
- reasoning: true,
7205
- input: ["text"],
7206
- cost: {
7207
- input: 0.045,
7208
- output: 0.15,
7209
- cacheRead: 0,
7210
- cacheWrite: 0,
7211
- },
7212
- contextWindow: 131072,
7213
- maxTokens: 131072,
7214
- },
7215
- "arcee-ai/trinity-mini:free": {
7216
- id: "arcee-ai/trinity-mini:free",
7217
- name: "Arcee AI: Trinity Mini (free)",
7611
+ "arcee-ai/trinity-mini": {
7612
+ id: "arcee-ai/trinity-mini",
7613
+ name: "Arcee AI: Trinity Mini",
7218
7614
  api: "openai-completions",
7219
7615
  provider: "openrouter",
7220
7616
  baseUrl: "https://openrouter.ai/api/v1",
7221
7617
  reasoning: true,
7222
7618
  input: ["text"],
7223
7619
  cost: {
7224
- input: 0,
7225
- output: 0,
7620
+ input: 0.045,
7621
+ output: 0.15,
7226
7622
  cacheRead: 0,
7227
7623
  cacheWrite: 0,
7228
7624
  },
7229
7625
  contextWindow: 131072,
7230
- maxTokens: 4096,
7626
+ maxTokens: 131072,
7231
7627
  },
7232
7628
  "arcee-ai/virtuoso-large": {
7233
7629
  id: "arcee-ai/virtuoso-large",
@@ -7414,7 +7810,7 @@ export const MODELS = {
7414
7810
  cacheWrite: 0,
7415
7811
  },
7416
7812
  contextWindow: 163840,
7417
- maxTokens: 163840,
7813
+ maxTokens: 16384,
7418
7814
  },
7419
7815
  "deepseek/deepseek-chat-v3-0324": {
7420
7816
  id: "deepseek/deepseek-chat-v3-0324",
@@ -7422,7 +7818,7 @@ export const MODELS = {
7422
7818
  api: "openai-completions",
7423
7819
  provider: "openrouter",
7424
7820
  baseUrl: "https://openrouter.ai/api/v1",
7425
- reasoning: true,
7821
+ reasoning: false,
7426
7822
  input: ["text"],
7427
7823
  cost: {
7428
7824
  input: 0.19999999999999998,
@@ -7431,7 +7827,7 @@ export const MODELS = {
7431
7827
  cacheWrite: 0,
7432
7828
  },
7433
7829
  contextWindow: 163840,
7434
- maxTokens: 4096,
7830
+ maxTokens: 16384,
7435
7831
  },
7436
7832
  "deepseek/deepseek-chat-v3.1": {
7437
7833
  id: "deepseek/deepseek-chat-v3.1",
@@ -7476,13 +7872,13 @@ export const MODELS = {
7476
7872
  reasoning: true,
7477
7873
  input: ["text"],
7478
7874
  cost: {
7479
- input: 0.44999999999999996,
7875
+ input: 0.5,
7480
7876
  output: 2.1500000000000004,
7481
- cacheRead: 0.22499999999999998,
7877
+ cacheRead: 0.35,
7482
7878
  cacheWrite: 0,
7483
7879
  },
7484
7880
  contextWindow: 163840,
7485
- maxTokens: 65536,
7881
+ maxTokens: 32768,
7486
7882
  },
7487
7883
  "deepseek/deepseek-v3.1-terminus": {
7488
7884
  id: "deepseek/deepseek-v3.1-terminus",
@@ -7495,11 +7891,11 @@ export const MODELS = {
7495
7891
  cost: {
7496
7892
  input: 0.21,
7497
7893
  output: 0.7899999999999999,
7498
- cacheRead: 0.1300000002,
7894
+ cacheRead: 0.13,
7499
7895
  cacheWrite: 0,
7500
7896
  },
7501
7897
  contextWindow: 163840,
7502
- maxTokens: 4096,
7898
+ maxTokens: 32768,
7503
7899
  },
7504
7900
  "deepseek/deepseek-v3.2": {
7505
7901
  id: "deepseek/deepseek-v3.2",
@@ -7510,13 +7906,13 @@ export const MODELS = {
7510
7906
  reasoning: true,
7511
7907
  input: ["text"],
7512
7908
  cost: {
7513
- input: 0.26,
7514
- output: 0.38,
7515
- cacheRead: 0.13,
7909
+ input: 0.252,
7910
+ output: 0.378,
7911
+ cacheRead: 0.0252,
7516
7912
  cacheWrite: 0,
7517
7913
  },
7518
- contextWindow: 163840,
7519
- maxTokens: 4096,
7914
+ contextWindow: 131072,
7915
+ maxTokens: 65536,
7520
7916
  },
7521
7917
  "deepseek/deepseek-v3.2-exp": {
7522
7918
  id: "deepseek/deepseek-v3.2-exp",
@@ -7535,6 +7931,40 @@ export const MODELS = {
7535
7931
  contextWindow: 163840,
7536
7932
  maxTokens: 65536,
7537
7933
  },
7934
+ "deepseek/deepseek-v4-flash": {
7935
+ id: "deepseek/deepseek-v4-flash",
7936
+ name: "DeepSeek: DeepSeek V4 Flash",
7937
+ api: "openai-completions",
7938
+ provider: "openrouter",
7939
+ baseUrl: "https://openrouter.ai/api/v1",
7940
+ reasoning: true,
7941
+ input: ["text"],
7942
+ cost: {
7943
+ input: 0.14,
7944
+ output: 0.28,
7945
+ cacheRead: 0.028,
7946
+ cacheWrite: 0,
7947
+ },
7948
+ contextWindow: 1048576,
7949
+ maxTokens: 384000,
7950
+ },
7951
+ "deepseek/deepseek-v4-pro": {
7952
+ id: "deepseek/deepseek-v4-pro",
7953
+ name: "DeepSeek: DeepSeek V4 Pro",
7954
+ api: "openai-completions",
7955
+ provider: "openrouter",
7956
+ baseUrl: "https://openrouter.ai/api/v1",
7957
+ reasoning: true,
7958
+ input: ["text"],
7959
+ cost: {
7960
+ input: 0.435,
7961
+ output: 0.87,
7962
+ cacheRead: 0.03625,
7963
+ cacheWrite: 0,
7964
+ },
7965
+ contextWindow: 1048576,
7966
+ maxTokens: 384000,
7967
+ },
7538
7968
  "essentialai/rnj-1-instruct": {
7539
7969
  id: "essentialai/rnj-1-instruct",
7540
7970
  name: "EssentialAI: Rnj 1 Instruct",
@@ -7566,7 +7996,7 @@ export const MODELS = {
7566
7996
  cacheRead: 0.024999999999999998,
7567
7997
  cacheWrite: 0.08333333333333334,
7568
7998
  },
7569
- contextWindow: 1048576,
7999
+ contextWindow: 1000000,
7570
8000
  maxTokens: 8192,
7571
8001
  },
7572
8002
  "google/gemini-2.0-flash-lite-001": {
@@ -7756,6 +8186,40 @@ export const MODELS = {
7756
8186
  contextWindow: 1048576,
7757
8187
  maxTokens: 65536,
7758
8188
  },
8189
+ "google/gemma-3-12b-it": {
8190
+ id: "google/gemma-3-12b-it",
8191
+ name: "Google: Gemma 3 12B",
8192
+ api: "openai-completions",
8193
+ provider: "openrouter",
8194
+ baseUrl: "https://openrouter.ai/api/v1",
8195
+ reasoning: false,
8196
+ input: ["text", "image"],
8197
+ cost: {
8198
+ input: 0.04,
8199
+ output: 0.13,
8200
+ cacheRead: 0,
8201
+ cacheWrite: 0,
8202
+ },
8203
+ contextWindow: 131072,
8204
+ maxTokens: 16384,
8205
+ },
8206
+ "google/gemma-3-27b-it": {
8207
+ id: "google/gemma-3-27b-it",
8208
+ name: "Google: Gemma 3 27B",
8209
+ api: "openai-completions",
8210
+ provider: "openrouter",
8211
+ baseUrl: "https://openrouter.ai/api/v1",
8212
+ reasoning: false,
8213
+ input: ["text", "image"],
8214
+ cost: {
8215
+ input: 0.08,
8216
+ output: 0.16,
8217
+ cacheRead: 0,
8218
+ cacheWrite: 0,
8219
+ },
8220
+ contextWindow: 131072,
8221
+ maxTokens: 16384,
8222
+ },
7759
8223
  "google/gemma-4-26b-a4b-it": {
7760
8224
  id: "google/gemma-4-26b-a4b-it",
7761
8225
  name: "Google: Gemma 4 26B A4B ",
@@ -7765,13 +8229,30 @@ export const MODELS = {
7765
8229
  reasoning: true,
7766
8230
  input: ["text", "image"],
7767
8231
  cost: {
7768
- input: 0.13,
7769
- output: 0.39999999999999997,
8232
+ input: 0.06,
8233
+ output: 0.33,
7770
8234
  cacheRead: 0,
7771
8235
  cacheWrite: 0,
7772
8236
  },
7773
8237
  contextWindow: 262144,
7774
- maxTokens: 262144,
8238
+ maxTokens: 4096,
8239
+ },
8240
+ "google/gemma-4-26b-a4b-it:free": {
8241
+ id: "google/gemma-4-26b-a4b-it:free",
8242
+ name: "Google: Gemma 4 26B A4B (free)",
8243
+ api: "openai-completions",
8244
+ provider: "openrouter",
