@earendil-works/pi-ai 0.77.0 → 0.78.1

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 (55) hide show
  1. package/README.md +9 -3
  2. package/dist/env-api-keys.d.ts.map +1 -1
  3. package/dist/env-api-keys.js +3 -0
  4. package/dist/env-api-keys.js.map +1 -1
  5. package/dist/image-models.generated.d.ts +15 -0
  6. package/dist/image-models.generated.d.ts.map +1 -1
  7. package/dist/image-models.generated.js +15 -0
  8. package/dist/image-models.generated.js.map +1 -1
  9. package/dist/models.generated.d.ts +1299 -341
  10. package/dist/models.generated.d.ts.map +1 -1
  11. package/dist/models.generated.js +1237 -545
  12. package/dist/models.generated.js.map +1 -1
  13. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  14. package/dist/providers/amazon-bedrock.js +82 -19
  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 +10 -7
  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 +5 -9
  21. package/dist/providers/azure-openai-responses.js.map +1 -1
  22. package/dist/providers/google-vertex.d.ts.map +1 -1
  23. package/dist/providers/google-vertex.js +1 -1
  24. package/dist/providers/google-vertex.js.map +1 -1
  25. package/dist/providers/google.d.ts.map +1 -1
  26. package/dist/providers/google.js +5 -3
  27. package/dist/providers/google.js.map +1 -1
  28. package/dist/providers/images/openrouter.d.ts.map +1 -1
  29. package/dist/providers/images/openrouter.js +2 -3
  30. package/dist/providers/images/openrouter.js.map +1 -1
  31. package/dist/providers/mistral.d.ts.map +1 -1
  32. package/dist/providers/mistral.js +2 -3
  33. package/dist/providers/mistral.js.map +1 -1
  34. package/dist/providers/openai-codex-responses.d.ts.map +1 -1
  35. package/dist/providers/openai-codex-responses.js +15 -5
  36. package/dist/providers/openai-codex-responses.js.map +1 -1
  37. package/dist/providers/openai-completions.d.ts.map +1 -1
  38. package/dist/providers/openai-completions.js +38 -21
  39. package/dist/providers/openai-completions.js.map +1 -1
  40. package/dist/providers/openai-responses.d.ts.map +1 -1
  41. package/dist/providers/openai-responses.js +5 -9
  42. package/dist/providers/openai-responses.js.map +1 -1
  43. package/dist/stream.d.ts.map +1 -1
  44. package/dist/stream.js +14 -2
  45. package/dist/stream.js.map +1 -1
  46. package/dist/types.d.ts +13 -5
  47. package/dist/types.d.ts.map +1 -1
  48. package/dist/types.js.map +1 -1
  49. package/dist/utils/oauth/github-copilot.d.ts.map +1 -1
  50. package/dist/utils/oauth/github-copilot.js +13 -1
  51. package/dist/utils/oauth/github-copilot.js.map +1 -1
  52. package/dist/utils/oauth/openai-codex.d.ts.map +1 -1
  53. package/dist/utils/oauth/openai-codex.js +4 -2
  54. package/dist/utils/oauth/openai-codex.js.map +1 -1
  55. package/package.json +1 -1
@@ -391,8 +391,8 @@ export const MODELS = {
391
391
  thinkingLevelMap: { "xhigh": "max" },
392
392
  input: ["text", "image"],
393
393
  cost: {
394
- input: 5,
395
- output: 25,
394
+ input: 5.5,
395
+ output: 27.5,
396
396
  cacheRead: 0.5,
397
397
  cacheWrite: 6.25,
398
398
  },
@@ -409,10 +409,10 @@ export const MODELS = {
409
409
  thinkingLevelMap: { "xhigh": "xhigh" },
410
410
  input: ["text", "image"],
411
411
  cost: {
412
- input: 5,
413
- output: 25,
414
- cacheRead: 0.5,
415
- cacheWrite: 6.25,
412
+ input: 5.5,
413
+ output: 27.5,
414
+ cacheRead: 0.55,
415
+ cacheWrite: 6.875,
416
416
  },
417
417
  contextWindow: 1000000,
418
418
  maxTokens: 128000,
@@ -427,10 +427,10 @@ export const MODELS = {
427
427
  thinkingLevelMap: { "xhigh": "xhigh" },
428
428
  input: ["text", "image"],
429
429
  cost: {
430
- input: 5,
431
- output: 25,
432
- cacheRead: 0.5,
433
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
434
434
  },
435
435
  contextWindow: 1000000,
436
436
  maxTokens: 128000,
@@ -444,10 +444,10 @@ export const MODELS = {
444
444
  reasoning: true,
445
445
  input: ["text", "image"],
446
446
  cost: {
447
- input: 3,
448
- output: 15,
449
- cacheRead: 0.3,
450
- cacheWrite: 3.75,
447
+ input: 3.3,
448
+ output: 16.5,
449
+ cacheRead: 0.33,
450
+ cacheWrite: 4.125,
451
451
  },
452
452
  contextWindow: 200000,
453
453
  maxTokens: 64000,
@@ -461,10 +461,10 @@ export const MODELS = {
461
461
  reasoning: true,
462
462
  input: ["text", "image"],
463
463
  cost: {
464
- input: 3,
465
- output: 15,
466
- cacheRead: 0.3,
467
- cacheWrite: 3.75,
464
+ input: 3.3,
465
+ output: 16.5,
466
+ cacheRead: 0.33,
467
+ cacheWrite: 4.125,
468
468
  },
469
469
  contextWindow: 1000000,
470
470
  maxTokens: 64000,
@@ -1549,6 +1549,63 @@ export const MODELS = {
1549
1549
  maxTokens: 101376,
1550
1550
  },
1551
1551
  },
1552
+ "ant-ling": {
1553
+ "Ling-2.6-1T": {
1554
+ id: "Ling-2.6-1T",
1555
+ name: "Ling 2.6 1T",
1556
+ api: "openai-completions",
1557
+ provider: "ant-ling",
1558
+ baseUrl: "https://api.ant-ling.com/v1",
1559
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsLongCacheRetention": false },
1560
+ reasoning: false,
1561
+ input: ["text"],
1562
+ cost: {
1563
+ input: 0.06,
1564
+ output: 0.25,
1565
+ cacheRead: 0,
1566
+ cacheWrite: 0,
1567
+ },
1568
+ contextWindow: 262144,
1569
+ maxTokens: 65536,
1570
+ },
1571
+ "Ling-2.6-flash": {
1572
+ id: "Ling-2.6-flash",
1573
+ name: "Ling 2.6 Flash",
1574
+ api: "openai-completions",
1575
+ provider: "ant-ling",
1576
+ baseUrl: "https://api.ant-ling.com/v1",
1577
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsLongCacheRetention": false },
1578
+ reasoning: false,
1579
+ input: ["text"],
1580
+ cost: {
1581
+ input: 0.01,
1582
+ output: 0.02,
1583
+ cacheRead: 0,
1584
+ cacheWrite: 0,
1585
+ },
1586
+ contextWindow: 262144,
1587
+ maxTokens: 65536,
1588
+ },
1589
+ "Ring-2.6-1T": {
1590
+ id: "Ring-2.6-1T",
1591
+ name: "Ring 2.6 1T",
1592
+ api: "openai-completions",
1593
+ provider: "ant-ling",
1594
+ baseUrl: "https://api.ant-ling.com/v1",
1595
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsLongCacheRetention": false, "thinkingFormat": "ant-ling" },
1596
+ reasoning: true,
1597
+ thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
1598
+ input: ["text"],
1599
+ cost: {
1600
+ input: 0.06,
1601
+ output: 0.25,
1602
+ cacheRead: 0,
1603
+ cacheWrite: 0,
1604
+ },
1605
+ contextWindow: 262144,
1606
+ maxTokens: 65536,
1607
+ },
1608
+ },
1552
1609
  "anthropic": {
1553
1610
  "claude-3-5-haiku-20241022": {
1554
1611
  id: "claude-3-5-haiku-20241022",
@@ -1847,7 +1904,7 @@ export const MODELS = {
1847
1904
  api: "anthropic-messages",
1848
1905
  provider: "anthropic",
1849
1906
  baseUrl: "https://api.anthropic.com",
1850
- compat: { "forceAdaptiveThinking": true },
1907
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
1851
1908
  reasoning: true,
1852
1909
  thinkingLevelMap: { "xhigh": "xhigh" },
1853
1910
  input: ["text", "image"],
@@ -1866,7 +1923,7 @@ export const MODELS = {
1866
1923
  api: "anthropic-messages",
1867
1924
  provider: "anthropic",
1868
1925
  baseUrl: "https://api.anthropic.com",
1869
- compat: { "forceAdaptiveThinking": true },
1926
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
1870
1927
  reasoning: true,
1871
1928
  thinkingLevelMap: { "xhigh": "xhigh" },
1872
1929
  input: ["text", "image"],
@@ -2955,7 +3012,26 @@ export const MODELS = {
2955
3012
  api: "anthropic-messages",
2956
3013
  provider: "cloudflare-ai-gateway",
2957
3014
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
2958
- compat: { "forceAdaptiveThinking": true },
3015
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3016
+ reasoning: true,
3017
+ thinkingLevelMap: { "xhigh": "xhigh" },
3018
+ input: ["text", "image"],
3019
+ cost: {
3020
+ input: 5,
3021
+ output: 25,
3022
+ cacheRead: 0.5,
3023
+ cacheWrite: 6.25,
3024
+ },
3025
+ contextWindow: 1000000,
3026
+ maxTokens: 128000,
3027
+ },
3028
+ "claude-opus-4-8": {
3029
+ id: "claude-opus-4-8",
3030
+ name: "Claude Opus 4.8",
3031
+ api: "anthropic-messages",
3032
+ provider: "cloudflare-ai-gateway",
3033
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3034
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
2959
3035
  reasoning: true,
2960
3036
  thinkingLevelMap: { "xhigh": "xhigh" },
2961
3037
  input: ["text", "image"],
@@ -3463,24 +3539,6 @@ export const MODELS = {
3463
3539
  contextWindow: 128000,
3464
3540
  maxTokens: 128000,
3465
3541
  },
3466
- "@cf/moonshotai/kimi-k2.5": {
3467
- id: "@cf/moonshotai/kimi-k2.5",
3468
- name: "Kimi K2.5",
3469
- api: "openai-completions",
3470
- provider: "cloudflare-workers-ai",
3471
- baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3472
- compat: { "sendSessionAffinityHeaders": true },
3473
- reasoning: true,
3474
- input: ["text", "image"],
3475
- cost: {
3476
- input: 0.6,
3477
- output: 3,
3478
- cacheRead: 0.1,
3479
- cacheWrite: 0,
3480
- },
3481
- contextWindow: 256000,
3482
- maxTokens: 256000,
3483
- },
3484
3542
  "@cf/moonshotai/kimi-k2.6": {
3485
3543
  id: "@cf/moonshotai/kimi-k2.6",
3486
3544
  name: "Kimi K2.6",
@@ -3808,8 +3866,8 @@ export const MODELS = {
3808
3866
  cacheRead: 0.1,
3809
3867
  cacheWrite: 0,
3810
3868
  },
3811
- contextWindow: 128000,
3812
- maxTokens: 8192,
3869
+ contextWindow: 262144,
3870
+ maxTokens: 65536,
3813
3871
  },
3814
3872
  "accounts/fireworks/routers/glm-5p1-fast": {
3815
3873
  id: "accounts/fireworks/routers/glm-5p1-fast",
@@ -3851,7 +3909,7 @@ export const MODELS = {
3851
3909
  "github-copilot": {
3852
3910
  "claude-haiku-4.5": {
3853
3911
  id: "claude-haiku-4.5",
3854
- name: "Claude Haiku 4.5",
3912
+ name: "Claude Haiku 4.5 (latest)",
3855
3913
  api: "anthropic-messages",
3856
3914
  provider: "github-copilot",
3857
3915
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3860,17 +3918,17 @@ export const MODELS = {
3860
3918
  reasoning: true,
3861
3919
  input: ["text", "image"],
3862
3920
  cost: {
3863
- input: 0,
3864
- output: 0,
3865
- cacheRead: 0,
3866
- cacheWrite: 0,
3921
+ input: 1,
3922
+ output: 5,
3923
+ cacheRead: 0.1,
3924
+ cacheWrite: 1.25,
3867
3925
  },
3868
- contextWindow: 144000,
3869
- maxTokens: 32000,
3926
+ contextWindow: 200000,
3927
+ maxTokens: 64000,
3870
3928
  },
3871
3929
  "claude-opus-4.5": {
3872
3930
  id: "claude-opus-4.5",
3873
- name: "Claude Opus 4.5",
3931
+ name: "Claude Opus 4.5 (latest)",
3874
3932
  api: "anthropic-messages",
3875
3933
  provider: "github-copilot",
3876
3934