8245
+ baseUrl: "https://openrouter.ai/api/v1",
8246
+ reasoning: true,
8247
+ input: ["text", "image"],
8248
+ cost: {
8249
+ input: 0,
8250
+ output: 0,
8251
+ cacheRead: 0,
8252
+ cacheWrite: 0,
8253
+ },
8254
+ contextWindow: 262144,
8255
+ maxTokens: 32768,
7775
8256
  },
7776
8257
  "google/gemma-4-31b-it": {
7777
8258
  id: "google/gemma-4-31b-it",
@@ -7782,30 +8263,30 @@ export const MODELS = {
7782
8263
  reasoning: true,
7783
8264
  input: ["text", "image"],
7784
8265
  cost: {
7785
- input: 0.14,
7786
- output: 0.39999999999999997,
8266
+ input: 0.13,
8267
+ output: 0.38,
7787
8268
  cacheRead: 0,
7788
8269
  cacheWrite: 0,
7789
8270
  },
7790
8271
  contextWindow: 262144,
7791
- maxTokens: 131072,
8272
+ maxTokens: 16384,
7792
8273
  },
7793
- "inception/mercury": {
7794
- id: "inception/mercury",
7795
- name: "Inception: Mercury",
8274
+ "google/gemma-4-31b-it:free": {
8275
+ id: "google/gemma-4-31b-it:free",
8276
+ name: "Google: Gemma 4 31B (free)",
7796
8277
  api: "openai-completions",
7797
8278
  provider: "openrouter",
7798
8279
  baseUrl: "https://openrouter.ai/api/v1",
7799
- reasoning: false,
7800
- input: ["text"],
8280
+ reasoning: true,
8281
+ input: ["text", "image"],
7801
8282
  cost: {
7802
- input: 0.25,
7803
- output: 0.75,
7804
- cacheRead: 0.024999999999999998,
8283
+ input: 0,
8284
+ output: 0,
8285
+ cacheRead: 0,
7805
8286
  cacheWrite: 0,
7806
8287
  },
7807
- contextWindow: 128000,
7808
- maxTokens: 32000,
8288
+ contextWindow: 262144,
8289
+ maxTokens: 32768,
7809
8290
  },
7810
8291
  "inception/mercury-2": {
7811
8292
  id: "inception/mercury-2",
@@ -7824,56 +8305,56 @@ export const MODELS = {
7824
8305
  contextWindow: 128000,
7825
8306
  maxTokens: 50000,
7826
8307
  },
7827
- "inception/mercury-coder": {
7828
- id: "inception/mercury-coder",
7829
- name: "Inception: Mercury Coder",
8308
+ "inclusionai/ling-2.6-1t:free": {
8309
+ id: "inclusionai/ling-2.6-1t:free",
8310
+ name: "inclusionAI: Ling-2.6-1T (free)",
7830
8311
  api: "openai-completions",
7831
8312
  provider: "openrouter",
7832
8313
  baseUrl: "https://openrouter.ai/api/v1",
7833
8314
  reasoning: false,
7834
8315
  input: ["text"],
7835
8316
  cost: {
7836
- input: 0.25,
7837
- output: 0.75,
7838
- cacheRead: 0.024999999999999998,
8317
+ input: 0,
8318
+ output: 0,
8319
+ cacheRead: 0,
7839
8320
  cacheWrite: 0,
7840
8321
  },
7841
- contextWindow: 128000,
7842
- maxTokens: 32000,
8322
+ contextWindow: 262144,
8323
+ maxTokens: 32768,
7843
8324
  },
7844
- "kwaipilot/kat-coder-pro-v2": {
7845
- id: "kwaipilot/kat-coder-pro-v2",
7846
- name: "Kwaipilot: KAT-Coder-Pro V2",
8325
+ "inclusionai/ling-2.6-flash:free": {
8326
+ id: "inclusionai/ling-2.6-flash:free",
8327
+ name: "inclusionAI: Ling-2.6-flash (free)",
7847
8328
  api: "openai-completions",
7848
8329
  provider: "openrouter",
7849
8330
  baseUrl: "https://openrouter.ai/api/v1",
7850
8331
  reasoning: false,
7851
8332
  input: ["text"],
7852
8333
  cost: {
7853
- input: 0.3,
7854
- output: 1.2,
7855
- cacheRead: 0.06,
8334
+ input: 0,
8335
+ output: 0,
8336
+ cacheRead: 0,
7856
8337
  cacheWrite: 0,
7857
8338
  },
7858
- contextWindow: 256000,
7859
- maxTokens: 80000,
8339
+ contextWindow: 262144,
8340
+ maxTokens: 32768,
7860
8341
  },
7861
- "meituan/longcat-flash-chat": {
7862
- id: "meituan/longcat-flash-chat",
7863
- name: "Meituan: LongCat Flash Chat",
8342
+ "kwaipilot/kat-coder-pro-v2": {
8343
+ id: "kwaipilot/kat-coder-pro-v2",
8344
+ name: "Kwaipilot: KAT-Coder-Pro V2",
7864
8345
  api: "openai-completions",
7865
8346
  provider: "openrouter",
7866
8347
  baseUrl: "https://openrouter.ai/api/v1",
7867
8348
  reasoning: false,
7868
8349
  input: ["text"],
7869
8350
  cost: {
7870
- input: 0.19999999999999998,
7871
- output: 0.7999999999999999,
7872
- cacheRead: 0.19999999999999998,
8351
+ input: 0.3,
8352
+ output: 1.2,
8353
+ cacheRead: 0.06,
7873
8354
  cacheWrite: 0,
7874
8355
  },
7875
- contextWindow: 131072,
7876
- maxTokens: 131072,
8356
+ contextWindow: 256000,
8357
+ maxTokens: 80000,
7877
8358
  },
7878
8359
  "meta-llama/llama-3-8b-instruct": {
7879
8360
  id: "meta-llama/llama-3-8b-instruct",
@@ -7907,7 +8388,7 @@ export const MODELS = {
7907
8388
  cacheWrite: 0,
7908
8389
  },
7909
8390
  contextWindow: 131072,
7910
- maxTokens: 4096,
8391
+ maxTokens: 16384,
7911
8392
  },
7912
8393
  "meta-llama/llama-3.1-8b-instruct": {
7913
8394
  id: "meta-llama/llama-3.1-8b-instruct",
@@ -7960,23 +8441,6 @@ export const MODELS = {
7960
8441
  contextWindow: 65536,
7961
8442
  maxTokens: 4096,
7962
8443
  },
7963
- "meta-llama/llama-4-maverick": {
7964
- id: "meta-llama/llama-4-maverick",
7965
- name: "Meta: Llama 4 Maverick",
7966
- api: "openai-completions",
7967
- provider: "openrouter",
7968
- baseUrl: "https://openrouter.ai/api/v1",
7969
- reasoning: false,
7970
- input: ["text", "image"],
7971
- cost: {
7972
- input: 0.15,
7973
- output: 0.6,
7974
- cacheRead: 0,
7975
- cacheWrite: 0,
7976
- },
7977
- contextWindow: 1048576,
7978
- maxTokens: 16384,
7979
- },
7980
8444
  "meta-llama/llama-4-scout": {
7981
8445
  id: "meta-llama/llama-4-scout",
7982
8446
  name: "Meta: Llama 4 Scout",
@@ -8037,13 +8501,13 @@ export const MODELS = {
8037
8501
  reasoning: true,
8038
8502
  input: ["text"],
8039
8503
  cost: {
8040
- input: 0.27,
8504
+ input: 0.29,
8041
8505
  output: 0.95,
8042
- cacheRead: 0.0290000007,
8506
+ cacheRead: 0.03,
8043
8507
  cacheWrite: 0,
8044
8508
  },
8045
8509
  contextWindow: 196608,
8046
- maxTokens: 4096,
8510
+ maxTokens: 196608,
8047
8511
  },
8048
8512
  "minimax/minimax-m2.5": {
8049
8513
  id: "minimax/minimax-m2.5",
@@ -8054,13 +8518,13 @@ export const MODELS = {
8054
8518
  reasoning: true,
8055
8519
  input: ["text"],
8056
8520
  cost: {
8057
- input: 0.118,
8058
- output: 0.9900000000000001,
8059
- cacheRead: 0.059,
8521
+ input: 0.15,
8522
+ output: 1.15,
8523
+ cacheRead: 0.03,
8060
8524
  cacheWrite: 0,
8061
8525
  },
8062
8526
  contextWindow: 196608,
8063
- maxTokens: 65536,
8527
+ maxTokens: 131072,
8064
8528
  },
8065
8529
  "minimax/minimax-m2.5:free": {
8066
8530
  id: "minimax/minimax-m2.5:free",
@@ -8077,7 +8541,7 @@ export const MODELS = {
8077
8541
  cacheWrite: 0,
8078
8542
  },
8079
8543
  contextWindow: 196608,
8080
- maxTokens: 196608,
8544
+ maxTokens: 8192,
8081
8545
  },
8082
8546
  "minimax/minimax-m2.7": {
8083
8547
  id: "minimax/minimax-m2.7",
@@ -8090,11 +8554,11 @@ export const MODELS = {
8090
8554
  cost: {
8091
8555
  input: 0.3,
8092
8556
  output: 1.2,
8093
- cacheRead: 0.06,
8557
+ cacheRead: 0.059,
8094
8558
  cacheWrite: 0,