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3878,12 +3936,12 @@ export const MODELS = {
3878
3936
  reasoning: true,
3879
3937
  input: ["text", "image"],
3880
3938
  cost: {
3881
- input: 0,
3882
- output: 0,
3883
- cacheRead: 0,
3884
- cacheWrite: 0,
3939
+ input: 5,
3940
+ output: 25,
3941
+ cacheRead: 0.5,
3942
+ cacheWrite: 6.25,
3885
3943
  },
3886
- contextWindow: 160000,
3944
+ contextWindow: 200000,
3887
3945
  maxTokens: 32000,
3888
3946
  },
3889
3947
  "claude-opus-4.6": {
@@ -3898,13 +3956,13 @@ export const MODELS = {
3898
3956
  thinkingLevelMap: { "xhigh": "max" },
3899
3957
  input: ["text", "image"],
3900
3958
  cost: {
3901
- input: 0,
3902
- output: 0,
3903
- cacheRead: 0,
3904
- cacheWrite: 0,
3959
+ input: 5,
3960
+ output: 25,
3961
+ cacheRead: 0.5,
3962
+ cacheWrite: 6.25,
3905
3963
  },
3906
3964
  contextWindow: 1000000,
3907
- maxTokens: 64000,
3965
+ maxTokens: 32000,
3908
3966
  },
3909
3967
  "claude-opus-4.7": {
3910
3968
  id: "claude-opus-4.7",
@@ -3913,22 +3971,61 @@ export const MODELS = {
3913
3971
  provider: "github-copilot",
3914
3972
  baseUrl: "https://api.individual.githubcopilot.com",
3915
3973
  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" },
3916
- compat: { "forceAdaptiveThinking": true },
3974
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3917
3975
  reasoning: true,
3918
3976
  thinkingLevelMap: { "xhigh": "xhigh" },
3919
3977
  input: ["text", "image"],
3920
3978
  cost: {
3921
- input: 0,
3922
- output: 0,
3923
- cacheRead: 0,
3924
- cacheWrite: 0,
3979
+ input: 5,
3980
+ output: 25,
3981
+ cacheRead: 0.5,
3982
+ cacheWrite: 6.25,
3983
+ },
3984
+ contextWindow: 200000,
3985
+ maxTokens: 32000,
3986
+ },
3987
+ "claude-opus-4.8": {
3988
+ id: "claude-opus-4.8",
3989
+ name: "Claude Opus 4.8",
3990
+ api: "anthropic-messages",
3991
+ provider: "github-copilot",
3992
+ baseUrl: "https://api.individual.githubcopilot.com",
3993
+ 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" },
3994
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
3995
+ reasoning: true,
3996
+ thinkingLevelMap: { "xhigh": "xhigh" },
3997
+ input: ["text", "image"],
3998
+ cost: {
3999
+ input: 5,
4000
+ output: 25,
4001
+ cacheRead: 0.5,
4002
+ cacheWrite: 6.25,
3925
4003
  },
3926
- contextWindow: 144000,
4004
+ contextWindow: 200000,
3927
4005
  maxTokens: 64000,
3928
4006
  },
4007
+ "claude-sonnet-4": {
4008
+ id: "claude-sonnet-4",
4009
+ name: "Claude Sonnet 4 (latest)",
4010
+ api: "anthropic-messages",
4011
+ provider: "github-copilot",
4012
+ baseUrl: "https://api.individual.githubcopilot.com",
4013
+ 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" },
4014
+ compat: { "supportsEagerToolInputStreaming": false },
4015
+ reasoning: true,
4016
+ input: ["text", "image"],
4017
+ cost: {
4018
+ input: 3,
4019
+ output: 15,
4020
+ cacheRead: 0.3,
4021
+ cacheWrite: 3.75,
4022
+ },
4023
+ contextWindow: 216000,
4024
+ maxTokens: 16000,
4025
+ },
3929
4026
  "claude-sonnet-4.5": {
3930
4027
  id: "claude-sonnet-4.5",
3931
- name: "Claude Sonnet 4.5",
4028
+ name: "Claude Sonnet 4.5 (latest)",
3932
4029
  api: "anthropic-messages",
3933
4030
  provider: "github-copilot",
3934
4031
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3937,12 +4034,12 @@ export const MODELS = {
3937
4034
  reasoning: true,
3938
4035
  input: ["text", "image"],
3939
4036
  cost: {
3940
- input: 0,
3941
- output: 0,
3942
- cacheRead: 0,
3943
- cacheWrite: 0,
4037
+ input: 3,
4038
+ output: 15,
4039
+ cacheRead: 0.3,
4040
+ cacheWrite: 3.75,
3944
4041
  },
3945
- contextWindow: 144000,
4042
+ contextWindow: 200000,
3946
4043
  maxTokens: 32000,
3947
4044
  },
3948
4045
  "claude-sonnet-4.6": {
@@ -3956,10 +4053,10 @@ export const MODELS = {
3956
4053
  reasoning: true,
3957
4054
  input: ["text", "image"],
3958
4055
  cost: {
3959
- input: 0,
3960
- output: 0,
3961
- cacheRead: 0,
3962
- cacheWrite: 0,
4056
+ input: 3,
4057
+ output: 15,
4058
+ cacheRead: 0.3,
4059
+ cacheWrite: 3.75,
3963
4060
  },
3964
4061
  contextWindow: 1000000,
3965
4062
  maxTokens: 32000,
@@ -3972,12 +4069,12 @@ export const MODELS = {
3972
4069
  baseUrl: "https://api.individual.githubcopilot.com",
3973
4070
  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" },
3974
4071
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3975
- reasoning: false,
4072
+ reasoning: true,
3976
4073
  input: ["text", "image"],
3977
4074
  cost: {
3978
- input: 0,
3979
- output: 0,
3980
- cacheRead: 0,
4075
+ input: 1.25,
4076
+ output: 10,
4077
+ cacheRead: 0.125,
3981
4078
  cacheWrite: 0,
3982
4079
  },
3983
4080
  contextWindow: 128000,
@@ -3985,7 +4082,7 @@ export const MODELS = {
3985
4082
  },
3986
4083
  "gemini-3-flash-preview": {
3987
4084
  id: "gemini-3-flash-preview",
3988
- name: "Gemini 3 Flash",
4085
+ name: "Gemini 3 Flash Preview",
3989
4086
  api: "openai-completions",
3990
4087
  provider: "github-copilot",
3991
4088
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3994,9 +4091,9 @@ export const MODELS = {
3994
4091
  reasoning: true,
3995
4092
  input: ["text", "image"],
3996
4093
  cost: {
3997
- input: 0,
3998
- output: 0,
3999
- cacheRead: 0,
4094
+ input: 0.5,
4095
+ output: 3,
4096
+ cacheRead: 0.05,
4000
4097
  cacheWrite: 0,
4001
4098
  },
4002
4099
  contextWindow: 128000,
@@ -4013,12 +4110,12 @@ export const MODELS = {
4013
4110
  reasoning: true,
4014
4111
  input: ["text", "image"],
4015
4112
  cost: {
4016
- input: 0,
4017
- output: 0,
4018
- cacheRead: 0,
4113
+ input: 2,
4114
+ output: 12,
4115
+ cacheRead: 0.2,
4019
4116
  cacheWrite: 0,
4020
4117
  },
4021
- contextWindow: 128000,
4118
+ contextWindow: 200000,
4022
4119
  maxTokens: 64000,
4023
4120
  },
4024
4121
  "gemini-3.5-flash": {
@@ -4032,12 +4129,12 @@ export const MODELS = {
4032
4129
  reasoning: true,
4033
4130
  input: ["text", "image"],
4034
4131
  cost: {
4035
- input: 0,
4036
- output: 0,
4037
- cacheRead: 0,
4132
+ input: 1.5,
4133
+ output: 9,
4134
+ cacheRead: 0.15,
4038
4135
  cacheWrite: 0,
4039
4136
  },
4040
- contextWindow: 128000,
4137
+ contextWindow: 200000,
4041
4138
  maxTokens: 64000,
4042
4139
  },
4043
4140
  "gpt-4.1": {
@@ -4051,47 +4148,28 @@ export const MODELS = {
4051
4148
  reasoning: false,
4052
4149
  input: ["text", "image"],
4053
4150
  cost: {
4054
- input: 0,
4055
- output: 0,
4056
- cacheRead: 0,
4151
+ input: 2,
4152
+ output: 8,
4153
+ cacheRead: 0.5,
4057
4154
  cacheWrite: 0,
4058
4155
  },
4059
4156
  contextWindow: 128000,
4060
4157
  maxTokens: 16384,
4061
4158
  },
4062
- "gpt-4o": {
4063
- id: "gpt-4o",
4064
- name: "GPT-4o",
4065
- api: "openai-completions",
4159
+ "gpt-5-mini": {
4160
+ id: "gpt-5-mini",
4161
+ name: "GPT-5 Mini",
4162
+ api: "openai-responses",
4066
4163
  provider: "github-copilot",
4067
4164
  baseUrl: "https://api.individual.githubcopilot.com",
4068
4165
  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" },
4069
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4070
- reasoning: false,
4166
+ reasoning: true,
4167
+ thinkingLevelMap: { "off": null, "minimal": "low" },
4071
4168
  input: ["text", "image"],
4072
4169
  cost: {
4073
- input: 0,
4074
- output: 0,
4075
- cacheRead: 0,
4076
- cacheWrite: 0,
4077
- },
4078
- contextWindow: 128000,
4079
- maxTokens: 4096,
4080
- },
4081
- "gpt-5-mini": {
4082
- id: "gpt-5-mini",
4083
- name: "GPT-5-mini",
4084
- api: "openai-responses",
4085
- provider: "github-copilot",
4086
- baseUrl: "https://api.individual.githubcopilot.com",
4087
- 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" },
4088
- reasoning: true,
4089
- thinkingLevelMap: { "off": null, "minimal": "low" },
4090
- input: ["text", "image"],
4091
- cost: {
4092
- input: 0,
4093
- output: 0,
4094
- cacheRead: 0,
4170
+ input: 0.25,
4171
+ output: 2,
4172
+ cacheRead: 0.025,
4095
4173
  cacheWrite: 0,
4096
4174
  },
4097
4175
  contextWindow: 264000,
@@ -4108,17 +4186,17 @@ export const MODELS = {
4108
4186
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4109
4187
  input: ["text", "image"],
4110
4188
  cost: {
4111
- input: 0,
4112
- output: 0,
4113
- cacheRead: 0,
4189
+ input: 1.75,
4190
+ output: 14,
4191
+ cacheRead: 0.175,
4114
4192
  cacheWrite: 0,
4115
4193
  },
4116
- contextWindow: 264000,
4117
- maxTokens: 64000,
4194
+ contextWindow: 400000,
4195
+ maxTokens: 128000,
4118
4196
  },
4119
4197
  "gpt-5.2-codex": {
4120
4198
  id: "gpt-5.2-codex",
4121
- name: "GPT-5.2-Codex",
4199
+ name: "GPT-5.2 Codex",
4122
4200
  api: "openai-responses",
4123
4201
  provider: "github-copilot",
4124
4202
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4127,9 +4205,9 @@ export const MODELS = {
4127
4205
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4128
4206
  input: ["text", "image"],
4129
4207
  cost: {
4130
- input: 0,
4131
- output: 0,
4132
- cacheRead: 0,
4208
+ input: 1.75,
4209
+ output: 14,
4210
+ cacheRead: 0.175,
4133
4211
  cacheWrite: 0,
4134
4212
  },
4135
4213
  contextWindow: 400000,
@@ -4137,7 +4215,7 @@ export const MODELS = {
4137
4215
  },
4138
4216
  "gpt-5.3-codex": {
4139
4217
  id: "gpt-5.3-codex",
4140
- name: "GPT-5.3-Codex",
4218
+ name: "GPT-5.3 Codex",
4141
4219
  api: "openai-responses",
4142
4220
  provider: "github-copilot",
4143
4221
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4146,9 +4224,9 @@ export const MODELS = {
4146
4224
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4147
4225
  input: ["text", "image"],
4148
4226
  cost: {
4149
- input: 0,
4150
- output: 0,
4151
- cacheRead: 0,
4227
+ input: 1.75,
4228
+ output: 14,
4229
+ cacheRead: 0.175,
4152
4230
  cacheWrite: 0,
4153
4231
  },
4154
4232
  contextWindow: 400000,
@@ -4165,9 +4243,9 @@ export const MODELS = {
4165
4243
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4166
4244
  input: ["text", "image"],
4167
4245
  cost: {
4168
- input: 0,
4169
- output: 0,
4170
- cacheRead: 0,
4246
+ input: 2.5,
4247
+ output: 15,
4248
+ cacheRead: 0.25,
4171
4249
  cacheWrite: 0,
4172
4250
  },
4173
4251
  contextWindow: 400000,
@@ -4175,7 +4253,7 @@ export const MODELS = {
4175
4253
  },
4176
4254
  "gpt-5.4-mini": {
4177
4255
  id: "gpt-5.4-mini",
4178
- name: "GPT-5.4 Mini",
4256
+ name: "GPT-5.4 mini",
4179
4257
  api: "openai-responses",
4180
4258
  provider: "github-copilot",
4181
4259
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4184,9 +4262,28 @@ export const MODELS = {