8095
8559
  },
8096
- contextWindow: 204800,
8097
- maxTokens: 131072,
8560
+ contextWindow: 196608,
8561
+ maxTokens: 4096,
8098
8562
  },
8099
8563
  "mistralai/codestral-2508": {
8100
8564
  id: "mistralai/codestral-2508",
@@ -8326,13 +8790,13 @@ export const MODELS = {
8326
8790
  reasoning: false,
8327
8791
  input: ["text"],
8328
8792
  cost: {
8329
- input: 0.02,
8330
- output: 0.04,
8793
+ input: 0.01,
8794
+ output: 0.03,
8331
8795
  cacheRead: 0,
8332
8796
  cacheWrite: 0,
8333
8797
  },
8334
8798
  contextWindow: 131072,
8335
- maxTokens: 16384,
8799
+ maxTokens: 4096,
8336
8800
  },
8337
8801
  "mistralai/mistral-saba": {
8338
8802
  id: "mistralai/mistral-saba",
@@ -8383,7 +8847,7 @@ export const MODELS = {
8383
8847
  cacheWrite: 0,
8384
8848
  },
8385
8849
  contextWindow: 128000,
8386
- maxTokens: 4096,
8850
+ maxTokens: 16384,
8387
8851
  },
8388
8852
  "mistralai/mistral-small-creative": {
8389
8853
  id: "mistralai/mistral-small-creative",
@@ -8485,7 +8949,7 @@ export const MODELS = {
8485
8949
  cacheWrite: 0,
8486
8950
  },
8487
8951
  contextWindow: 131072,
8488
- maxTokens: 131072,
8952
+ maxTokens: 32768,
8489
8953
  },
8490
8954
  "moonshotai/kimi-k2-0905": {
8491
8955
  id: "moonshotai/kimi-k2-0905",
@@ -8498,11 +8962,11 @@ export const MODELS = {
8498
8962
  cost: {
8499
8963
  input: 0.39999999999999997,
8500
8964
  output: 2,
8501
- cacheRead: 0.15,
8965
+ cacheRead: 0,
8502
8966
  cacheWrite: 0,
8503
8967
  },
8504
- contextWindow: 131072,
8505
- maxTokens: 4096,
8968
+ contextWindow: 262144,
8969
+ maxTokens: 262144,
8506
8970
  },
8507
8971
  "moonshotai/kimi-k2-thinking": {
8508
8972
  id: "moonshotai/kimi-k2-thinking",
@@ -8513,13 +8977,13 @@ export const MODELS = {
8513
8977
  reasoning: true,
8514
8978
  input: ["text"],
8515
8979
  cost: {
8516
- input: 0.47,
8517
- output: 2,
8518
- cacheRead: 0.14100000000000001,
8980
+ input: 0.6,
8981
+ output: 2.5,
8982
+ cacheRead: 0.15,
8519
8983
  cacheWrite: 0,
8520
8984
  },
8521
- contextWindow: 131072,
8522
- maxTokens: 4096,
8985
+ contextWindow: 262144,
8986
+ maxTokens: 262144,
8523
8987
  },
8524
8988
  "moonshotai/kimi-k2.5": {
8525
8989
  id: "moonshotai/kimi-k2.5",
@@ -8538,6 +9002,23 @@ export const MODELS = {
8538
9002
  contextWindow: 262144,
8539
9003
  maxTokens: 4096,
8540
9004
  },
9005
+ "moonshotai/kimi-k2.6": {
9006
+ id: "moonshotai/kimi-k2.6",
9007
+ name: "MoonshotAI: Kimi K2.6",
9008
+ api: "openai-completions",
9009
+ provider: "openrouter",
9010
+ baseUrl: "https://openrouter.ai/api/v1",
9011
+ reasoning: true,
9012
+ input: ["text", "image"],
9013
+ cost: {
9014
+ input: 0.7448,
9015
+ output: 4.655,
9016
+ cacheRead: 0.1463,
9017
+ cacheWrite: 0,
9018
+ },
9019
+ contextWindow: 256000,
9020
+ maxTokens: 65536,
9021
+ },
8541
9022
  "nex-agi/deepseek-v3.1-nex-n1": {
8542
9023
  id: "nex-agi/deepseek-v3.1-nex-n1",
8543
9024
  name: "Nex AGI: DeepSeek V3.1 Nex N1",
@@ -8587,7 +9068,7 @@ export const MODELS = {
8587
9068
  cacheWrite: 0,
8588
9069
  },
8589
9070
  contextWindow: 131072,
8590
- maxTokens: 4096,
9071
+ maxTokens: 16384,
8591
9072
  },
8592
9073
  "nvidia/nemotron-3-nano-30b-a3b": {
8593
9074
  id: "nvidia/nemotron-3-nano-30b-a3b",
@@ -8604,7 +9085,7 @@ export const MODELS = {
8604
9085
  cacheWrite: 0,
8605
9086
  },
8606
9087
  contextWindow: 262144,
8607
- maxTokens: 4096,
9088
+ maxTokens: 228000,
8608
9089
  },
8609
9090
  "nvidia/nemotron-3-nano-30b-a3b:free": {
8610
9091
  id: "nvidia/nemotron-3-nano-30b-a3b:free",
@@ -8632,9 +9113,9 @@ export const MODELS = {
8632
9113
  reasoning: true,
8633
9114
  input: ["text"],
8634
9115
  cost: {
8635
- input: 0.09999999999999999,
8636
- output: 0.5,
8637
- cacheRead: 0.09999999999999999,
9116
+ input: 0.09,
9117
+ output: 0.44999999999999996,
9118
+ cacheRead: 0,
8638
9119
  cacheWrite: 0,
8639
9120
  },
8640
9121
  contextWindow: 262144,
@@ -8689,7 +9170,7 @@ export const MODELS = {
8689
9170
  cacheWrite: 0,
8690
9171
  },
8691
9172
  contextWindow: 131072,
8692
- maxTokens: 4096,
9173
+ maxTokens: 16384,
8693
9174
  },
8694
9175
  "nvidia/nemotron-nano-9b-v2:free": {
8695
9176
  id: "nvidia/nemotron-nano-9b-v2:free",
@@ -8859,7 +9340,7 @@ export const MODELS = {
8859
9340
  cacheWrite: 0,
8860
9341
  },
8861
9342
  contextWindow: 1047576,
8862
- maxTokens: 32768,
9343
+ maxTokens: 4096,
8863
9344
  },
8864
9345
  "openai/gpt-4.1-mini": {
8865
9346
  id: "openai/gpt-4.1-mini",
@@ -8906,7 +9387,7 @@ export const MODELS = {
8906
9387
  cost: {
8907
9388
  input: 2.5,
8908
9389
  output: 10,
8909
- cacheRead: 1.25,
9390
+ cacheRead: 0,
8910
9391
  cacheWrite: 0,
8911
9392
  },
8912
9393
  contextWindow: 128000,
@@ -8998,38 +9479,21 @@ export const MODELS = {
8998
9479
  maxTokens: 16384,
8999
9480
  },
9000
9481
  "openai/gpt-4o-mini-2024-07-18": {
9001
- id: "openai/gpt-4o-mini-2024-07-18",
9002
- name: "OpenAI: GPT-4o-mini (2024-07-18)",
9003
- api: "openai-completions",
9004
- provider: "openrouter",
9005
- baseUrl: "https://openrouter.ai/api/v1",
9006
- reasoning: false,
9007
- input: ["text", "image"],
9008
- cost: {
9009
- input: 0.15,
9010
- output: 0.6,
9011
- cacheRead: 0.075,
9012
- cacheWrite: 0,
9013
- },
9014
- contextWindow: 128000,
9015
- maxTokens: 16384,
9016
- },
9017
- "openai/gpt-4o:extended": {
9018
- id: "openai/gpt-4o:extended",
9019
- name: "OpenAI: GPT-4o (extended)",
9482
+ id: "openai/gpt-4o-mini-2024-07-18",
9483
+ name: "OpenAI: GPT-4o-mini (2024-07-18)",
9020
9484
  api: "openai-completions",
9021
9485
  provider: "openrouter",
9022
9486
  baseUrl: "https://openrouter.ai/api/v1",
9023
9487
  reasoning: false,
9024
9488
  input: ["text", "image"],
9025
9489
  cost: {
9026
- input: 6,
9027
- output: 18,
9028
- cacheRead: 0,
9490
+ input: 0.15,
9491
+ output: 0.6,
9492
+ cacheRead: 0.075,
9029
9493
  cacheWrite: 0,
9030
9494
  },
9031
9495
  contextWindow: 128000,
9032
- maxTokens: 64000,
9496
+ maxTokens: 16384,
9033
9497
  },
9034
9498
  "openai/gpt-5": {
9035
9499
  id: "openai/gpt-5",
@@ -9065,40 +9529,6 @@ export const MODELS = {
9065
9529
  contextWindow: 400000,
9066
9530
  maxTokens: 128000,
9067
9531
  },
9068
- "openai/gpt-5-image": {
9069
- id: "openai/gpt-5-image",
9070
- name: "OpenAI: GPT-5 Image",
9071
- api: "openai-completions",
9072
- provider: "openrouter",
9073
- baseUrl: "https://openrouter.ai/api/v1",
9074
- reasoning: true,
9075
- input: ["text", "image"],
9076
- cost: {
9077
- input: 10,
9078
- output: 10,
9079
- cacheRead: 1.25,
9080
- cacheWrite: 0,
9081
- },
9082
- contextWindow: 400000,
9083
- maxTokens: 128000,
9084
- },
9085
- "openai/gpt-5-image-mini": {
9086
- id: "openai/gpt-5-image-mini",
9087
- name: "OpenAI: GPT-5 Image Mini",
9088
- api: "openai-completions",
9089
- provider: "openrouter",
9090
- baseUrl: "https://openrouter.ai/api/v1",
9091
- reasoning: true,
9092
- input: ["text", "image"],
9093
- cost: {
9094
- input: 2.5,
9095
- output: 2,
9096
- cacheRead: 0.25,
9097
- cacheWrite: 0,
9098
- },
9099
- contextWindow: 400000,
9100
- maxTokens: 128000,
9101
- },
9102
9532
  "openai/gpt-5-mini": {
9103
9533
  id: "openai/gpt-5-mini",
9104
9534
  name: "OpenAI: GPT-5 Mini",
@@ -9127,11 +9557,11 @@ export const MODELS = {
9127
9557
  cost: {
9128
9558
  input: 0.049999999999999996,
9129
9559
  output: 0.39999999999999997,
9130
- cacheRead: 0.005,
9560
+ cacheRead: 0.01,
9131
9561
  cacheWrite: 0,
9132
9562
  },
9133
9563
  contextWindow: 400000,
9134
- maxTokens: 128000,
9564
+ maxTokens: 4096,
9135
9565
  },
9136
9566
  "openai/gpt-5-pro": {
9137
9567
  id: "openai/gpt-5-pro",
@@ -9161,7 +9591,7 @@ export const MODELS = {
9161
9591
  cost: {
9162
9592
  input: 1.25,
9163
9593
  output: 10,
9164
- cacheRead: 0.125,
9594
+ cacheRead: 0.13,
9165
9595
  cacheWrite: 0,
9166
9596
  },
9167
9597
  contextWindow: 400000,
@@ -9229,11 +9659,11 @@ export const MODELS = {
9229
9659
  cost: {
9230
9660
  input: 0.25,
9231
9661
  output: 2,
9232
- cacheRead: 0.024999999999999998,
9662
+ cacheRead: 0.03,
9233
9663
  cacheWrite: 0,
9234
9664
  },
9235
9665
  contextWindow: 400000,
9236
- maxTokens: 100000,
9666
+ maxTokens: 128000,
9237
9667
  },
9238
9668
  "openai/gpt-5.2": {
9239
9669
  id: "openai/gpt-5.2",
@@ -9267,7 +9697,7 @@ export const MODELS = {
9267
9697
  cacheWrite: 0,
9268
9698
  },
9269
9699
  contextWindow: 128000,
9270
- maxTokens: 16384,
9700
+ maxTokens: 32000,
9271
9701
  },
9272
9702
  "openai/gpt-5.2-codex": {
9273
9703
  id: "openai/gpt-5.2-codex",
@@ -9405,6 +9835,40 @@ export const MODELS = {
9405
9835
  contextWindow: 1050000,
9406
9836
  maxTokens: 128000,
9407
9837
  },
9838
+ "openai/gpt-5.5": {
9839
+ id: "openai/gpt-5.5",
9840
+ name: "OpenAI: GPT-5.5",
9841
+ api: "openai-completions",
9842
+ provider: "openrouter",
9843
+ baseUrl: "https://openrouter.ai/api/v1",
9844
+ reasoning: true,
9845
+ input: ["text", "image"],
9846
+ cost: {
9847
+ input: 5,
9848
+ output: 30,
9849
+ cacheRead: 0.5,
9850
+ cacheWrite: 0,
9851
+ },
9852
+ contextWindow: 1050000,
9853
+ maxTokens: 128000,
9854
+ },
9855
+ "openai/gpt-5.5-pro": {
9856
+ id: "openai/gpt-5.5-pro",
9857
+ name: "OpenAI: GPT-5.5 Pro",
9858
+ api: "openai-completions",
9859
+ provider: "openrouter",
9860
+ baseUrl: "https://openrouter.ai/api/v1",
9861
+ reasoning: true,
9862
+ input: ["text", "image"],
9863
+ cost: {
9864
+ input: 30,
9865
+ output: 180,
9866
+ cacheRead: 0,
9867
+ cacheWrite: 0,
9868
+ },
9869
+ contextWindow: 1050000,
9870
+ maxTokens: 128000,
9871
+ },
9408
9872
  "openai/gpt-audio": {
9409
9873
  id: "openai/gpt-audio",
9410
9874
  name: "OpenAI: GPT Audio",
@@ -9454,7 +9918,7 @@ export const MODELS = {
9454
9918
  cacheWrite: 0,
9455
9919
  },
9456
9920
  contextWindow: 131072,
9457
- maxTokens: 4096,
9921
+ maxTokens: 131072,
9458
9922
  },
9459
9923
  "openai/gpt-oss-120b:free": {
9460
9924
  id: "openai/gpt-oss-120b:free",
@@ -9483,8 +9947,8 @@ export const MODELS = {
9483
9947
  input: ["text"],
9484
9948
  cost: {
9485
9949
  input: 0.03,
9486
- output: 0.11,
9487
- cacheRead: 0.015,
9950
+ output: 0.14,
9951
+ cacheRead: 0,
9488
9952
  cacheWrite: 0,
9489
9953
  },
9490
9954
  contextWindow: 131072,
@@ -9505,7 +9969,7 @@ export const MODELS = {
9505
9969
  cacheWrite: 0,
9506
9970
  },
9507
9971
  contextWindow: 131072,
9508
- maxTokens: 131072,
9972
+ maxTokens: 8192,
9509
9973
  },
9510
9974
  "openai/gpt-oss-safeguard-20b": {
9511
9975
  id: "openai/gpt-oss-safeguard-20b",
@@ -9791,7 +10255,7 @@ export const MODELS = {
9791
10255
  input: 0.26,
9792
10256
  output: 0.78,
9793
10257
  cacheRead: 0.052000000000000005,
9794
- cacheWrite: 0,
10258
+ cacheWrite: 0.325,
9795
10259
  },
9796
10260
  contextWindow: 1000000,
9797
10261
  maxTokens: 32768,
@@ -9808,7 +10272,7 @@ export const MODELS = {
9808
10272
  input: 0.26,
9809
10273
  output: 0.78,
9810
10274
  cacheRead: 0,
9811
- cacheWrite: 0,
10275
+ cacheWrite: 0.325,
9812
10276
  },
9813
10277
  contextWindow: 1000000,
9814
10278
  maxTokens: 32768,
@@ -9825,7 +10289,7 @@ export const MODELS = {
9825
10289
  input: 0.26,
9826
10290
  output: 0.78,
9827
10291
  cacheRead: 0,
9828
- cacheWrite: 0,
10292
+ cacheWrite: 0.325,
9829
10293
  },
9830
10294
  contextWindow: 1000000,
9831
10295
  maxTokens: 32768,
@@ -9904,7 +10368,7 @@ export const MODELS = {
9904
10368
  api: "openai-completions",
9905
10369
  provider: "openrouter",
9906
10370
  baseUrl: "https://openrouter.ai/api/v1",
9907
- reasoning: true,
10371
+ reasoning: false,
9908
10372
  input: ["text"],
9909
10373
  cost: {
9910
10374
  input: 0.071,
@@ -9913,7 +10377,7 @@ export const MODELS = {
9913
10377
  cacheWrite: 0,
9914
10378
  },
9915
10379
  contextWindow: 262144,
9916
- maxTokens: 4096,
10380
+ maxTokens: 16384,
9917
10381
  },
9918
10382
  "qwen/qwen3-235b-a22b-thinking-2507": {
9919
10383
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -9947,7 +10411,7 @@ export const MODELS = {
9947
10411
  cacheWrite: 0,
9948
10412
  },
9949
10413
  contextWindow: 40960,
9950
- maxTokens: 40960,
10414
+ maxTokens: 16384,
9951
10415
  },
9952
10416
  "qwen/qwen3-30b-a3b-instruct-2507": {
9953
10417
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -10032,7 +10496,7 @@ export const MODELS = {
10032
10496
  cacheWrite: 0,
10033
10497
  },
10034
10498
  contextWindow: 262144,
10035
- maxTokens: 4096,
10499
+ maxTokens: 65536,
10036
10500
  },
10037
10501
  "qwen/qwen3-coder-30b-a3b-instruct": {
10038
10502
  id: "qwen/qwen3-coder-30b-a3b-instruct",
@@ -10063,7 +10527,7 @@ export const MODELS = {
10063
10527
  input: 0.195,
10064
10528
  output: 0.975,
10065
10529
  cacheRead: 0.039,
10066
- cacheWrite: 0,
10530
+ cacheWrite: 0.24375,
10067
10531
  },
10068
10532
  contextWindow: 1000000,
10069
10533
  maxTokens: 65536,
@@ -10077,13 +10541,13 @@ export const MODELS = {
10077
10541
  reasoning: false,
10078
10542
  input: ["text"],
10079
10543
  cost: {
10080
- input: 0.12,
10081
- output: 0.75,
10082
- cacheRead: 0.06,
10544
+ input: 0.14,
10545
+ output: 0.7999999999999999,
10546
+ cacheRead: 0.09,
10083
10547
  cacheWrite: 0,
10084
10548
  },
10085
10549
  contextWindow: 262144,
10086
- maxTokens: 65536,
10550
+ maxTokens: 262144,
10087
10551
  },
10088
10552