4184
4262
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4185
4263
  input: ["text", "image"],
4186
4264
  cost: {
4187
- input: 0,
4188
- output: 0,
4189
- cacheRead: 0,
4265
+ input: 0.75,
4266
+ output: 4.5,
4267
+ cacheRead: 0.075,
4268
+ cacheWrite: 0,
4269
+ },
4270
+ contextWindow: 400000,
4271
+ maxTokens: 128000,
4272
+ },
4273
+ "gpt-5.4-nano": {
4274
+ id: "gpt-5.4-nano",
4275
+ name: "GPT-5.4 nano",
4276
+ api: "openai-responses",
4277
+ provider: "github-copilot",
4278
+ baseUrl: "https://api.individual.githubcopilot.com",
4279
+ 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" },
4280
+ reasoning: true,
4281
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4282
+ input: ["text", "image"],
4283
+ cost: {
4284
+ input: 0.2,
4285
+ output: 1.25,
4286
+ cacheRead: 0.02,
4190
4287
  cacheWrite: 0,
4191
4288
  },
4192
4289
  contextWindow: 400000,
@@ -4203,32 +4300,32 @@ export const MODELS = {
4203
4300
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4204
4301
  input: ["text", "image"],
4205
4302
  cost: {
4206
- input: 0,
4207
- output: 0,
4208
- cacheRead: 0,
4303
+ input: 5,
4304
+ output: 30,
4305
+ cacheRead: 0.5,
4209
4306
  cacheWrite: 0,
4210
4307
  },
4211
4308
  contextWindow: 400000,
4212
4309
  maxTokens: 128000,
4213
4310
  },
4214
- "grok-code-fast-1": {
4215
- id: "grok-code-fast-1",
4216
- name: "Grok Code Fast 1",
4311
+ "raptor-mini": {
4312
+ id: "raptor-mini",
4313
+ name: "Raptor mini",
4217
4314
  api: "openai-completions",
4218
4315
  provider: "github-copilot",
4219
4316
  baseUrl: "https://api.individual.githubcopilot.com",
4220
4317
  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" },
4221
4318
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4222
4319
  reasoning: true,
4223
- input: ["text"],
4320
+ input: ["text", "image"],
4224
4321
  cost: {
4225
- input: 0,
4226
- output: 0,
4227
- cacheRead: 0,
4322
+ input: 0.25,
4323
+ output: 2,
4324
+ cacheRead: 0.025,
4228
4325
  cacheWrite: 0,
4229
4326
  },
4230
- contextWindow: 128000,
4231
- maxTokens: 64000,
4327
+ contextWindow: 400000,
4328
+ maxTokens: 128000,
4232
4329
  },
4233
4330
  },
4234
4331
  "google": {
@@ -5521,6 +5618,23 @@ export const MODELS = {
5521
5618
  contextWindow: 204800,
5522
5619
  maxTokens: 131072,
5523
5620
  },
5621
+ "MiniMax-M3": {
5622
+ id: "MiniMax-M3",
5623
+ name: "MiniMax-M3",
5624
+ api: "anthropic-messages",
5625
+ provider: "minimax",
5626
+ baseUrl: "https://api.minimax.io/anthropic",
5627
+ reasoning: true,
5628
+ input: ["text", "image"],
5629
+ cost: {
5630
+ input: 0.6,
5631
+ output: 2.4,
5632
+ cacheRead: 0.12,
5633
+ cacheWrite: 0,
5634
+ },
5635
+ contextWindow: 512000,
5636
+ maxTokens: 128000,
5637
+ },
5524
5638
  },
5525
5639
  "minimax-cn": {
5526
5640
  "MiniMax-M2.7": {
@@ -5557,6 +5671,23 @@ export const MODELS = {
5557
5671
  contextWindow: 204800,
5558
5672
  maxTokens: 131072,
5559
5673
  },
5674
+ "MiniMax-M3": {
5675
+ id: "MiniMax-M3",
5676
+ name: "MiniMax-M3",
5677
+ api: "anthropic-messages",
5678
+ provider: "minimax-cn",
5679
+ baseUrl: "https://api.minimaxi.com/anthropic",
5680
+ reasoning: true,
5681
+ input: ["text", "image"],
5682
+ cost: {
5683
+ input: 0.6,
5684
+ output: 2.4,
5685
+ cacheRead: 0.12,
5686
+ cacheWrite: 0,
5687
+ },
5688
+ contextWindow: 512000,
5689
+ maxTokens: 128000,
5690
+ },
5560
5691
  },
5561
5692
  "mistral": {
5562
5693
  "codestral-latest": {
@@ -5593,6 +5724,23 @@ export const MODELS = {
5593
5724
  contextWindow: 262144,
5594
5725
  maxTokens: 262144,
5595
5726
  },
5727
+ "devstral-latest": {
5728
+ id: "devstral-latest",
5729
+ name: "Devstral 2",
5730
+ api: "mistral-conversations",
5731
+ provider: "mistral",
5732
+ baseUrl: "https://api.mistral.ai",
5733
+ reasoning: false,
5734
+ input: ["text"],
5735
+ cost: {
5736
+ input: 0.4,
5737
+ output: 2,
5738
+ cacheRead: 0,
5739
+ cacheWrite: 0,
5740
+ },
5741
+ contextWindow: 262144,
5742
+ maxTokens: 262144,
5743
+ },
5596
5744
  "devstral-medium-2507": {
5597
5745
  id: "devstral-medium-2507",
5598
5746
  name: "Devstral Medium",
@@ -5967,6 +6115,23 @@ export const MODELS = {
5967
6115
  contextWindow: 8000,
5968
6116
  maxTokens: 8000,
5969
6117
  },
6118
+ "open-mistral-nemo": {
6119
+ id: "open-mistral-nemo",
6120
+ name: "Open Mistral Nemo",
6121
+ api: "mistral-conversations",
6122
+ provider: "mistral",
6123
+ baseUrl: "https://api.mistral.ai",
6124
+ reasoning: false,
6125
+ input: ["text"],
6126
+ cost: {
6127
+ input: 0.15,
6128
+ output: 0.15,
6129
+ cacheRead: 0,
6130
+ cacheWrite: 0,
6131
+ },
6132
+ contextWindow: 128000,
6133
+ maxTokens: 128000,
6134
+ },
5970
6135
  "open-mixtral-8x22b": {
5971
6136
  id: "open-mixtral-8x22b",
5972
6137
  name: "Mixtral 8x22B",
@@ -6255,41 +6420,423 @@ export const MODELS = {
6255
6420
  contextWindow: 262144,
6256
6421
  maxTokens: 262144,
6257
6422
  },
6258
- "kimi-k2.5": {
6259
- id: "kimi-k2.5",
6260
- name: "Kimi K2.5",
6423
+ "kimi-k2.5": {
6424
+ id: "kimi-k2.5",
6425
+ name: "Kimi K2.5",
6426
+ api: "openai-completions",
6427
+ provider: "moonshotai-cn",
6428
+ baseUrl: "https://api.moonshot.cn/v1",
6429
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6430
+ reasoning: true,
6431
+ input: ["text", "image"],
6432
+ cost: {
6433
+ input: 0.6,
6434
+ output: 3,
6435
+ cacheRead: 0.1,
6436
+ cacheWrite: 0,
6437
+ },
6438
+ contextWindow: 262144,
6439
+ maxTokens: 262144,
6440
+ },
6441
+ "kimi-k2.6": {
6442
+ id: "kimi-k2.6",
6443
+ name: "Kimi K2.6",
6444
+ api: "openai-completions",
6445
+ provider: "moonshotai-cn",
6446
+ baseUrl: "https://api.moonshot.cn/v1",
6447
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6448
+ reasoning: true,
6449
+ input: ["text", "image"],
6450
+ cost: {
6451
+ input: 0.95,
6452
+ output: 4,
6453
+ cacheRead: 0.16,
6454
+ cacheWrite: 0,
6455
+ },
6456
+ contextWindow: 262144,
6457
+ maxTokens: 262144,
6458
+ },
6459
+ },
6460
+ "nvidia": {
6461
+ "meta/llama-3.1-70b-instruct": {
6462
+ id: "meta/llama-3.1-70b-instruct",
6463
+ name: "Llama 3.1 70b Instruct",
6464
+ api: "openai-completions",
6465
+ provider: "nvidia",
6466
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6467
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6468
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6469
+ reasoning: false,
6470
+ input: ["text"],
6471
+ cost: {
6472
+ input: 0,
6473
+ output: 0,
6474
+ cacheRead: 0,
6475
+ cacheWrite: 0,
6476
+ },
6477
+ contextWindow: 128000,
6478
+ maxTokens: 4096,
6479
+ },
6480
+ "meta/llama-3.1-8b-instruct": {
6481
+ id: "meta/llama-3.1-8b-instruct",
6482
+ name: "Llama 3.1 8B Instruct",
6483
+ api: "openai-completions",
6484
+ provider: "nvidia",
6485
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6486
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6487
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6488
+ reasoning: false,
6489
+ input: ["text"],
6490
+ cost: {
6491
+ input: 0,
6492
+ output: 0,
6493
+ cacheRead: 0,
6494
+ cacheWrite: 0,
6495
+ },
6496
+ contextWindow: 16000,
6497
+ maxTokens: 4096,
6498
+ },
6499
+ "meta/llama-3.2-11b-vision-instruct": {
6500
+ id: "meta/llama-3.2-11b-vision-instruct",
6501
+ name: "Llama 3.2 11b Vision Instruct",
6502
+ api: "openai-completions",
6503
+ provider: "nvidia",
6504
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6505
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6506
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6507
+ reasoning: false,
6508
+ input: ["text", "image"],
6509
+ cost: {
6510
+ input: 0,
6511
+ output: 0,
6512
+ cacheRead: 0,
6513
+ cacheWrite: 0,
6514
+ },
6515
+ contextWindow: 128000,
6516
+ maxTokens: 4096,
6517
+ },
6518
+ "meta/llama-3.2-90b-vision-instruct": {
6519
+ id: "meta/llama-3.2-90b-vision-instruct",
6520
+ name: "Llama-3.2-90B-Vision-Instruct",
6521
+ api: "openai-completions",
6522
+ provider: "nvidia",
6523
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6524
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6525
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6526
+ reasoning: false,
6527
+ input: ["text", "image"],
6528
+ cost: {
6529
+ input: 0,
6530
+ output: 0,
6531
+ cacheRead: 0,
6532
+ cacheWrite: 0,
6533
+ },
6534
+ contextWindow: 128000,
6535
+ maxTokens: 8192,
6536
+ },
6537
+ "meta/llama-3.3-70b-instruct": {
6538
+ id: "meta/llama-3.3-70b-instruct",
6539
+ name: "Llama 3.3 70b Instruct",
6540
+ api: "openai-completions",
6541
+ provider: "nvidia",
6542
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6543
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6544
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6545
+ reasoning: false,
6546
+ input: ["text"],
6547
+ cost: {
6548
+ input: 0,
6549
+ output: 0,
6550
+ cacheRead: 0,
6551
+ cacheWrite: 0,
6552
+ },
6553
+ contextWindow: 128000,
6554
+ maxTokens: 4096,
6555
+ },
6556
+ "mistralai/mistral-large-3-675b-instruct-2512": {
6557
+ id: "mistralai/mistral-large-3-675b-instruct-2512",
6558
+ name: "Mistral Large 3 675B Instruct 2512",
6559
+ api: "openai-completions",
6560
+ provider: "nvidia",
6561
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6562
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6563
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6564
+ reasoning: false,
6565
+ input: ["text", "image"],
6566
+ cost: {
6567
+ input: 0,
6568
+ output: 0,
6569
+ cacheRead: 0,
6570
+ cacheWrite: 0,
6571
+ },
6572
+ contextWindow: 262144,
6573
+ maxTokens: 262144,
6574
+ },
6575
+ "mistralai/mistral-small-4-119b-2603": {
6576
+ id: "mistralai/mistral-small-4-119b-2603",
6577
+ name: "mistral-small-4-119b-2603",
6578
+ api: "openai-completions",
6579
+ provider: "nvidia",
6580
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6581
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6582
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6583
+ reasoning: false,
6584
+ input: ["text"],
6585
+ cost: {
6586
+ input: 0,
6587
+ output: 0,
6588
+ cacheRead: 0,
6589
+ cacheWrite: 0,
6590
+ },
6591
+ contextWindow: 128000,
6592
+ maxTokens: 8192,
6593
+ },
6594
+ "moonshotai/kimi-k2.6": {
6595
+ id: "moonshotai/kimi-k2.6",
6596
+ name: "Kimi K2.6",
6597
+ api: "openai-completions",
6598
+ provider: "nvidia",
6599
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6600
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6601