  "qwen/qwen3-coder-plus": {
10089
10553
  id: "qwen/qwen3-coder-plus",
@@ -10097,7 +10561,7 @@ export const MODELS = {
10097
10561
  input: 0.65,
10098
10562
  output: 3.25,
10099
10563
  cacheRead: 0.13,
10100
- cacheWrite: 0,
10564
+ cacheWrite: 0.8125,
10101
10565
  },
10102
10566
  contextWindow: 1000000,
10103
10567
  maxTokens: 65536,
@@ -10131,7 +10595,7 @@ export const MODELS = {
10131
10595
  input: 0.78,
10132
10596
  output: 3.9,
10133
10597
  cacheRead: 0.156,
10134
- cacheWrite: 0,
10598
+ cacheWrite: 0.975,
10135
10599
  },
10136
10600
  contextWindow: 262144,
10137
10601
  maxTokens: 32768,
@@ -10168,7 +10632,7 @@ export const MODELS = {
10168
10632
  cacheWrite: 0,
10169
10633
  },
10170
10634
  contextWindow: 262144,
10171
- maxTokens: 4096,
10635
+ maxTokens: 16384,
10172
10636
  },
10173
10637
  "qwen/qwen3-next-80b-a3b-instruct:free": {
10174
10638
  id: "qwen/qwen3-next-80b-a3b-instruct:free",
@@ -10219,7 +10683,7 @@ export const MODELS = {
10219
10683
  cacheWrite: 0,
10220
10684
  },
10221
10685
  contextWindow: 262144,
10222
- maxTokens: 4096,
10686
+ maxTokens: 16384,
10223
10687
  },
10224
10688
  "qwen/qwen3-vl-235b-a22b-thinking": {
10225
10689
  id: "qwen/qwen3-vl-235b-a22b-thinking",
@@ -10385,7 +10849,7 @@ export const MODELS = {
10385
10849
  cost: {
10386
10850
  input: 0.39,
10387
10851
  output: 2.34,
10388
- cacheRead: 0,
10852
+ cacheRead: 0.195,
10389
10853
  cacheWrite: 0,
10390
10854
  },
10391
10855
  contextWindow: 262144,
@@ -10400,13 +10864,13 @@ export const MODELS = {
10400
10864
  reasoning: true,
10401
10865
  input: ["text", "image"],
10402
10866
  cost: {
10403
- input: 0.049999999999999996,
10867
+ input: 0.09999999999999999,
10404
10868
  output: 0.15,
10405
10869
  cacheRead: 0,
10406
10870
  cacheWrite: 0,
10407
10871
  },
10408
- contextWindow: 256000,
10409
- maxTokens: 32768,
10872
+ contextWindow: 262144,
10873
+ maxTokens: 4096,
10410
10874
  },
10411
10875
  "qwen/qwen3.5-flash-02-23": {
10412
10876
  id: "qwen/qwen3.5-flash-02-23",
@@ -10420,7 +10884,7 @@ export const MODELS = {
10420
10884
  input: 0.065,
10421
10885
  output: 0.26,
10422
10886
  cacheRead: 0,
10423
- cacheWrite: 0,
10887
+ cacheWrite: 0.08125,
10424
10888
  },
10425
10889
  contextWindow: 1000000,
10426
10890
  maxTokens: 65536,
@@ -10437,24 +10901,24 @@ export const MODELS = {
10437
10901
  input: 0.26,
10438
10902
  output: 1.56,
10439
10903
  cacheRead: 0,
10440
- cacheWrite: 0,
10904
+ cacheWrite: 0.325,
10441
10905
  },
10442
10906
  contextWindow: 1000000,
10443
10907
  maxTokens: 65536,
10444
10908
  },
10445
- "qwen/qwen3.6-plus:free": {
10446
- id: "qwen/qwen3.6-plus:free",
10447
- name: "Qwen: Qwen3.6 Plus (free)",
10909
+ "qwen/qwen3.6-plus": {
10910
+ id: "qwen/qwen3.6-plus",
10911
+ name: "Qwen: Qwen3.6 Plus",
10448
10912
  api: "openai-completions",
10449
10913
  provider: "openrouter",
10450
10914
  baseUrl: "https://openrouter.ai/api/v1",
10451
10915
  reasoning: true,
10452
10916
  input: ["text", "image"],
10453
10917
  cost: {
10454
- input: 0,
10455
- output: 0,
10918
+ input: 0.325,
10919
+ output: 1.95,
10456
10920
  cacheRead: 0,
10457
- cacheWrite: 0,
10921
+ cacheWrite: 0.40625,
10458
10922
  },
10459
10923
  contextWindow: 1000000,
10460
10924
  maxTokens: 65536,
@@ -10561,9 +11025,9 @@ export const MODELS = {
10561
11025
  contextWindow: 262144,
10562
11026
  maxTokens: 65536,
10563
11027
  },
10564
- "stepfun/step-3.5-flash:free": {
10565
- id: "stepfun/step-3.5-flash:free",
10566
- name: "StepFun: Step 3.5 Flash (free)",
11028
+ "tencent/hy3-preview:free": {
11029
+ id: "tencent/hy3-preview:free",
11030
+ name: "Tencent: Hy3 preview (free)",
10567
11031
  api: "openai-completions",
10568
11032
  provider: "openrouter",
10569
11033
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10575,8 +11039,8 @@ export const MODELS = {
10575
11039
  cacheRead: 0,
10576
11040
  cacheWrite: 0,
10577
11041
  },
10578
- contextWindow: 256000,
10579
- maxTokens: 256000,
11042
+ contextWindow: 262144,
11043
+ maxTokens: 262144,
10580
11044
  },
10581
11045
  "thedrummer/rocinante-12b": {
10582
11046
  id: "thedrummer/rocinante-12b",
@@ -10850,6 +11314,40 @@ export const MODELS = {
10850
11314
  contextWindow: 1048576,
10851
11315
  maxTokens: 131072,
10852
11316
  },
11317
+ "xiaomi/mimo-v2.5": {
11318
+ id: "xiaomi/mimo-v2.5",
11319
+ name: "Xiaomi: MiMo-V2.5",
11320
+ api: "openai-completions",
11321
+ provider: "openrouter",
11322
+ baseUrl: "https://openrouter.ai/api/v1",
11323
+ reasoning: true,
11324
+ input: ["text", "image"],
11325
+ cost: {
11326
+ input: 0.39999999999999997,
11327
+ output: 2,
11328
+ cacheRead: 0.08,
11329
+ cacheWrite: 0,
11330
+ },
11331
+ contextWindow: 1048576,
11332
+ maxTokens: 131072,
11333
+ },
11334
+ "xiaomi/mimo-v2.5-pro": {
11335
+ id: "xiaomi/mimo-v2.5-pro",
11336
+ name: "Xiaomi: MiMo-V2.5-Pro",
11337
+ api: "openai-completions",
11338
+ provider: "openrouter",
11339
+ baseUrl: "https://openrouter.ai/api/v1",
11340
+ reasoning: true,
11341
+ input: ["text"],
11342
+ cost: {
11343
+ input: 1,
11344
+ output: 3,
11345
+ cacheRead: 0.19999999999999998,
11346
+ cacheWrite: 0,
11347
+ },
11348
+ contextWindow: 1048576,
11349
+ maxTokens: 131072,
11350
+ },
10853
11351
  "z-ai/glm-4-32b": {
10854
11352
  id: "z-ai/glm-4-32b",
10855
11353
  name: "Z.ai: GLM 4 32B ",
@@ -10963,11 +11461,11 @@ export const MODELS = {
10963
11461
  cost: {
10964
11462
  input: 0.3,
10965
11463
  output: 0.8999999999999999,
10966
- cacheRead: 0,
11464
+ cacheRead: 0.049999999999999996,
10967
11465
  cacheWrite: 0,
10968
11466
  },
10969
11467
  contextWindow: 131072,
10970
- maxTokens: 131072,
11468
+ maxTokens: 24000,
10971
11469
  },
10972
11470
  "z-ai/glm-4.7": {
10973
11471
  id: "z-ai/glm-4.7",
@@ -10978,13 +11476,13 @@ export const MODELS = {
10978
11476
  reasoning: true,
10979
11477
  input: ["text"],
10980
11478
  cost: {
10981
- input: 0.39,
10982
- output: 1.75,
10983
- cacheRead: 0.195,
11479
+ input: 0.38,
11480
+ output: 1.74,
11481
+ cacheRead: 0,
10984
11482
  cacheWrite: 0,
10985
11483
  },
10986
11484
  contextWindow: 202752,
10987
- maxTokens: 65535,
11485
+ maxTokens: 4096,
10988
11486
  },
10989
11487
  "z-ai/glm-4.7-flash": {
10990
11488
  id: "z-ai/glm-4.7-flash",
@@ -10997,11 +11495,11 @@ export const MODELS = {
10997
11495
  cost: {
10998
11496
  input: 0.06,
10999
11497
  output: 0.39999999999999997,
11000
- cacheRead: 0.0100000002,
11498
+ cacheRead: 0.01,
11001
11499
  cacheWrite: 0,
11002
11500
  },
11003
11501
  contextWindow: 202752,
11004
- maxTokens: 4096,
11502
+ maxTokens: 16384,
11005
11503
  },
11006
11504
  "z-ai/glm-5": {
11007
11505
  id: "z-ai/glm-5",
@@ -11017,8 +11515,8 @@ export const MODELS = {
11017
11515
  cacheRead: 0.119,
11018
11516
  cacheWrite: 0,
11019
11517
  },
11020
- contextWindow: 80000,
11021
- maxTokens: 131072,
11518
+ contextWindow: 202752,
11519
+ maxTokens: 16384,
11022
11520
  },
11023
11521
  "z-ai/glm-5-turbo": {
11024
11522
  id: "z-ai/glm-5-turbo",
@@ -11037,6 +11535,23 @@ export const MODELS = {
11037
11535
  contextWindow: 202752,
11038
11536
  maxTokens: 131072,
11039
11537
  },
11538
+ "z-ai/glm-5.1": {
11539
+ id: "z-ai/glm-5.1",
11540
+ name: "Z.ai: GLM 5.1",
11541
+ api: "openai-completions",
11542
+ provider: "openrouter",
11543
+ baseUrl: "https://openrouter.ai/api/v1",
11544
+ reasoning: true,
11545
+ input: ["text"],
11546
+ cost: {
11547
+ input: 1.0499999999999998,
11548
+ output: 3.5,
11549
+ cacheRead: 0.5249999999999999,
11550
+ cacheWrite: 0,
11551
+ },
11552
+ contextWindow: 202752,
11553
+ maxTokens: 65535,
11554
+ },
11040
11555
  "z-ai/glm-5v-turbo": {
11041
11556
  id: "z-ai/glm-5v-turbo",
11042
11557
  name: "Z.ai: GLM 5V Turbo",
@@ -11054,6 +11569,23 @@ export const MODELS = {
11054
11569
  contextWindow: 202752,
11055
11570
  maxTokens: 131072,
11056
11571
  },
11572
+ "~anthropic/claude-opus-latest": {
11573
+ id: "~anthropic/claude-opus-latest",
11574
+ name: "Anthropic: Claude Opus Latest",
11575
+ api: "openai-completions",
11576
+ provider: "openrouter",
11577
+ baseUrl: "https://openrouter.ai/api/v1",
11578
+ reasoning: true,
11579
+ input: ["text", "image"],
11580
+ cost: {
11581
+ input: 5,
11582
+ output: 25,
11583
+ cacheRead: 0.5,
11584
+ cacheWrite: 6.25,
11585
+ },
11586
+ contextWindow: 1000000,
11587
+ maxTokens: 128000,
11588
+ },
11057
11589
  },
11058
11590
  "vercel-ai-gateway": {
11059
11591
  "alibaba/qwen-3-14b": {
@@ -11082,13 +11614,13 @@ export const MODELS = {
11082
11614
  reasoning: false,
11083
11615
  input: ["text"],
11084
11616
  cost: {
11085
- input: 0.22,
11086
- output: 0.88,
11087
- cacheRead: 0.11,
11617
+ input: 0.6,
11618
+ output: 1.2,
11619
+ cacheRead: 0.6,
11088
11620
  cacheWrite: 0,
11089
11621
  },
11090
- contextWindow: 32768,
11091
- maxTokens: 16384,
11622
+ contextWindow: 131000,
11623
+ maxTokens: 40000,
11092
11624
  },
11093
11625
  "alibaba/qwen-3-30b": {
11094
11626
  id: "alibaba/qwen-3-30b",
@@ -11124,6 +11656,23 @@ export const MODELS = {
11124
11656
  contextWindow: 128000,
11125
11657
  maxTokens: 8192,
11126
11658
  },
11659
+ "alibaba/qwen-3.6-max-preview": {
11660
+ id: "alibaba/qwen-3.6-max-preview",
11661
+ name: "Qwen 3.6 Max Preview",
11662
+ api: "anthropic-messages",
11663
+ provider: "vercel-ai-gateway",
11664
+ baseUrl: "https://ai-gateway.vercel.sh",
11665
+ reasoning: true,
11666
+ input: ["text", "image"],
11667
+ cost: {
11668
+ input: 1.3,
11669
+ output: 7.8,
11670
+ cacheRead: 0.26,
11671
+ cacheWrite: 1.625,
11672
+ },
11673
+ contextWindow: 240000,
11674
+ maxTokens: 64000,
11675
+ },
11127
11676
  "alibaba/qwen3-235b-a22b-thinking": {
11128
11677
  id: "alibaba/qwen3-235b-a22b-thinking",
11129
11678
  name: "Qwen3 235B A22B Thinking 2507",
@@ -11464,6 +12013,23 @@ export const MODELS = {
11464
12013
  contextWindow: 1000000,
11465
12014
  maxTokens: 128000,
11466
12015
  },
12016
+ "anthropic/claude-opus-4.7": {
12017
+ id: "anthropic/claude-opus-4.7",
12018
+ name: "Claude Opus 4.7",
12019
+ api: "anthropic-messages",
12020
+ provider: "vercel-ai-gateway",
12021
+ baseUrl: "https://ai-gateway.vercel.sh",
12022
+ reasoning: true,
12023
+ input: ["text", "image"],
12024
+ cost: {
12025
+ input: 5,
12026
+ output: 25,
12027
+ cacheRead: 0.5,
12028
+ cacheWrite: 6.25,
12029
+ },
12030
+ contextWindow: 1000000,
12031
+ maxTokens: 128000,
12032
+ },
11467
12033
  "anthropic/claude-sonnet-4": {
11468
12034
  id: "anthropic/claude-sonnet-4",
11469
12035
  name: "Claude Sonnet 4",
@@ -11660,30 +12226,64 @@ export const MODELS = {
11660
12226
  reasoning: false,
11661
12227
  input: ["text"],
11662
12228
  cost: {
11663
- input: 0.28,
11664
- output: 0.42,
12229
+ input: 0.28,
12230
+ output: 0.42,
12231
+ cacheRead: 0.028,
12232
+ cacheWrite: 0,
12233
+ },
12234
+ contextWindow: 128000,
12235
+ maxTokens: 8000,
12236
+ },
12237
+ "deepseek/deepseek-v3.2-thinking": {
12238
+ id: "deepseek/deepseek-v3.2-thinking",
12239
+ name: "DeepSeek V3.2 Thinking",
12240
+ api: "anthropic-messages",
12241
+ provider: "vercel-ai-gateway",
12242
+ baseUrl: "https://ai-gateway.vercel.sh",
12243
+ reasoning: true,
12244
+ input: ["text"],
12245
+ cost: {
12246
+ input: 0.28,
12247
+ output: 0.42,
12248
+ cacheRead: 0.028,
12249
+ cacheWrite: 0,
12250
+ },
12251
+ contextWindow: 128000,
12252
+ maxTokens: 64000,
12253
+ },
12254
+ "deepseek/deepseek-v4-flash": {
12255
+ id: "deepseek/deepseek-v4-flash",
12256
+ name: "DeepSeek V4 Flash",
12257
+ api: "anthropic-messages",
12258
+ provider: "vercel-ai-gateway",
12259
+ baseUrl: "https://ai-gateway.vercel.sh",
12260
+ reasoning: true,
12261
+ input: ["text"],
12262
+ cost: {
12263
+ input: 0.14,
12264
+ output: 0.28,
11665
12265
  cacheRead: 0.028,
11666
12266
  cacheWrite: 0,
11667
12267
  },
11668
- contextWindow: 128000,
11669
- maxTokens: 8000,
12268
+ contextWindow: 1000000,
12269
+ maxTokens: 384000,
11670
12270
  },
11671
- "deepseek/deepseek-v3.2-thinking": {
11672
- id: "deepseek/deepseek-v3.2-thinking",
11673
- name: "DeepSeek V3.2 Thinking",
12271
+ "deepseek/deepseek-v4-pro": {
12272
+ id: "deepseek/deepseek-v4-pro",
12273
+ name: "DeepSeek V4 Pro",
11674
12274
  api: "anthropic-messages",
11675
12275
  provider: "vercel-ai-gateway",
11676
12276
  baseUrl: "https://ai-gateway.vercel.sh",
11677
12277
  reasoning: true,
11678
12278
  input: ["text"],
11679
12279
  cost: {
11680
- input: 0.28,
11681
- output: 0.42,
11682
- cacheRead: 0.028,
12280
+ input: 1.74,
12281
+ output: 3.48,
12282
+ cacheRead: 0.145,
11683
12283
  cacheWrite: 0,
11684
12284
  },
11685
- contextWindow: 128000,
11686
- maxTokens: 64000,
12285
+ contextWindow: 1000000,
12286
+ maxTokens: 384000,
11687
12287
  },
11688
12288
  "google/gemini-2.0-flash": {
11689
12289
  id: "google/gemini-2.0-flash",
@@ -11815,7 +12415,7 @@ export const MODELS = {
11815
12415
  cost: {
11816
12416
  input: 0.25,
11817
12417
  output: 1.5,
11818
- cacheRead: 0,
12418
+ cacheRead: 0.03,
11819
12419
  cacheWrite: 0,
11820
12420
  },
11821
12421
  contextWindow: 1000000,
@@ -11938,7 +12538,7 @@ export const MODELS = {
11938
12538
  cacheWrite: 0,
11939
12539
  },
11940
12540
  contextWindow: 128000,
11941