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6602
+ reasoning: true,
6603
+ input: ["text", "image"],
6604
+ cost: {
6605
+ input: 0,
6606
+ output: 0,
6607
+ cacheRead: 0,
6608
+ cacheWrite: 0,
6609
+ },
6610
+ contextWindow: 262144,
6611
+ maxTokens: 262144,
6612
+ },
6613
+ "nvidia/llama-3.3-nemotron-super-49b-v1": {
6614
+ id: "nvidia/llama-3.3-nemotron-super-49b-v1",
6615
+ name: "Llama 3.3 Nemotron Super 49B v1",
6616
+ api: "openai-completions",
6617
+ provider: "nvidia",
6618
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6619
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6620
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6621
+ reasoning: true,
6622
+ input: ["text"],
6623
+ cost: {
6624
+ input: 0,
6625
+ output: 0,
6626
+ cacheRead: 0,
6627
+ cacheWrite: 0,
6628
+ },
6629
+ contextWindow: 131072,
6630
+ maxTokens: 131072,
6631
+ },
6632
+ "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
6633
+ id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
6634
+ name: "Llama 3.3 Nemotron Super 49B v1.5",
6635
+ api: "openai-completions",
6636
+ provider: "nvidia",
6637
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6638
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6639
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6640
+ reasoning: true,
6641
+ input: ["text"],
6642
+ cost: {
6643
+ input: 0,
6644
+ output: 0,
6645
+ cacheRead: 0,
6646
+ cacheWrite: 0,
6647
+ },
6648
+ contextWindow: 131072,
6649
+ maxTokens: 131072,
6650
+ },
6651
+ "nvidia/nemotron-3-nano-30b-a3b": {
6652
+ id: "nvidia/nemotron-3-nano-30b-a3b",
6653
+ name: "nemotron-3-nano-30b-a3b",
6654
+ api: "openai-completions",
6655
+ provider: "nvidia",
6656
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6657
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6658
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6659
+ reasoning: true,
6660
+ input: ["text"],
6661
+ cost: {
6662
+ input: 0,
6663
+ output: 0,
6664
+ cacheRead: 0,
6665
+ cacheWrite: 0,
6666
+ },
6667
+ contextWindow: 131072,
6668
+ maxTokens: 131072,
6669
+ },
6670
+ "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": {
6671
+ id: "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning",
6672
+ name: "Nemotron 3 Nano Omni",
6673
+ api: "openai-completions",
6674
+ provider: "nvidia",
6675
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6676
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6677
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6678
+ reasoning: true,
6679
+ input: ["text", "image"],
6680
+ cost: {
6681
+ input: 0,
6682
+ output: 0,
6683
+ cacheRead: 0,
6684
+ cacheWrite: 0,
6685
+ },
6686
+ contextWindow: 256000,
6687
+ maxTokens: 65536,
6688
+ },
6689
+ "nvidia/nemotron-3-super-120b-a12b": {
6690
+ id: "nvidia/nemotron-3-super-120b-a12b",
6691
+ name: "Nemotron 3 Super",
6692
+ api: "openai-completions",
6693
+ provider: "nvidia",
6694
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6695
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6696
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6697
+ reasoning: true,
6698
+ input: ["text"],
6699
+ cost: {
6700
+ input: 0.2,
6701
+ output: 0.8,
6702
+ cacheRead: 0,
6703
+ cacheWrite: 0,
6704
+ },
6705
+ contextWindow: 262144,
6706
+ maxTokens: 262144,
6707
+ },
6708
+ "nvidia/nvidia-nemotron-nano-9b-v2": {
6709
+ id: "nvidia/nvidia-nemotron-nano-9b-v2",
6710
+ name: "nvidia-nemotron-nano-9b-v2",
6711
+ api: "openai-completions",
6712
+ provider: "nvidia",
6713
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6714
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6715
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6716
+ reasoning: true,
6717
+ input: ["text"],
6718
+ cost: {
6719
+ input: 0,
6720
+ output: 0,
6721
+ cacheRead: 0,
6722
+ cacheWrite: 0,
6723
+ },
6724
+ contextWindow: 131072,
6725
+ maxTokens: 131072,
6726
+ },
6727
+ "openai/gpt-oss-20b": {
6728
+ id: "openai/gpt-oss-20b",
6729
+ name: "GPT OSS 20B",
6730
+ api: "openai-completions",
6731
+ provider: "nvidia",
6732
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6733
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6734
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6735
+ reasoning: true,
6736
+ input: ["text"],
6737
+ cost: {
6738
+ input: 0,
6739
+ output: 0,
6740
+ cacheRead: 0,
6741
+ cacheWrite: 0,
6742
+ },
6743
+ contextWindow: 131072,
6744
+ maxTokens: 32768,
6745
+ },
6746
+ "qwen/qwen3-coder-480b-a35b-instruct": {
6747
+ id: "qwen/qwen3-coder-480b-a35b-instruct",
6748
+ name: "Qwen3 Coder 480B A35B Instruct",
6749
+ api: "openai-completions",
6750
+ provider: "nvidia",
6751
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6752
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6753
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6754
+ reasoning: false,
6755
+ input: ["text"],
6756
+ cost: {
6757
+ input: 0,
6758
+ output: 0,
6759
+ cacheRead: 0,
6760
+ cacheWrite: 0,
6761
+ },
6762
+ contextWindow: 262144,
6763
+ maxTokens: 66536,
6764
+ },
6765
+ "qwen/qwen3.5-122b-a10b": {
6766
+ id: "qwen/qwen3.5-122b-a10b",
6767
+ name: "Qwen3.5 122B-A10B",
6768
+ api: "openai-completions",
6769
+ provider: "nvidia",
6770
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6771
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6772
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6773
+ reasoning: true,
6774
+ input: ["text", "image"],
6775
+ cost: {
6776
+ input: 0,
6777
+ output: 0,
6778
+ cacheRead: 0,
6779
+ cacheWrite: 0,
6780
+ },
6781
+ contextWindow: 262144,
6782
+ maxTokens: 65536,
6783
+ },
6784
+ "stepfun-ai/step-3.5-flash": {
6785
+ id: "stepfun-ai/step-3.5-flash",
6786
+ name: "Step 3.5 Flash",
6787
+ api: "openai-completions",
6788
+ provider: "nvidia",
6789
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6790
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6791
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6792
+ reasoning: true,
6793
+ input: ["text"],
6794
+ cost: {
6795
+ input: 0,
6796
+ output: 0,
6797
+ cacheRead: 0,
6798
+ cacheWrite: 0,
6799
+ },
6800
+ contextWindow: 256000,
6801
+ maxTokens: 16384,
6802
+ },
6803
+ "stepfun-ai/step-3.7-flash": {
6804
+ id: "stepfun-ai/step-3.7-flash",
6805
+ name: "Step 3.7 Flash",
6261
6806
  api: "openai-completions",
6262
- provider: "moonshotai-cn",
6263
- baseUrl: "https://api.moonshot.cn/v1",
6264
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6807
+ provider: "nvidia",
6808
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6809
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6810
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6265
6811
  reasoning: true,
6266
6812
  input: ["text", "image"],
6267
6813
  cost: {
6268
- input: 0.6,
6269
- output: 3,
6270
- cacheRead: 0.1,
6814
+ input: 0,
6815
+ output: 0,
6816
+ cacheRead: 0,
6271
6817
  cacheWrite: 0,
6272
6818
  },
6273
- contextWindow: 262144,
6274
- maxTokens: 262144,
6819
+ contextWindow: 256000,
6820
+ maxTokens: 16384,
6275
6821
  },
6276
- "kimi-k2.6": {
6277
- id: "kimi-k2.6",
6278
- name: "Kimi K2.6",
6822
+ "z-ai/glm-5.1": {
6823
+ id: "z-ai/glm-5.1",
6824
+ name: "GLM-5.1",
6279
6825
  api: "openai-completions",
6280
- provider: "moonshotai-cn",
6281
- baseUrl: "https://api.moonshot.cn/v1",
6282
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6826
+ provider: "nvidia",
6827
+ baseUrl: "https://integrate.api.nvidia.com/v1",
6828
+ headers: { "NVCF-POLL-SECONDS": "3600" },
6829
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
6283
6830
  reasoning: true,
6284
- input: ["text", "image"],
6831
+ input: ["text"],
6285
6832
  cost: {
6286
- input: 0.95,
6287
- output: 4,
6288
- cacheRead: 0.16,
6833
+ input: 0,
6834
+ output: 0,
6835
+ cacheRead: 0,
6289
6836
  cacheWrite: 0,
6290
6837
  },
6291
- contextWindow: 262144,
6292
- maxTokens: 262144,
6838
+ contextWindow: 131072,
6839
+ maxTokens: 131072,
6293
6840
  },
6294
6841
  },
6295
6842
  "openai": {
@@ -6866,7 +7413,7 @@ export const MODELS = {
6866
7413
  provider: "openai",
6867
7414
  baseUrl: "https://api.openai.com/v1",
6868
7415
  reasoning: true,
6869
- thinkingLevelMap: { "off": "none", "xhigh": "xhigh" },
7416
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
6870
7417
  input: ["text", "image"],
6871
7418
  cost: {
6872
7419
  input: 5,
@@ -7033,42 +7580,6 @@ export const MODELS = {
7033
7580
  },
7034
7581
  },
7035
7582
  "openai-codex": {
7036
- "gpt-5.2": {
7037
- id: "gpt-5.2",
7038
- name: "GPT-5.2",
7039
- api: "openai-codex-responses",
7040
- provider: "openai-codex",
7041
- baseUrl: "https://chatgpt.com/backend-api",
7042
- reasoning: true,
7043
- thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
7044
- input: ["text", "image"],
7045
- cost: {
7046
- input: 1.75,
7047
- output: 14,
7048
- cacheRead: 0.175,
7049
- cacheWrite: 0,
7050
- },
7051
- contextWindow: 272000,
7052
- maxTokens: 128000,
7053
- },
7054
- "gpt-5.3-codex": {
7055
- id: "gpt-5.3-codex",
7056
- name: "GPT-5.3 Codex",
7057
- api: "openai-codex-responses",
7058
- provider: "openai-codex",
7059
- baseUrl: "https://chatgpt.com/backend-api",
7060
- reasoning: true,
7061
- thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
7062
- input: ["text", "image"],
7063
- cost: {
7064
- input: 1.75,
7065
- output: 14,
7066
- cacheRead: 0.175,
7067
- cacheWrite: 0,
7068
- },
7069
- contextWindow: 272000,
7070
- maxTokens: 128000,
7071
- },
7072
7583
  "gpt-5.3-codex-spark": {
7073
7584
  id: "gpt-5.3-codex-spark",
7074
7585
  name: "GPT-5.3 Codex Spark",
@@ -7236,7 +7747,7 @@ export const MODELS = {
7236
7747
  api: "anthropic-messages",
7237
7748
  provider: "opencode",
7238
7749
  baseUrl: "https://opencode.ai/zen",
7239
- compat: { "forceAdaptiveThinking": true },
7750
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
7240
7751
  reasoning: true,
7241
7752
  thinkingLevelMap: { "xhigh": "xhigh" },
7242
7753
  input: ["text", "image"],
@@ -7255,7 +7766,7 @@ export const MODELS = {
7255
7766
  api: "anthropic-messages",
7256
7767
  provider: "opencode",
7257
7768
  baseUrl: "https://opencode.ai/zen",
7258
- compat: { "forceAdaptiveThinking": true },
7769
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
7259
7770
  reasoning: true,
7260
7771
  thinkingLevelMap: { "xhigh": "xhigh" },
7261
7772
  input: ["text", "image"],
@@ -7320,6 +7831,25 @@ export const MODELS = {
7320
7831
  contextWindow: 1000000,
7321
7832
  maxTokens: 64000,
7322
7833
  },
7834
+ "deepseek-v4-flash": {
7835
+ id: "deepseek-v4-flash",
7836
+ name: "DeepSeek V4 Flash",
7837
+ api: "openai-completions",
7838
+ provider: "opencode",
7839
+ baseUrl: "https://opencode.ai/zen/v1",
7840
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7841
+ reasoning: true,
7842
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7843
+ input: ["text"],
7844
+ cost: {
7845
+ input: 0.14,
7846
+ output: 0.28,
7847
+ cacheRead: 0.03,
7848
+ cacheWrite: 0,
7849
+ },
7850
+ contextWindow: 1000000,
7851
+ maxTokens: 384000,
7852
+ },
7323
7853
  "deepseek-v4-flash-free": {
7324
7854
  id: "deepseek-v4-flash-free",
7325
7855
  name: "DeepSeek V4 Flash Free",
@@ -7721,7 +8251,9 @@ export const MODELS = {
7721
8251
  api: "openai-completions",
7722
8252
  provider: "opencode",
7723
8253
  baseUrl: "https://opencode.ai/zen/v1",
8254
+ compat: { "supportsReasoningEffort": false },
7724
8255
  reasoning: true,
8256
+ thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null },
7725
8257
  input: ["text", "image"],
7726
8258
  cost: {
7727
8259
  input: 1,
@@ -7755,6 +8287,7 @@ export const MODELS = {
7755
8287
  api: "openai-completions",
7756
8288
  provider: "opencode",
7757
8289
  baseUrl: "https://opencode.ai/zen/v1",
8290
+ compat: { "thinkingFormat": "deepseek", "supportsReasoningEffort": false },
7758
8291
  reasoning: true,
7759
8292
  input: ["text", "image"],
7760
8293
  cost: {
@@ -7780,8 +8313,8 @@ export const MODELS = {
7780
8313
  cacheRead: 0,
7781
8314
  cacheWrite: 0,
7782
8315
  },
7783
- contextWindow: 1000000,
7784
- maxTokens: 128000,
8316
+ contextWindow: 200000,
8317
+ maxTokens: 32000,
7785
8318
  },
7786
8319
  "minimax-m2.5": {
7787
8320
  id: "minimax-m2.5",
@@ -7817,9 +8350,26 @@ export const MODELS = {
7817
8350
  contextWindow: 204800,
7818
8351
  maxTokens: 131072,
7819
8352
  },
7820
- "nemotron-3-super-free": {
7821
- id: "nemotron-3-super-free",
7822
- name: "Nemotron 3 Super Free",
8353
+ "minimax-m3-free": {
8354
+ id: "minimax-m3-free",
8355
+ name: "MiniMax M3 Free",
8356
+ api: "anthropic-messages",
8357
+ provider: "opencode",
8358
+ baseUrl: "https://opencode.ai/zen",
8359
+ reasoning: true,
8360
+ input: ["text", "image"],
8361
+ cost: {
8362
+ input: 0,
8363
+ output: 0,
8364
+ cacheRead: 0,
8365
+ cacheWrite: 0,
8366
+ },
8367
+ contextWindow: 200000,
8368
+ maxTokens: 32000,
8369
+ },
8370
+ "nemotron-3-ultra-free": {
8371
+ id: "nemotron-3-ultra-free",
8372
+ name: "Nemotron 3 Ultra Free",
7823
8373
  api: "openai-completions",
7824
8374
  provider: "opencode",
7825
8375
  baseUrl: "https://opencode.ai/zen/v1",
@@ -7831,7 +8381,7 @@ export const MODELS = {
7831
8381
  cacheRead: 0,
7832
8382
  cacheWrite: 0,
7833
8383
  },
7834
- contextWindow: 204800,
8384
+ contextWindow: 1000000,
7835
8385
  maxTokens: 128000,
7836
8386
  },
7837
8387
  "qwen3.5-plus": {
@@ -7965,9 +8515,9 @@ export const MODELS = {
7965
8515
  api: "openai-completions",
7966
8516
  provider: "opencode-go",
7967
8517
  baseUrl: "https://opencode.ai/zen/go/v1",
7968
- compat: { "thinkingFormat": "string-thinking" },
8518
+ compat: { "thinkingFormat": "deepseek", "supportsReasoningEffort": false },
7969
8519
  reasoning: true,
7970
- thinkingLevelMap: { "off": "none" },
8520
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
7971
8521
  input: ["text", "image"],
7972
8522
  cost: {
7973
8523
  input: 0.95,
@@ -8046,23 +8596,22 @@ export const MODELS = {
8046
8596
  contextWindow: 204800,
8047
8597
  maxTokens: 131072,
8048
8598
  },
8049
- "qwen3.5-plus": {
8050
- id: "qwen3.5-plus",
8051
- name: "Qwen3.5 Plus",
8052
- api: "openai-completions",
8599
+ "minimax-m3": {
8600
+ id: "minimax-m3",
8601
+ name: "MiniMax M3",
8602
+ api: "anthropic-messages",
8053
8603
  provider: "opencode-go",
8054
- baseUrl: "https://opencode.ai/zen/go/v1",
8055
- compat: { "thinkingFormat": "qwen" },
8604
+ baseUrl: "https://opencode.ai/zen/go",
8056
8605
  reasoning: true,
8057
8606
  input: ["text", "image"],
8058
8607
  cost: {
8059
- input: 0.2,
8060
- output: 1.2,
8061
- cacheRead: 0.02,
8062
- cacheWrite: 0.25,
8608
+ input: 0.6,
8609
+ output: 2.4,
8610
+ cacheRead: 0.12,
8611
+ cacheWrite: 0,
8063
8612
  },
8064
- contextWindow: 262144,
8065
- maxTokens: 65536,
8613
+ contextWindow: 512000,
8614
+ maxTokens: 131072,
8066
8615
  },
8067
8616
  "qwen3.6-plus": {
8068
8617
  id: "qwen3.6-plus",
@@ -8099,6 +8648,23 @@ export const MODELS = {
8099
8648
  contextWindow: 1000000,
8100
8649
  maxTokens: 65536,
8101
8650
  },
8651
+ "qwen3.7-plus": {
8652
+ id: "qwen3.7-plus",
8653
+ name: "Qwen3.7 Plus",
8654
+ api: "anthropic-messages",
8655
+ provider: "opencode-go",
8656
+ baseUrl: "https://opencode.ai/zen/go",
8657
+ reasoning: true,
8658
+ input: ["text", "image"],
8659
+ cost: {
8660
+ input: 0.4,
8661
+ output: 1.6,
8662
+ cacheRead: 0.04,
8663
+ cacheWrite: 0.5,
8664
+ },
8665
+ contextWindow: 262144,
8666
+ maxTokens: 65536,
8667
+ },
8102
8668
  },
8103
8669
  "openrouter": {
8104
8670
  "ai21/jamba-large-1.7": {
@@ -8532,23 +9098,6 @@ export const MODELS = {
8532
9098
  contextWindow: 2000000,
8533
9099
  maxTokens: 30000,
8534
9100
  },
8535
- "baidu/ernie-4.5-21b-a3b": {
8536
- id: "baidu/ernie-4.5-21b-a3b",
8537
- name: "Baidu: ERNIE 4.5 21B A3B",
8538
- api: "openai-completions",
8539
- provider: "openrouter",
8540
- baseUrl: "https://openrouter.ai/api/v1",
8541
- reasoning: false,
8542
- input: ["text"],
8543
- cost: {
8544
- input: 0.07,
8545
- output: 0.28,
8546
- cacheRead: 0,
8547
- cacheWrite: 0,
8548
- },
8549
- contextWindow: 131072,
8550
- maxTokens: 8000,
8551
- },
8552
9101
  "baidu/ernie-4.5-vl-28b-a3b": {
8553
9102
  id: "baidu/ernie-4.5-vl-28b-a3b",
8554
9103
  name: "Baidu: ERNIE 4.5 VL 28B A3B",
@@ -8677,8 +9226,8 @@ export const MODELS = {
8677
9226
  reasoning: false,
8678
9227
  input: ["text"],
8679
9228
  cost: {
8680
- input: 0.2288,
8681
- output: 0.9144,
9229
+ input: 0.20020000000000002,
9230
+ output: 0.8000999999999999,
8682
9231
  cacheRead: 0,
8683
9232
  cacheWrite: 0,
8684
9233
  },
@@ -8779,13 +9328,13 @@ export const MODELS = {
8779
9328
  reasoning: true,
8780
9329
  input: ["text"],
8781
9330
  cost: {
8782
- input: 0.252,
8783
- output: 0.378,
8784
- cacheRead: 0.0252,
9331
+ input: 0.2288,
9332
+ output: 0.3432,
9333
+ cacheRead: 0,
8785
9334
  cacheWrite: 0,
8786
9335
  },
8787
9336
  contextWindow: 131072,
8788
- maxTokens: 65536,
9337
+ maxTokens: 64000,
8789
9338
  },
8790
9339
  "deepseek/deepseek-v3.2-exp": {
8791
9340
  id: "deepseek/deepseek-v3.2-exp",
@@ -8815,32 +9364,13 @@ export const MODELS = {
8815
9364
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
8816
9365
  input: ["text"],
8817
9366
  cost: {
8818
- input: 0.09999999999999999,
8819
- output: 0.19999999999999998,
8820
- cacheRead: 0.02,
8821
- cacheWrite: 0,
8822
- },
8823
- contextWindow: 1048576,
8824
- maxTokens: 16384,
8825
- },
8826
- "deepseek/deepseek-v4-flash:free": {
8827
- id: "deepseek/deepseek-v4-flash:free",
8828
- name: "DeepSeek: DeepSeek V4 Flash (free)",
8829
- api: "openai-completions",
8830
- provider: "openrouter",
8831
- baseUrl: "https://openrouter.ai/api/v1",
8832
- compat: { "requiresReasoningContentOnAssistantMessages": true },
8833
- reasoning: true,
8834
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
8835
- input: ["text"],
8836
- cost: {
8837
- input: 0,
8838
- output: 0,
8839
- cacheRead: 0,
9367
+ input: 0.0983,
9368
+ output: 0.1966,
9369
+ cacheRead: 0.019700000000000002,
8840
9370
  cacheWrite: 0,
8841
9371
  },
8842
9372
  contextWindow: 1048576,
8843
- maxTokens: 384000,
9373
+ maxTokens: 131072,
8844
9374
  },
8845
9375
  "deepseek/deepseek-v4-pro": {
8846
9376
  id: "deepseek/deepseek-v4-pro",
@@ -8878,40 +9408,6 @@ export const MODELS = {
8878
9408
  contextWindow: 32768,
8879
9409
  maxTokens: 4096,
8880
9410
  },
8881
- "google/gemini-2.0-flash-001": {
8882
- id: "google/gemini-2.0-flash-001",
8883
- name: "Google: Gemini 2.0 Flash",
8884
- api: "openai-completions",
8885
- provider: "openrouter",
8886
- baseUrl: "https://openrouter.ai/api/v1",
8887
- reasoning: false,
8888
- input: ["text", "image"],
8889
- cost: {
8890
- input: 0.09999999999999999,
8891
- output: 0.39999999999999997,
8892
- cacheRead: 0.024999999999999998,
8893
- cacheWrite: 0.08333333333333334,
8894
- },
8895
- contextWindow: 1000000,
8896
- maxTokens: 8192,
8897
- },
8898
- "google/gemini-2.0-flash-lite-001": {
8899
- id: "google/gemini-2.0-flash-lite-001",
8900
- name: "Google: Gemini 2.0 Flash Lite",
8901
- api: "openai-completions",
8902
- provider: "openrouter",
8903
- baseUrl: "https://openrouter.ai/api/v1",
8904
- reasoning: false,
8905
- input: ["text", "image"],
8906
- cost: {
8907
- input: 0.075,
8908
- output: 0.3,
8909
- cacheRead: 0,
8910
- cacheWrite: 0,
8911
- },
8912
- contextWindow: 1048576,
8913
- maxTokens: 8192,
8914
- },
8915
9411
  "google/gemini-2.5-flash": {
8916
9412
  id: "google/gemini-2.5-flash",
8917
9413
  name: "Google: Gemini 2.5 Flash",
@@ -9389,6 +9885,23 @@ export const MODELS = {
9389
9885
  contextWindow: 131072,
9390
9886
  maxTokens: 4096,
9391
9887
  },
9888
+ "meta-llama/llama-4-maverick": {
9889
+ id: "meta-llama/llama-4-maverick",
9890
+ name: "Meta: Llama 4 Maverick",
9891
+ api: "openai-completions",
9892
+ provider: "openrouter",
9893
+ baseUrl: "https://openrouter.ai/api/v1",
9894
+ reasoning: false,
9895
+ input: ["text", "image"],
9896
+ cost: {
9897
+ input: 0.15,
9898
+ output: 0.6,
9899
+ cacheRead: 0,
9900
+ cacheWrite: 0,
9901
+ },
9902
+ contextWindow: 1048576,
9903
+ maxTokens: 16384,
9904
+ },
9392
9905
  "meta-llama/llama-4-scout": {
9393
9906
  id: "meta-llama/llama-4-scout",
9394
9907
  name: "Meta: Llama 4 Scout",
@@ -9474,39 +9987,39 @@ export const MODELS = {
9474
9987
  contextWindow: 204800,
9475
9988
  maxTokens: 196608,
9476
9989
  },
9477
- "minimax/minimax-m2.5:free": {
9478
- id: "minimax/minimax-m2.5:free",
9479
- name: "MiniMax: MiniMax M2.5 (free)",
9990
+ "minimax/minimax-m2.7": {
9991
+ id: "minimax/minimax-m2.7",
9992
+ name: "MiniMax: MiniMax M2.7",
9480
9993
  api: "openai-completions",
9481
9994
  provider: "openrouter",
9482
9995
  baseUrl: "https://openrouter.ai/api/v1",
9483
9996
  reasoning: true,