- maxTokens: 8192,
12541
+ maxTokens: 100000,
11942
12542
  },
11943
12543
  "meta/llama-3.1-70b": {
11944
12544
  id: "meta/llama-3.1-70b",
@@ -12450,6 +13050,23 @@ export const MODELS = {
12450
13050
  contextWindow: 262114,
12451
13051
  maxTokens: 262114,
12452
13052
  },
13053
+ "moonshotai/kimi-k2.6": {
13054
+ id: "moonshotai/kimi-k2.6",
13055
+ name: "Kimi K2.6",
13056
+ api: "anthropic-messages",
13057
+ provider: "vercel-ai-gateway",
13058
+ baseUrl: "https://ai-gateway.vercel.sh",
13059
+ reasoning: true,
13060
+ input: ["text", "image"],
13061
+ cost: {
13062
+ input: 0.95,
13063
+ output: 4,
13064
+ cacheRead: 0.16,
13065
+ cacheWrite: 0,
13066
+ },
13067
+ contextWindow: 262000,
13068
+ maxTokens: 262000,
13069
+ },
12453
13070
  "nvidia/nemotron-nano-12b-v2-vl": {
12454
13071
  id: "nvidia/nemotron-nano-12b-v2-vl",
12455
13072
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -12943,26 +13560,43 @@ export const MODELS = {
12943
13560
  contextWindow: 1050000,
12944
13561
  maxTokens: 128000,
12945
13562
  },
12946
- "openai/gpt-oss-120b": {
12947
- id: "openai/gpt-oss-120b",
12948
- name: "gpt-oss-120b",
13563
+ "openai/gpt-5.5": {
13564
+ id: "openai/gpt-5.5",
13565
+ name: "GPT 5.5",
12949
13566
  api: "anthropic-messages",
12950
13567
  provider: "vercel-ai-gateway",
12951
13568
  baseUrl: "https://ai-gateway.vercel.sh",
12952
13569
  reasoning: true,
12953
- input: ["text"],
13570
+ input: ["text", "image"],
12954
13571
  cost: {
12955
- input: 0.15,
12956
- output: 0.6,
13572
+ input: 5,
13573
+ output: 30,
13574
+ cacheRead: 0.5,
13575
+ cacheWrite: 0,
13576
+ },
13577
+ contextWindow: 1000000,
13578
+ maxTokens: 128000,
13579
+ },
13580
+ "openai/gpt-5.5-pro": {
13581
+ id: "openai/gpt-5.5-pro",
13582
+ name: "GPT 5.5 Pro",
13583
+ api: "anthropic-messages",
13584
+ provider: "vercel-ai-gateway",
13585
+ baseUrl: "https://ai-gateway.vercel.sh",
13586
+ reasoning: true,
13587
+ input: ["text", "image"],
13588
+ cost: {
13589
+ input: 30,
13590
+ output: 180,
12957
13591
  cacheRead: 0,
12958
13592
  cacheWrite: 0,
12959
13593
  },
12960
- contextWindow: 128000,
12961
- maxTokens: 8192,
13594
+ contextWindow: 1000000,
13595
+ maxTokens: 128000,
12962
13596
  },
12963
13597
  "openai/gpt-oss-20b": {
12964
13598
  id: "openai/gpt-oss-20b",
12965
- name: "gpt-oss-20b",
13599
+ name: "GPT OSS 120B",
12966
13600
  api: "anthropic-messages",
12967
13601
  provider: "vercel-ai-gateway",
12968
13602
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -12979,7 +13613,7 @@ export const MODELS = {
12979
13613
  },
12980
13614
  "openai/gpt-oss-safeguard-20b": {
12981
13615
  id: "openai/gpt-oss-safeguard-20b",
12982
- name: "gpt-oss-safeguard-20b",
13616
+ name: "GPT OSS Safeguard 20B",
12983
13617
  api: "anthropic-messages",
12984
13618
  provider: "vercel-ai-gateway",
12985
13619
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13147,23 +13781,6 @@ export const MODELS = {
13147
13781
  contextWindow: 131072,
13148
13782
  maxTokens: 131072,
13149
13783
  },
13150
- "xai/grok-2-vision": {
13151
- id: "xai/grok-2-vision",
13152
- name: "Grok 2 Vision",
13153
- api: "anthropic-messages",
13154
- provider: "vercel-ai-gateway",
13155
- baseUrl: "https://ai-gateway.vercel.sh",
13156
- reasoning: false,
13157
- input: ["text", "image"],
13158
- cost: {
13159
- input: 2,
13160
- output: 10,
13161
- cacheRead: 0,
13162
- cacheWrite: 0,
13163
- },
13164
- contextWindow: 32768,
13165
- maxTokens: 32768,
13166
- },
13167
13784
  "xai/grok-3": {
13168
13785
  id: "xai/grok-3",
13169
13786
  name: "Grok 3 Beta",
@@ -13256,7 +13873,7 @@ export const MODELS = {
13256
13873
  provider: "vercel-ai-gateway",
13257
13874
  baseUrl: "https://ai-gateway.vercel.sh",
13258
13875
  reasoning: false,
13259
- input: ["text"],
13876
+ input: ["text", "image"],
13260
13877
  cost: {
13261
13878
  input: 0.19999999999999998,
13262
13879
  output: 0.5,
@@ -13273,7 +13890,7 @@ export const MODELS = {
13273
13890
  provider: "vercel-ai-gateway",
13274
13891
  baseUrl: "https://ai-gateway.vercel.sh",
13275
13892
  reasoning: true,
13276
- input: ["text"],
13893
+ input: ["text", "image"],
13277
13894
  cost: {
13278
13895
  input: 0.19999999999999998,
13279
13896
  output: 0.5,
@@ -13290,7 +13907,7 @@ export const MODELS = {
13290
13907
  provider: "vercel-ai-gateway",
13291
13908
  baseUrl: "https://ai-gateway.vercel.sh",
13292
13909
  reasoning: false,
13293
- input: ["text"],
13910
+ input: ["text", "image"],
13294
13911
  cost: {
13295
13912
  input: 0.19999999999999998,
13296
13913
  output: 0.5,
@@ -13307,7 +13924,7 @@ export const MODELS = {
13307
13924
  provider: "vercel-ai-gateway",
13308
13925
  baseUrl: "https://ai-gateway.vercel.sh",
13309
13926
  reasoning: true,
13310
- input: ["text"],
13927
+ input: ["text", "image"],
13311
13928
  cost: {
13312
13929
  input: 0.19999999999999998,
13313
13930
  output: 0.5,
@@ -13324,7 +13941,7 @@ export const MODELS = {
13324
13941
  provider: "vercel-ai-gateway",
13325
13942
  baseUrl: "https://ai-gateway.vercel.sh",
13326
13943
  reasoning: true,
13327
- input: ["text"],
13944
+ input: ["text", "image"],
13328
13945
  cost: {
13329
13946
  input: 2,
13330
13947
  output: 6,
@@ -13341,7 +13958,7 @@ export const MODELS = {
13341
13958
  provider: "vercel-ai-gateway",
13342
13959
  baseUrl: "https://ai-gateway.vercel.sh",
13343
13960
  reasoning: true,
13344
- input: ["text"],
13961
+ input: ["text", "image"],
13345
13962
  cost: {
13346
13963
  input: 2,
13347
13964
  output: 6,
@@ -13445,9 +14062,9 @@ export const MODELS = {
13445
14062
  reasoning: true,
13446
14063
  input: ["text"],
13447
14064
  cost: {
13448
- input: 0.09,
13449
- output: 0.29,
13450
- cacheRead: 0.045,
14065
+ input: 0.09999999999999999,
14066
+ output: 0.3,
14067
+ cacheRead: 0.01,
13451
14068
  cacheWrite: 0,
13452
14069
  },
13453
14070
  contextWindow: 262144,
@@ -13581,13 +14198,13 @@ export const MODELS = {
13581
14198
  reasoning: true,
13582
14199
  input: ["text"],
13583
14200
  cost: {
13584
- input: 0.6,
13585
- output: 2.2,
13586
- cacheRead: 0.11,
14201
+ input: 2.25,
14202
+ output: 2.75,
14203
+ cacheRead: 2.25,
13587
14204
  cacheWrite: 0,
13588
14205
  },
13589
- contextWindow: 200000,
13590
- maxTokens: 120000,
14206
+ contextWindow: 131000,
14207
+ maxTokens: 40000,
13591
14208
  },
13592
14209
  "zai/glm-4.7-flash": {
13593
14210
  id: "zai/glm-4.7-flash",
@@ -13657,6 +14274,23 @@ export const MODELS = {
13657
14274
  contextWindow: 202800,
13658
14275
  maxTokens: 131100,
13659
14276
  },
14277
+ "zai/glm-5.1": {
14278
+ id: "zai/glm-5.1",