9484
9997
  input: ["text"],
9485
9998
  cost: {
9486
- input: 0,
9487
- output: 0,
9999
+ input: 0.27899999999999997,
10000
+ output: 1.2,
9488
10001
  cacheRead: 0,
9489
10002
  cacheWrite: 0,
9490
10003
  },
9491
10004
  contextWindow: 204800,
9492
- maxTokens: 8192,
10005
+ maxTokens: 131072,
9493
10006
  },
9494
- "minimax/minimax-m2.7": {
9495
- id: "minimax/minimax-m2.7",
9496
- name: "MiniMax: MiniMax M2.7",
10007
+ "minimax/minimax-m3": {
10008
+ id: "minimax/minimax-m3",
10009
+ name: "MiniMax: MiniMax M3",
9497
10010
  api: "openai-completions",
9498
10011
  provider: "openrouter",
9499
10012
  baseUrl: "https://openrouter.ai/api/v1",
9500
10013
  reasoning: true,
9501
- input: ["text"],
10014
+ input: ["text", "image"],
9502
10015
  cost: {
9503
- input: 0.27899999999999997,
10016
+ input: 0.3,
9504
10017
  output: 1.2,
9505
- cacheRead: 0,
10018
+ cacheRead: 0.06,
9506
10019
  cacheWrite: 0,
9507
10020
  },
9508
- contextWindow: 204800,
9509
- maxTokens: 131072,
10021
+ contextWindow: 1048576,
10022
+ maxTokens: 512000,
9510
10023
  },
9511
10024
  "mistralai/codestral-2508": {
9512
10025
  id: "mistralai/codestral-2508",
@@ -9542,40 +10055,6 @@ export const MODELS = {
9542
10055
  contextWindow: 262144,
9543
10056
  maxTokens: 4096,
9544
10057
  },
9545
- "mistralai/devstral-medium": {
9546
- id: "mistralai/devstral-medium",
9547
- name: "Mistral: Devstral Medium",
9548
- api: "openai-completions",
9549
- provider: "openrouter",
9550
- baseUrl: "https://openrouter.ai/api/v1",
9551
- reasoning: false,
9552
- input: ["text"],
9553
- cost: {
9554
- input: 0.39999999999999997,
9555
- output: 2,
9556
- cacheRead: 0.04,
9557
- cacheWrite: 0,
9558
- },
9559
- contextWindow: 131072,
9560
- maxTokens: 4096,
9561
- },
9562
- "mistralai/devstral-small": {
9563
- id: "mistralai/devstral-small",
9564
- name: "Mistral: Devstral Small 1.1",
9565
- api: "openai-completions",
9566
- provider: "openrouter",
9567
- baseUrl: "https://openrouter.ai/api/v1",
9568
- reasoning: false,
9569
- input: ["text"],
9570
- cost: {
9571
- input: 0.09999999999999999,
9572
- output: 0.3,
9573
- cacheRead: 0.01,
9574
- cacheWrite: 0,
9575
- },
9576
- contextWindow: 131072,
9577
- maxTokens: 4096,
9578
- },
9579
10058
  "mistralai/ministral-14b-2512": {
9580
10059
  id: "mistralai/ministral-14b-2512",
9581
10060
  name: "Mistral: Ministral 3 14B 2512",
@@ -9661,23 +10140,6 @@ export const MODELS = {
9661
10140
  contextWindow: 131072,
9662
10141
  maxTokens: 4096,
9663
10142
  },
9664
- "mistralai/mistral-large-2411": {
9665
- id: "mistralai/mistral-large-2411",
9666
- name: "Mistral Large 2411",
9667
- api: "openai-completions",
9668
- provider: "openrouter",
9669
- baseUrl: "https://openrouter.ai/api/v1",
9670
- reasoning: false,
9671
- input: ["text"],
9672
- cost: {
9673
- input: 2,
9674
- output: 6,
9675
- cacheRead: 0.19999999999999998,
9676
- cacheWrite: 0,
9677
- },
9678
- contextWindow: 131072,
9679
- maxTokens: 4096,
9680
- },
9681
10143
  "mistralai/mistral-large-2512": {
9682
10144
  id: "mistralai/mistral-large-2512",
9683
10145
  name: "Mistral: Mistral Large 3 2512",
@@ -9831,23 +10293,6 @@ export const MODELS = {
9831
10293
  contextWindow: 65536,
9832
10294
  maxTokens: 4096,
9833
10295
  },
9834
- "mistralai/pixtral-large-2411": {
9835
- id: "mistralai/pixtral-large-2411",
9836
- name: "Mistral: Pixtral Large 2411",
9837
- api: "openai-completions",
9838
- provider: "openrouter",
9839
- baseUrl: "https://openrouter.ai/api/v1",
9840
- reasoning: false,
9841
- input: ["text", "image"],
9842
- cost: {
9843
- input: 2,
9844
- output: 6,
9845
- cacheRead: 0.19999999999999998,
9846
- cacheWrite: 0,
9847
- },
9848
- contextWindow: 131072,
9849
- maxTokens: 4096,
9850
- },
9851
10296
  "mistralai/voxtral-small-24b-2507": {
9852
10297
  id: "mistralai/voxtral-small-24b-2507",
9853
10298
  name: "Mistral: Voxtral Small 24B 2507",
@@ -9939,16 +10384,17 @@ export const MODELS = {
9939
10384
  api: "openai-completions",
9940
10385
  provider: "openrouter",
9941
10386
  baseUrl: "https://openrouter.ai/api/v1",
10387
+ compat: { "supportsDeveloperRole": false, "requiresReasoningContentOnAssistantMessages": true },
9942
10388
  reasoning: true,
9943
10389
  input: ["text", "image"],
9944
10390
  cost: {
9945
- input: 0.73,
9946
- output: 3.49,
9947
- cacheRead: 0.25,
10391
+ input: 0.684,
10392
+ output: 3.42,
10393
+ cacheRead: 0.144,
9948
10394
  cacheWrite: 0,
9949
10395
  },
9950
10396
  contextWindow: 262144,
9951
- maxTokens: 262142,
10397
+ maxTokens: 262144,
9952
10398
  },
9953
10399
  "moonshotai/kimi-k2.6:free": {
9954
10400
  id: "moonshotai/kimi-k2.6:free",
@@ -9956,6 +10402,7 @@ export const MODELS = {
9956
10402
  api: "openai-completions",
9957
10403
  provider: "openrouter",
9958
10404
  baseUrl: "https://openrouter.ai/api/v1",
10405
+ compat: { "supportsDeveloperRole": false, "requiresReasoningContentOnAssistantMessages": true },
9959
10406
  reasoning: true,
9960
10407
  input: ["text", "image"],
9961
10408
  cost: {
@@ -10086,6 +10533,40 @@ export const MODELS = {
10086
10533
  contextWindow: 1000000,
10087
10534
  maxTokens: 262144,
10088
10535
  },
10536
+ "nvidia/nemotron-3-ultra-550b-a55b": {
10537
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
10538
+ name: "NVIDIA: Nemotron 3 Ultra",
10539
+ api: "openai-completions",
10540
+ provider: "openrouter",
10541
+ baseUrl: "https://openrouter.ai/api/v1",
10542
+ reasoning: true,
10543
+ input: ["text"],
10544
+ cost: {
10545
+ input: 0.5,
10546
+ output: 2.5,
10547
+ cacheRead: 0.15,
10548
+ cacheWrite: 0,
10549
+ },
10550
+ contextWindow: 1000000,
10551
+ maxTokens: 16384,
10552
+ },
10553
+ "nvidia/nemotron-3-ultra-550b-a55b:free": {
10554
+ id: "nvidia/nemotron-3-ultra-550b-a55b:free",
10555
+ name: "NVIDIA: Nemotron 3 Ultra (free)",
10556
+ api: "openai-completions",
10557
+ provider: "openrouter",
10558
+ baseUrl: "https://openrouter.ai/api/v1",
10559
+ reasoning: true,
10560
+ input: ["text"],
10561
+ cost: {
10562
+ input: 0,
10563
+ output: 0,
10564
+ cacheRead: 0,
10565
+ cacheWrite: 0,
10566
+ },
10567
+ contextWindow: 1000000,
10568
+ maxTokens: 65536,
10569
+ },
10089
10570
  "nvidia/nemotron-nano-12b-v2-vl:free": {
10090
10571
  id: "nvidia/nemotron-nano-12b-v2-vl:free",
10091
10572
  name: "NVIDIA: Nemotron Nano 12B 2 VL (free)",
@@ -10205,23 +10686,6 @@ export const MODELS = {
10205
10686
  contextWindow: 8191,
10206
10687
  maxTokens: 4096,
10207
10688
  },
10208
- "openai/gpt-4-0314": {
10209
- id: "openai/gpt-4-0314",
10210
- name: "OpenAI: GPT-4 (older v0314)",
10211
- api: "openai-completions",
10212
- provider: "openrouter",
10213
- baseUrl: "https://openrouter.ai/api/v1",
10214
- reasoning: false,
10215
- input: ["text"],
10216
- cost: {
10217
- input: 30,
10218
- output: 60,
10219
- cacheRead: 0,
10220
- cacheWrite: 0,
10221
- },
10222
- contextWindow: 8191,
10223
- maxTokens: 4096,
10224
- },
10225
10689
  "openai/gpt-4-1106-preview": {
10226
10690
  id: "openai/gpt-4-1106-preview",
10227
10691
  name: "OpenAI: GPT-4 Turbo (older v1106)",
@@ -10392,23 +10856,6 @@ export const MODELS = {
10392
10856
  contextWindow: 128000,
10393
10857
  maxTokens: 16384,
10394
10858
  },
10395
- "openai/gpt-4o-audio-preview": {
10396
- id: "openai/gpt-4o-audio-preview",
10397
- name: "OpenAI: GPT-4o Audio",
10398
- api: "openai-completions",
10399
- provider: "openrouter",
10400
- baseUrl: "https://openrouter.ai/api/v1",
10401
- reasoning: false,
10402
- input: ["text"],
10403
- cost: {
10404
- input: 2.5,
10405
- output: 10,
10406
- cacheRead: 0,
10407
- cacheWrite: 0,
10408
- },
10409
- contextWindow: 128000,
10410
- maxTokens: 16384,
10411
- },
10412
10859
  "openai/gpt-4o-mini": {
10413
10860
  id: "openai/gpt-4o-mini",
10414
10861
  name: "OpenAI: GPT-4o-mini",
@@ -10923,13 +11370,13 @@ export const MODELS = {
10923
11370
  reasoning: true,
10924
11371
  input: ["text"],
10925
11372
  cost: {
10926
- input: 0.03,
11373
+ input: 0.029,
10927
11374
  output: 0.14,
10928
11375
  cacheRead: 0,
10929
11376
  cacheWrite: 0,
10930
11377
  },
10931
11378
  contextWindow: 131072,
10932
- maxTokens: 131072,
11379
+ maxTokens: 4096,
10933
11380
  },
10934
11381
  "openai/gpt-oss-20b:free": {
10935
11382
  id: "openai/gpt-oss-20b:free",
@@ -11237,23 +11684,6 @@ export const MODELS = {
11237
11684
  contextWindow: 131072,
11238
11685
  maxTokens: 16384,
11239
11686
  },
11240
- "qwen/qwen-2.5-7b-instruct": {
11241
- id: "qwen/qwen-2.5-7b-instruct",
11242
- name: "Qwen: Qwen2.5 7B Instruct",
11243
- api: "openai-completions",
11244
- provider: "openrouter",
11245
- baseUrl: "https://openrouter.ai/api/v1",
11246
- reasoning: false,
11247
- input: ["text"],
11248
- cost: {
11249
- input: 0.04,
11250
- output: 0.09999999999999999,
11251
- cacheRead: 0,
11252
- cacheWrite: 0,
11253
- },
11254
- contextWindow: 131072,
11255
- maxTokens: 32768,
11256
- },
11257
11687
  "qwen/qwen-plus": {
11258
11688
  id: "qwen/qwen-plus",
11259
11689
  name: "Qwen: Qwen-Plus",
@@ -11365,13 +11795,13 @@ export const MODELS = {
11365
11795
  reasoning: true,
11366
11796
  input: ["text"],
11367
11797
  cost: {
11368
- input: 0.14950000000000002,
11369
- output: 1.495,
11370
- cacheRead: 0,
11798
+ input: 0.09999999999999999,
11799
+ output: 0.09999999999999999,
11800
+ cacheRead: 0.09999999999999999,
11371
11801
  cacheWrite: 0,
11372
11802
  },
11373
11803
  contextWindow: 262144,
11374
- maxTokens: 4096,
11804
+ maxTokens: 262144,
11375
11805
  },
11376
11806
  "qwen/qwen3-30b-a3b": {
11377
11807
  id: "qwen/qwen3-30b-a3b",
@@ -11399,13 +11829,13 @@ export const MODELS = {
11399
11829
  reasoning: false,
11400
11830
  input: ["text"],
11401
11831
  cost: {
11402
- input: 0.09,
11403
- output: 0.3,
11832
+ input: 0.04815,
11833
+ output: 0.19305,
11404
11834
  cacheRead: 0,
11405
11835
  cacheWrite: 0,
11406
11836
  },
11407
- contextWindow: 262144,
11408
- maxTokens: 262144,
11837
+ contextWindow: 131072,
11838
+ maxTokens: 32000,
11409
11839
  },
11410
11840
  "qwen/qwen3-30b-a3b-thinking-2507": {
11411
11841
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -11807,13 +12237,13 @@ export const MODELS = {
11807
12237
  reasoning: true,
11808
12238
  input: ["text", "image"],
11809
12239
  cost: {
11810
- input: 0.13899999999999998,
12240
+ input: 0.14,
11811
12241
  output: 1,
11812
- cacheRead: 0,
12242
+ cacheRead: 0.049999999999999996,
11813
12243
  cacheWrite: 0,
11814
12244
  },
11815
12245
  contextWindow: 262144,
11816
- maxTokens: 4096,
12246
+ maxTokens: 262144,
11817
12247
  },
11818
12248
  "qwen/qwen3.5-397b-a17b": {
11819
12249
  id: "qwen/qwen3.5-397b-a17b",
@@ -12002,6 +12432,23 @@ export const MODELS = {
12002
12432
  contextWindow: 1000000,
12003
12433
  maxTokens: 65536,
12004
12434
  },
12435
+ "qwen/qwen3.7-plus": {
12436
+ id: "qwen/qwen3.7-plus",
12437
+ name: "Qwen: Qwen3.7 Plus",
12438
+ api: "openai-completions",
12439
+ provider: "openrouter",
12440
+ baseUrl: "https://openrouter.ai/api/v1",
12441
+ reasoning: true,
12442
+ input: ["text", "image"],
12443
+ cost: {
12444
+ input: 0.39999999999999997,
12445
+ output: 1.5999999999999999,
12446
+ cacheRead: 0.08,
12447
+ cacheWrite: 0.5,
12448
+ },
12449
+ contextWindow: 1000000,
12450
+ maxTokens: 65536,
12451
+ },
12005
12452
  "rekaai/reka-edge": {
12006
12453
  id: "rekaai/reka-edge",
12007
12454
  name: "Reka Edge",
@@ -12087,6 +12534,23 @@ export const MODELS = {
12087
12534
  contextWindow: 262144,
12088
12535
  maxTokens: 16384,
12089
12536
  },
12537
+ "stepfun/step-3.7-flash": {
12538
+ id: "stepfun/step-3.7-flash",
12539
+ name: "StepFun: Step 3.7 Flash",
12540
+ api: "openai-completions",
12541
+ provider: "openrouter",
12542
+ baseUrl: "https://openrouter.ai/api/v1",
12543
+ reasoning: true,
12544
+ input: ["text", "image"],
12545
+ cost: {
12546
+ input: 0.19999999999999998,
12547
+ output: 1.15,
12548
+ cacheRead: 0.04,
12549
+ cacheWrite: 0,
12550
+ },
12551
+ contextWindow: 256000,
12552
+ maxTokens: 256000,
12553
+ },
12090
12554
  "tencent/hy3-preview": {
12091
12555
  id: "tencent/hy3-preview",
12092
12556
  name: "Tencent: Hy3 preview",
@@ -12223,40 +12687,6 @@ export const MODELS = {
12223
12687
  contextWindow: 262144,
12224
12688
  maxTokens: 65536,
12225
12689
  },
12226
- "xiaomi/mimo-v2-omni": {
12227
- id: "xiaomi/mimo-v2-omni",
12228
- name: "Xiaomi: MiMo-V2-Omni",
12229
- api: "openai-completions",
12230
- provider: "openrouter",
12231
- baseUrl: "https://openrouter.ai/api/v1",
12232
- reasoning: true,
12233
- input: ["text", "image"],
12234
- cost: {
12235
- input: 0.39999999999999997,
12236
- output: 2,
12237
- cacheRead: 0.08,
12238
- cacheWrite: 0,
12239
- },
12240
- contextWindow: 262144,
12241
- maxTokens: 65536,
12242
- },
12243
- "xiaomi/mimo-v2-pro": {
12244
- id: "xiaomi/mimo-v2-pro",
12245
- name: "Xiaomi: MiMo-V2-Pro",
12246
- api: "openai-completions",
12247
- provider: "openrouter",
12248
- baseUrl: "https://openrouter.ai/api/v1",
12249
- reasoning: true,
12250
- input: ["text"],
12251
- cost: {
12252
- input: 1,
12253
- output: 3,
12254
- cacheRead: 0.19999999999999998,
12255
- cacheWrite: 0,
12256
- },
12257
- contextWindow: 1048576,
12258
- maxTokens: 131072,
12259
- },
12260
12690
  "xiaomi/mimo-v2.5": {
12261
12691
  id: "xiaomi/mimo-v2.5",
12262
12692
  name: "Xiaomi: MiMo-V2.5",
@@ -12606,13 +13036,13 @@ export const MODELS = {
12606
13036
  reasoning: true,
12607
13037
  input: ["text", "image"],
12608
13038
  cost: {
12609
- input: 0.73,
12610
- output: 3.49,
12611
- cacheRead: 0.25,
13039
+ input: 0.684,
13040
+ output: 3.42,
13041
+ cacheRead: 0.144,
12612
13042
  cacheWrite: 0,
12613
13043
  },
12614
13044
  contextWindow: 262144,
12615
- maxTokens: 262142,
13045
+ maxTokens: 262144,
12616
13046
  },
12617
13047
  "~openai/gpt-latest": {
12618
13048
  id: "~openai/gpt-latest",
@@ -13010,20 +13440,20 @@ export const MODELS = {
13010
13440
  },
13011
13441
  "alibaba/qwen-3-235b": {
13012
13442
  id: "alibaba/qwen-3-235b",
13013
- name: "Qwen3 235B A22b Instruct 2507",
13443
+ name: "Qwen3 235B A22B",
13014
13444
  api: "anthropic-messages",
13015
13445
  provider: "vercel-ai-gateway",
13016
13446
  baseUrl: "https://ai-gateway.vercel.sh",
13017
- reasoning: false,
13447
+ reasoning: true,
13018
13448
  input: ["text"],
13019
13449
  cost: {
13020
- input: 0.6,
13021
- output: 1.2,
13022
- cacheRead: 0.6,
13450
+ input: 0.22,
13451
+ output: 0.88,
13452
+ cacheRead: 0,
13023
13453
  cacheWrite: 0,
13024
13454
  },
13025
- contextWindow: 131000,
13026
- maxTokens: 40000,
13455
+ contextWindow: 262144,
13456
+ maxTokens: 16384,
13027
13457
  },
13028
13458
  "alibaba/qwen-3-30b": {
13029
13459
  id: "alibaba/qwen-3-30b",
@@ -13066,7 +13496,7 @@ export const MODELS = {
13066
13496
  provider: "vercel-ai-gateway",
13067
13497
  baseUrl: "https://ai-gateway.vercel.sh",
13068
13498
  reasoning: true,
13069
- input: ["text", "image"],
13499
+ input: ["text"],
13070
13500
  cost: {
13071
13501
  input: 1.3,
13072
13502
  output: 7.8,
@@ -13099,7 +13529,7 @@ export const MODELS = {
13099
13529
  api: "anthropic-messages",
13100
13530
  provider: "vercel-ai-gateway",
13101
13531
  baseUrl: "https://ai-gateway.vercel.sh",
13102
- reasoning: false,
13532
+ reasoning: true,
13103
13533
  input: ["text"],
13104
13534
  cost: {
13105
13535
  input: 1.5,
@@ -13133,7 +13563,7 @@ export const MODELS = {
13133
13563
  api: "anthropic-messages",
13134
13564
  provider: "vercel-ai-gateway",
13135
13565
  baseUrl: "https://ai-gateway.vercel.sh",
13136
- reasoning: false,
13566
+ reasoning: true,
13137
13567
  input: ["text"],
13138
13568
  cost: {
13139
13569
  input: 0.5,
@@ -13192,25 +13622,59 @@ export const MODELS = {
13192
13622
  cacheRead: 0.24,
13193
13623
  cacheWrite: 0,
13194
13624
  },
13195
- contextWindow: 262144,
13625
+ contextWindow: 262144,
13626
+ maxTokens: 32768,
13627
+ },
13628
+ "alibaba/qwen3-max-thinking": {
13629
+ id: "alibaba/qwen3-max-thinking",
13630
+ name: "Qwen 3 Max Thinking",
13631
+ api: "anthropic-messages",
13632
+ provider: "vercel-ai-gateway",
13633
+ baseUrl: "https://ai-gateway.vercel.sh",
13634
+ reasoning: true,
13635
+ input: ["text"],
13636
+ cost: {
13637
+ input: 1.2,
13638
+ output: 6,
13639
+ cacheRead: 0.24,
13640
+ cacheWrite: 0,
13641
+ },
13642
+ contextWindow: 256000,
13643
+ maxTokens: 65536,
13644
+ },
13645
+ "alibaba/qwen3-next-80b-a3b-instruct": {
13646
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
13647
+ name: "Qwen3 Next 80B A3B Instruct",
13648
+ api: "anthropic-messages",
13649
+ provider: "vercel-ai-gateway",
13650
+ baseUrl: "https://ai-gateway.vercel.sh",
13651
+ reasoning: false,
13652
+ input: ["text"],
13653
+ cost: {
13654
+ input: 0.15,
13655
+ output: 1.2,
13656
+ cacheRead: 0,
13657
+ cacheWrite: 0,
13658
+ },
13659
+ contextWindow: 131072,
13196
13660
  maxTokens: 32768,
13197
13661
  },
13198
- "alibaba/qwen3-max-thinking": {
13199
- id: "alibaba/qwen3-max-thinking",
13200
- name: "Qwen 3 Max Thinking",
13662
+ "alibaba/qwen3-next-80b-a3b-thinking": {
13663
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
13664
+ name: "Qwen3 Next 80B A3B Thinking",
13201
13665
  api: "anthropic-messages",
13202
13666
  provider: "vercel-ai-gateway",
13203
13667
  baseUrl: "https://ai-gateway.vercel.sh",
13204
13668
  reasoning: true,
13205
13669
  input: ["text"],
13206
13670
  cost: {
13207
- input: 1.2,
13208
- output: 6,
13209
- cacheRead: 0.24,
13671
+ input: 0.15,
13672
+ output: 1.2,
13673
+ cacheRead: 0,
13210
13674
  cacheWrite: 0,
13211
13675
  },
13212
- contextWindow: 256000,
13213
- maxTokens: 65536,
13676
+ contextWindow: 131072,
13677
+ maxTokens: 32768,
13214
13678
  },
13215
13679
  "alibaba/qwen3-vl-thinking": {
13216
13680
  id: "alibaba/qwen3-vl-thinking",
@@ -13304,7 +13768,7 @@ export const MODELS = {
13304
13768
  provider: "vercel-ai-gateway",
13305
13769
  baseUrl: "https://ai-gateway.vercel.sh",
13306
13770
  reasoning: true,
13307
- input: ["text", "image"],
13771
+ input: ["text"],
13308
13772
  cost: {
13309
13773
  input: 1.25,
13310
13774
  output: 3.75,
@@ -13314,6 +13778,23 @@ export const MODELS = {
13314
13778
  contextWindow: 991000,
13315
13779
  maxTokens: 64000,
13316
13780
  },
13781
+ "alibaba/qwen3.7-plus": {
13782
+ id: "alibaba/qwen3.7-plus",
13783
+ name: "Qwen 3.7 Plus",
13784
+ api: "anthropic-messages",
13785
+ provider: "vercel-ai-gateway",
13786
+ baseUrl: "https://ai-gateway.vercel.sh",
13787
+ reasoning: true,
13788
+ input: ["text", "image"],
13789
+ cost: {
13790
+ input: 0.39999999999999997,
13791
+ output: 1.5999999999999999,
13792
+ cacheRead: 0.08,
13793
+ cacheWrite: 0.5,
13794
+ },
13795
+ contextWindow: 1000000,
13796
+ maxTokens: 64000,
13797
+ },
13317
13798
  "anthropic/claude-3-haiku": {
13318
13799
  id: "anthropic/claude-3-haiku",
13319
13800
  name: "Claude 3 Haiku",
@@ -13441,7 +13922,7 @@ export const MODELS = {
13441
13922
  api: "anthropic-messages",
13442
13923
  provider: "vercel-ai-gateway",
13443
13924
  baseUrl: "https://ai-gateway.vercel.sh",
13444
- compat: { "forceAdaptiveThinking": true },
13925
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
13445
13926
  reasoning: true,
13446
13927
  thinkingLevelMap: { "xhigh": "xhigh" },
13447
13928
  input: ["text", "image"],
@@ -13460,7 +13941,7 @@ export const MODELS = {
13460
13941
  api: "anthropic-messages",
13461
13942
  provider: "vercel-ai-gateway",
13462
13943
  baseUrl: "https://ai-gateway.vercel.sh",
13463
- compat: { "forceAdaptiveThinking": true },
13944
+ compat: { "forceAdaptiveThinking": true, "supportsTemperature": false },
13464
13945
  reasoning: true,
13465
13946
  thinkingLevelMap: { "xhigh": "xhigh" },
13466
13947
  input: ["text", "image"],
@@ -13619,17 +14100,17 @@ export const MODELS = {
13619
14100
  reasoning: false,
13620
14101
  input: ["text"],
13621
14102
  cost: {
13622
- input: 0.77,
13623
- output: 0.77,
13624
- cacheRead: 0,
14103
+ input: 0.27,
14104
+ output: 1.12,
14105
+ cacheRead: 0.135,
13625
14106
  cacheWrite: 0,
13626
14107
  },
13627
14108
  contextWindow: 163840,
13628
- maxTokens: 16384,
14109
+ maxTokens: 163840,
13629
14110
  },
13630
14111
  "deepseek/deepseek-v3.1": {
13631
14112
  id: "deepseek/deepseek-v3.1",
13632
- name: "DeepSeek-V3.1",
14113
+ name: "DeepSeek V3.1",
13633
14114
  api: "anthropic-messages",
13634
14115
  provider: "vercel-ai-gateway",
13635
14116
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13667,8 +14148,8 @@ export const MODELS = {
13667
14148
  api: "anthropic-messages",
13668
14149
  provider: "vercel-ai-gateway",
13669
14150
  baseUrl: "https://ai-gateway.vercel.sh",
13670
- reasoning: false,
13671
- input: ["text"],
14151
+ reasoning: true,
14152