14279
+ name: "GLM 5.1",
14280
+ api: "anthropic-messages",
14281
+ provider: "vercel-ai-gateway",
14282
+ baseUrl: "https://ai-gateway.vercel.sh",
14283
+ reasoning: true,
14284
+ input: ["text"],
14285
+ cost: {
14286
+ input: 1.4,
14287
+ output: 4.4,
14288
+ cacheRead: 0.26,
14289
+ cacheWrite: 0,
14290
+ },
14291
+ contextWindow: 202800,
14292
+ maxTokens: 64000,
14293
+ },
13660
14294
  "zai/glm-5v-turbo": {
13661
14295
  id: "zai/glm-5v-turbo",
13662
14296
  name: "GLM 5V Turbo",
@@ -14086,24 +14720,6 @@ export const MODELS = {
14086
14720
  },
14087
14721
  },
14088
14722
  "zai": {
14089
- "glm-4.5": {
14090
- id: "glm-4.5",
14091
- name: "GLM-4.5",
14092
- api: "openai-completions",
14093
- provider: "zai",
14094
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14095
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14096
- reasoning: true,
14097
- input: ["text"],
14098
- cost: {
14099
- input: 0.6,
14100
- output: 2.2,
14101
- cacheRead: 0.11,
14102
- cacheWrite: 0,
14103
- },
14104
- contextWindow: 131072,
14105
- maxTokens: 98304,
14106
- },
14107
14723
  "glm-4.5-air": {
14108
14724
  id: "glm-4.5-air",
14109
14725
  name: "GLM-4.5-Air",
@@ -14113,24 +14729,6 @@ export const MODELS = {
14113
14729
  compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14114
14730
  reasoning: true,
14115
14731
  input: ["text"],
14116
- cost: {
14117
- input: 0.2,
14118
- output: 1.1,
14119
- cacheRead: 0.03,
14120
- cacheWrite: 0,
14121
- },
14122
- contextWindow: 131072,
14123
- maxTokens: 98304,
14124
- },
14125
- "glm-4.5-flash": {
14126
- id: "glm-4.5-flash",
14127
- name: "GLM-4.5-Flash",
14128
- api: "openai-completions",
14129
- provider: "zai",
14130
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14131
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14132
- reasoning: true,
14133
- input: ["text"],
14134
14732
  cost: {
14135
14733
  input: 0,
14136
14734
  output: 0,
@@ -14140,60 +14738,6 @@ export const MODELS = {
14140
14738
  contextWindow: 131072,
14141
14739
  maxTokens: 98304,
14142
14740
  },
14143
- "glm-4.5v": {
14144
- id: "glm-4.5v",
14145
- name: "GLM-4.5V",
14146
- api: "openai-completions",
14147
- provider: "zai",
14148
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14149
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
14150
- reasoning: true,
14151
- input: ["text", "image"],
14152
- cost: {
14153
- input: 0.6,
14154
- output: 1.8,
14155
- cacheRead: 0,
14156
- cacheWrite: 0,
14157
- },
14158
- contextWindow: 64000,
14159
- maxTokens: 16384,
14160
- },
14161
- "glm-4.6": {
14162
- id: "glm-4.6",
14163
- name: "GLM-4.6",
14164
- api: "openai-completions",
14165
- provider: "zai",
14166
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14167
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14168
- reasoning: true,
14169
- input: ["text"],
14170
- cost: {
14171
- input: 0.6,
14172
- output: 2.2,
14173
- cacheRead: 0.11,
14174
- cacheWrite: 0,
14175
- },
14176
- contextWindow: 204800,
14177
- maxTokens: 131072,
14178
- },
14179
- "glm-4.6v": {
14180
- id: "glm-4.6v",
14181
- name: "GLM-4.6V",
14182
- api: "openai-completions",
14183
- provider: "zai",
14184
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14185
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14186
- reasoning: true,
14187
- input: ["text", "image"],
14188
- cost: {
14189
- input: 0.3,
14190
- output: 0.9,
14191
- cacheRead: 0,
14192
- cacheWrite: 0,
14193
- },
14194
- contextWindow: 128000,
14195
- maxTokens: 32768,
14196
- },
14197
14741
  "glm-4.7": {
14198
14742
  id: "glm-4.7",
14199
14743
  name: "GLM-4.7",
@@ -14203,66 +14747,12 @@ export const MODELS = {
14203
14747
  compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14204
14748
  reasoning: true,
14205
14749
  input: ["text"],
14206
- cost: {
14207
- input: 0.6,
14208
- output: 2.2,
14209
- cacheRead: 0.11,
14210
- cacheWrite: 0,
14211
- },
14212
- contextWindow: 204800,
14213
- maxTokens: 131072,
14214
- },
14215
- "glm-4.7-flash": {
14216
- id: "glm-4.7-flash",
14217
- name: "GLM-4.7-Flash",
14218
- api: "openai-completions",
14219
- provider: "zai",
14220
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14221
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14222
- reasoning: true,
14223
- input: ["text"],
14224
14750
  cost: {
14225
14751
  input: 0,
14226
14752
  output: 0,
14227
14753
  cacheRead: 0,
14228
14754
  cacheWrite: 0,
14229
14755
  },
14230
- contextWindow: 200000,
14231
- maxTokens: 131072,
14232
- },
14233
- "glm-4.7-flashx": {
14234
- id: "glm-4.7-flashx",
14235
- name: "GLM-4.7-FlashX",
14236
- api: "openai-completions",
14237
- provider: "zai",
14238
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14239
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14240
- reasoning: true,
14241
- input: ["text"],
14242
- cost: {
14243
- input: 0.07,
14244
- output: 0.4,
14245
- cacheRead: 0.01,
14246
- cacheWrite: 0,
14247
- },
14248
- contextWindow: 200000,
14249
- maxTokens: 131072,
14250
- },
14251
- "glm-5": {
14252
- id: "glm-5",
14253
- name: "GLM-5",
14254
- api: "openai-completions",
14255
- provider: "zai",
14256
- baseUrl: "https://api.z.ai/api/coding/paas/v4",
14257
- compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14258
- reasoning: true,
14259
- input: ["text"],
14260
- cost: {
14261
- input: 1,
14262
- output: 3.2,
14263
- cacheRead: 0.2,
14264
- cacheWrite: 0,
14265
- },
14266
14756
  contextWindow: 204800,
14267
14757
  maxTokens: 131072,
14268
14758
  },
@@ -14276,27 +14766,27 @@ export const MODELS = {
14276
14766
  reasoning: true,
14277
14767
  input: ["text"],
14278
14768
  cost: {
14279
- input: 1.2,
14280
- output: 4,
14281
- cacheRead: 0.24,
14769
+ input: 0,
14770
+ output: 0,
14771
+ cacheRead: 0,
14282
14772
  cacheWrite: 0,
14283
14773
  },
14284
14774
  contextWindow: 200000,
14285
14775
  maxTokens: 131072,
14286
14776
  },
14287
- "glm-5v-turbo": {
14288
- id: "glm-5v-turbo",
14289
- name: "glm-5v-turbo",
14777
+ "glm-5.1": {
14778
+ id: "glm-5.1",
14779
+ name: "GLM-5.1",
14290
14780
  api: "openai-completions",
14291
14781
  provider: "zai",
14292
14782
  baseUrl: "https://api.z.ai/api/coding/paas/v4",
14293
14783
  compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14294
14784
  reasoning: true,
14295
- input: ["text", "image"],
14785
+ input: ["text"],
14296
14786
  cost: {
14297
- input: 1.2,
14298
- output: 4,
14299
- cacheRead: 0.24,
14787
+ input: 0,
14788
+ output: 0,
14789
+ cacheRead: 0,
14300
14790
  cacheWrite: 0,
14301
14791
  },
14302
14792
  contextWindow: 200000,