+ input: ["text", "image"],
13672
14153
  cost: {
13673
14154
  input: 0.28,
13674
14155
  output: 0.42,
@@ -13684,8 +14165,8 @@ export const MODELS = {
13684
14165
  api: "anthropic-messages",
13685
14166
  provider: "vercel-ai-gateway",
13686
14167
  baseUrl: "https://ai-gateway.vercel.sh",
13687
- reasoning: false,
13688
- input: ["text"],
14168
+ reasoning: true,
14169
+ input: ["text", "image"],
13689
14170
  cost: {
13690
14171
  input: 0.62,
13691
14172
  output: 1.85,
@@ -13702,7 +14183,7 @@ export const MODELS = {
13702
14183
  provider: "vercel-ai-gateway",
13703
14184
  baseUrl: "https://ai-gateway.vercel.sh",
13704
14185
  reasoning: true,
13705
- input: ["text"],
14186
+ input: ["text", "image"],
13706
14187
  cost: {
13707
14188
  input: 0.14,
13708
14189
  output: 0.28,
@@ -13922,12 +14403,12 @@ export const MODELS = {
13922
14403
  api: "anthropic-messages",
13923
14404
  provider: "vercel-ai-gateway",
13924
14405
  baseUrl: "https://ai-gateway.vercel.sh",
13925
- reasoning: false,
14406
+ reasoning: true,
13926
14407
  input: ["text", "image"],
13927
14408
  cost: {
13928
- input: 0.13,
13929
- output: 0.39999999999999997,
13930
- cacheRead: 0,
14409
+ input: 0.15,
14410
+ output: 0.6,
14411
+ cacheRead: 0.015,
13931
14412
  cacheWrite: 0,
13932
14413
  },
13933
14414
  contextWindow: 262144,
@@ -14229,7 +14710,7 @@ export const MODELS = {
14229
14710
  provider: "vercel-ai-gateway",
14230
14711
  baseUrl: "https://ai-gateway.vercel.sh",
14231
14712
  reasoning: true,
14232
- input: ["text", "image"],
14713
+ input: ["text"],
14233
14714
  cost: {
14234
14715
  input: 0.3,
14235
14716
  output: 1.2,
@@ -14246,7 +14727,7 @@ export const MODELS = {
14246
14727
  provider: "vercel-ai-gateway",
14247
14728
  baseUrl: "https://ai-gateway.vercel.sh",
14248
14729
  reasoning: true,
14249
- input: ["text", "image"],
14730
+ input: ["text"],
14250
14731
  cost: {
14251
14732
  input: 0.6,
14252
14733
  output: 2.4,
@@ -14256,6 +14737,23 @@ export const MODELS = {
14256
14737
  contextWindow: 204800,
14257
14738
  maxTokens: 131100,
14258
14739
  },
14740
+ "minimax/minimax-m3": {
14741
+ id: "minimax/minimax-m3",
14742
+ name: "MiniMax M3",
14743
+ api: "anthropic-messages",
14744
+ provider: "vercel-ai-gateway",
14745
+ baseUrl: "https://ai-gateway.vercel.sh",
14746
+ reasoning: true,
14747
+ input: ["text", "image"],
14748
+ cost: {
14749
+ input: 0.3,
14750
+ output: 1.2,
14751
+ cacheRead: 0.06,
14752
+ cacheWrite: 0,
14753
+ },
14754
+ contextWindow: 1000000,
14755
+ maxTokens: 1000000,
14756
+ },
14259
14757
  "mistral/codestral": {
14260
14758
  id: "mistral/codestral",
14261
14759
  name: "Mistral Codestral",
@@ -14392,6 +14890,23 @@ export const MODELS = {
14392
14890
  contextWindow: 256000,
14393
14891
  maxTokens: 256000,
14394
14892
  },
14893
+ "mistral/mistral-nemo": {
14894
+ id: "mistral/mistral-nemo",
14895
+ name: "Mistral Nemo 12B",
14896
+ api: "anthropic-messages",
14897
+ provider: "vercel-ai-gateway",
14898
+ baseUrl: "https://ai-gateway.vercel.sh",
14899
+ reasoning: false,
14900
+ input: ["text"],
14901
+ cost: {
14902
+ input: 0.02,
14903
+ output: 0.04,
14904
+ cacheRead: 0,
14905
+ cacheWrite: 0,
14906
+ },
14907
+ contextWindow: 131072,
14908
+ maxTokens: 131072,
14909
+ },
14395
14910
  "mistral/mistral-small": {
14396
14911
  id: "mistral/mistral-small",
14397
14912
  name: "Mistral Small",
@@ -14545,6 +15060,40 @@ export const MODELS = {
14545
15060
  contextWindow: 262000,
14546
15061
  maxTokens: 262000,
14547
15062
  },
15063
+ "nvidia/nemotron-3-super-120b-a12b": {
15064
+ id: "nvidia/nemotron-3-super-120b-a12b",
15065
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
15066
+ api: "anthropic-messages",
15067
+ provider: "vercel-ai-gateway",
15068
+ baseUrl: "https://ai-gateway.vercel.sh",
15069
+ reasoning: true,
15070
+ input: ["text"],
15071
+ cost: {
15072
+ input: 0.15,
15073
+ output: 0.65,
15074
+ cacheRead: 0,
15075
+ cacheWrite: 0,
15076
+ },
15077
+ contextWindow: 256000,
15078
+ maxTokens: 32000,
15079
+ },
15080
+ "nvidia/nemotron-3-ultra-550b-a55b": {
15081
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
15082
+ name: "Nemotron 3 Ultra",
15083
+ api: "anthropic-messages",
15084
+ provider: "vercel-ai-gateway",
15085
+ baseUrl: "https://ai-gateway.vercel.sh",
15086
+ reasoning: true,
15087
+ input: ["text"],
15088
+ cost: {
15089
+ input: 0.5,
15090
+ output: 2.5,
15091
+ cacheRead: 0.15,
15092
+ cacheWrite: 0,
15093
+ },
15094
+ contextWindow: 262144,
15095
+ maxTokens: 65000,
15096
+ },
14548
15097
  "nvidia/nemotron-nano-12b-v2-vl": {
14549
15098
  id: "nvidia/nemotron-nano-12b-v2-vl",
14550
15099
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -14722,7 +15271,7 @@ export const MODELS = {
14722
15271
  provider: "vercel-ai-gateway",
14723
15272
  baseUrl: "https://ai-gateway.vercel.sh",
14724
15273
  reasoning: true,
14725
- input: ["text"],
15274
+ input: ["text", "image"],
14726
15275
  cost: {
14727
15276
  input: 1.25,
14728
15277
  output: 10,
@@ -15084,6 +15633,23 @@ export const MODELS = {
15084
15633
  contextWindow: 1000000,
15085
15634
  maxTokens: 128000,
15086
15635
  },
15636
+ "openai/gpt-oss-120b": {
15637
+ id: "openai/gpt-oss-120b",
15638
+ name: "GPT OSS 120B",
15639
+ api: "anthropic-messages",
15640
+ provider: "vercel-ai-gateway",
15641
+ baseUrl: "https://ai-gateway.vercel.sh",
15642
+ reasoning: true,
15643
+ input: ["text"],
15644
+ cost: {
15645
+ input: 0.35,
15646
+ output: 0.75,
15647
+ cacheRead: 0.25,
15648
+ cacheWrite: 0,
15649
+ },
15650
+ contextWindow: 131072,
15651
+ maxTokens: 131000,
15652
+ },
15087
15653
  "openai/gpt-oss-20b": {
15088
15654
  id: "openai/gpt-oss-20b",
15089
15655
  name: "GPT OSS 20B",
@@ -15254,6 +15820,40 @@ export const MODELS = {
15254
15820
  contextWindow: 200000,
15255
15821
  maxTokens: 8000,
15256
15822
  },
15823
+ "stepfun/step-3.5-flash": {
15824
+ id: "stepfun/step-3.5-flash",
15825
+ name: "StepFun 3.5 Flash",
15826
+ api: "anthropic-messages",
15827
+ provider: "vercel-ai-gateway",
15828
+ baseUrl: "https://ai-gateway.vercel.sh",
15829
+ reasoning: true,
15830
+ input: ["text"],
15831
+ cost: {
15832
+ input: 0.09,
15833
+ output: 0.3,
15834
+ cacheRead: 0,
15835
+ cacheWrite: 0.02,
15836
+ },
15837
+ contextWindow: 262114,
15838
+ maxTokens: 262114,
15839
+ },
15840
+ "stepfun/step-3.7-flash": {
15841
+ id: "stepfun/step-3.7-flash",
15842
+ name: "Step 3.7 Flash",
15843
+ api: "anthropic-messages",
15844
+ provider: "vercel-ai-gateway",
15845
+ baseUrl: "https://ai-gateway.vercel.sh",
15846
+ reasoning: true,
15847
+ input: ["text", "image"],
15848
+ cost: {
15849
+ input: 0.19999999999999998,
15850
+ output: 1.15,
15851
+ cacheRead: 0.04,
15852
+ cacheWrite: 0,
15853
+ },
15854
+ contextWindow: 256000,
15855
+ maxTokens: 256000,
15856
+ },
15257
15857
  "xai/grok-4.1-fast-non-reasoning": {
15258
15858
  id: "xai/grok-4.1-fast-non-reasoning",
15259
15859
  name: "Grok 4.1 Fast Non-Reasoning",
@@ -15532,7 +16132,7 @@ export const MODELS = {
15532
16132
  api: "anthropic-messages",
15533
16133
  provider: "vercel-ai-gateway",
15534
16134
  baseUrl: "https://ai-gateway.vercel.sh",
15535
- reasoning: false,
16135
+ reasoning: true,
15536
16136
  input: ["text", "image"],
15537
16137
  cost: {
15538
16138
  input: 0.6,
@@ -15686,7 +16286,7 @@ export const MODELS = {
15686
16286
  provider: "vercel-ai-gateway",
15687
16287
  baseUrl: "https://ai-gateway.vercel.sh",
15688
16288
  reasoning: true,
15689
- input: ["text"],
16289
+ input: ["text", "image"],
15690
16290
  cost: {
15691
16291
  input: 1.4,
15692
16292
  output: 4.4,
@@ -16241,5 +16841,97 @@ export const MODELS = {
16241
16841
  maxTokens: 131072,
16242
16842
  },
16243
16843
  },
16844
+ "zai-coding-cn": {
16845
+ "glm-4.5-air": {
16846
+ id: "glm-4.5-air",
16847
+ name: "GLM-4.5-Air",
16848
+ api: "openai-completions",
16849
+ provider: "zai-coding-cn",
16850
+ baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
16851
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai" },
16852
+ reasoning: true,
16853
+ input: ["text"],
16854
+ cost: {
16855
+ input: 0,
16856
+ output: 0,
16857
+ cacheRead: 0,
16858
+ cacheWrite: 0,
16859
+ },
16860
+ contextWindow: 131072,
16861
+ maxTokens: 98304,
16862
+ },
16863
+ "glm-4.7": {
16864
+ id: "glm-4.7",
16865
+ name: "GLM-4.7",
16866
+ api: "openai-completions",
16867
+ provider: "zai-coding-cn",
16868
+ baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
16869
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16870
+ reasoning: true,
16871
+ input: ["text"],
16872
+ cost: {
16873
+ input: 0,
16874
+ output: 0,
16875
+ cacheRead: 0,
16876
+ cacheWrite: 0,
16877
+ },
16878
+ contextWindow: 204800,
16879
+ maxTokens: 131072,
16880
+ },
16881
+ "glm-5-turbo": {
16882
+ id: "glm-5-turbo",
16883
+ name: "GLM-5-Turbo",
16884
+ api: "openai-completions",
16885
+ provider: "zai-coding-cn",
16886
+ baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
16887
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16888
+ reasoning: true,
16889
+ input: ["text"],
16890
+ cost: {
16891
+ input: 0,
16892
+ output: 0,
16893
+ cacheRead: 0,
16894
+ cacheWrite: 0,
16895
+ },
16896
+ contextWindow: 200000,
16897
+ maxTokens: 131072,
16898
+ },
16899
+ "glm-5.1": {
16900
+ id: "glm-5.1",
16901
+ name: "GLM-5.1",
16902
+ api: "openai-completions",
16903
+ provider: "zai-coding-cn",
16904
+ baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
16905
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16906
+ reasoning: true,
16907
+ input: ["text"],
16908
+ cost: {
16909
+ input: 0,
16910
+ output: 0,
16911
+ cacheRead: 0,
16912
+ cacheWrite: 0,
16913
+ },
16914
+ contextWindow: 200000,
16915
+ maxTokens: 131072,
16916
+ },
16917
+ "glm-5v-turbo": {
16918
+ id: "glm-5v-turbo",
16919
+ name: "GLM-5V-Turbo",
16920
+ api: "openai-completions",
16921
+ provider: "zai-coding-cn",
16922
+ baseUrl: "https://open.bigmodel.cn/api/coding/paas/v4",
16923
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16924
+ reasoning: true,
16925
+ input: ["text", "image"],
16926
+ cost: {
16927
+ input: 0,
16928
+ output: 0,
16929
+ cacheRead: 0,
16930
+ cacheWrite: 0,
16931
+ },
16932
+ contextWindow: 200000,
16933
+ maxTokens: 131072,
16934
+ },
16935
+ },
16244
16936
  };
16245
16937
  //# sourceMappingURL=models.generated.js.map