@draht/ai 2026.4.5 → 2026.4.23

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 (40) hide show
  1. package/README.md +21 -12
  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.generated.d.ts +683 -357
  6. package/dist/models.generated.d.ts.map +1 -1
  7. package/dist/models.generated.js +897 -586
  8. package/dist/models.generated.js.map +1 -1
  9. package/dist/providers/amazon-bedrock.d.ts.map +1 -1
  10. package/dist/providers/amazon-bedrock.js +37 -9
  11. package/dist/providers/amazon-bedrock.js.map +1 -1
  12. package/dist/providers/anthropic.d.ts.map +1 -1
  13. package/dist/providers/anthropic.js +38 -16
  14. package/dist/providers/anthropic.js.map +1 -1
  15. package/dist/providers/google-gemini-cli.d.ts.map +1 -1
  16. package/dist/providers/google-gemini-cli.js +2 -2
  17. package/dist/providers/google-gemini-cli.js.map +1 -1
  18. package/dist/providers/google-vertex.d.ts.map +1 -1
  19. package/dist/providers/google-vertex.js +1 -1
  20. package/dist/providers/google-vertex.js.map +1 -1
  21. package/dist/providers/google.d.ts.map +1 -1
  22. package/dist/providers/google.js +29 -4
  23. package/dist/providers/google.js.map +1 -1
  24. package/dist/providers/openai-completions.d.ts.map +1 -1
  25. package/dist/providers/openai-completions.js +13 -6
  26. package/dist/providers/openai-completions.js.map +1 -1
  27. package/dist/providers/openai-responses-shared.d.ts.map +1 -1
  28. package/dist/providers/openai-responses-shared.js +8 -1
  29. package/dist/providers/openai-responses-shared.js.map +1 -1
  30. package/dist/providers/openai-responses.d.ts.map +1 -1
  31. package/dist/providers/openai-responses.js +2 -2
  32. package/dist/providers/openai-responses.js.map +1 -1
  33. package/dist/types.d.ts +62 -5
  34. package/dist/types.d.ts.map +1 -1
  35. package/dist/types.js.map +1 -1
  36. package/dist/utils/overflow.d.ts +3 -2
  37. package/dist/utils/overflow.d.ts.map +1 -1
  38. package/dist/utils/overflow.js +24 -11
  39. package/dist/utils/overflow.js.map +1 -1
  40. 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",
@@ -410,6 +427,23 @@ export const MODELS = {
410
427
  contextWindow: 1000000,
411
428
  maxTokens: 128000,
412
429
  },
430
+ "eu.anthropic.claude-opus-4-7": {
431
+ id: "eu.anthropic.claude-opus-4-7",
432
+ name: "Claude Opus 4.7 (EU)",
433
+ api: "bedrock-converse-stream",
434
+ provider: "amazon-bedrock",
435
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
436
+ reasoning: true,
437
+ input: ["text", "image"],
438
+ cost: {
439
+ input: 5,
440
+ output: 25,
441
+ cacheRead: 0.5,
442
+ cacheWrite: 6.25,
443
+ },
444
+ contextWindow: 1000000,
445
+ maxTokens: 128000,
446
+ },
413
447
  "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
414
448
  id: "eu.anthropic.claude-sonnet-4-20250514-v1:0",
415
449
  name: "Claude Sonnet 4 (EU)",
@@ -512,6 +546,23 @@ export const MODELS = {
512
546
  contextWindow: 1000000,
513
547
  maxTokens: 128000,
514
548
  },
549
+ "global.anthropic.claude-opus-4-7": {
550
+ id: "global.anthropic.claude-opus-4-7",
551
+ name: "Claude Opus 4.7 (Global)",
552
+ api: "bedrock-converse-stream",
553
+ provider: "amazon-bedrock",
554
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
555
+ reasoning: true,
556
+ input: ["text", "image"],
557
+ cost: {
558
+ input: 5,
559
+ output: 25,
560
+ cacheRead: 0.5,
561
+ cacheWrite: 6.25,
562
+ },
563
+ contextWindow: 1000000,
564
+ maxTokens: 128000,
565
+ },
515
566
  "global.anthropic.claude-sonnet-4-20250514-v1:0": {
516
567
  id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
517
568
  name: "Claude Sonnet 4 (Global)",
@@ -1209,6 +1260,23 @@ export const MODELS = {
1209
1260
  contextWindow: 131072,
1210
1261
  maxTokens: 65536,
1211
1262
  },
1263
+ "qwen.qwen3-coder-next": {
1264
+ id: "qwen.qwen3-coder-next",
1265
+ name: "Qwen3 Coder Next",
1266
+ api: "bedrock-converse-stream",
1267
+ provider: "amazon-bedrock",
1268
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1269
+ reasoning: true,
1270
+ input: ["text"],
1271
+ cost: {
1272
+ input: 0.22,
1273
+ output: 1.8,
1274
+ cacheRead: 0,
1275
+ cacheWrite: 0,
1276
+ },
1277
+ contextWindow: 131072,
1278
+ maxTokens: 65536,
1279
+ },
1212
1280
  "qwen.qwen3-next-80b-a3b": {
1213
1281
  id: "qwen.qwen3-next-80b-a3b",
1214
1282
  name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
@@ -1328,6 +1396,23 @@ export const MODELS = {
1328
1396
  contextWindow: 1000000,
1329
1397
  maxTokens: 128000,
1330
1398
  },
1399
+ "us.anthropic.claude-opus-4-7": {
1400
+ id: "us.anthropic.claude-opus-4-7",
1401
+ name: "Claude Opus 4.7 (US)",
1402
+ api: "bedrock-converse-stream",
1403
+ provider: "amazon-bedrock",
1404
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1405
+ reasoning: true,
1406
+ input: ["text", "image"],
1407
+ cost: {
1408
+ input: 5,
1409
+ output: 25,
1410
+ cacheRead: 0.5,
1411
+ cacheWrite: 6.25,
1412
+ },
1413
+ contextWindow: 1000000,
1414
+ maxTokens: 128000,
1415
+ },
1331
1416
  "us.anthropic.claude-sonnet-4-20250514-v1:0": {
1332
1417
  id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
1333
1418
  name: "Claude Sonnet 4 (US)",
@@ -1755,6 +1840,23 @@ export const MODELS = {
1755
1840
  contextWindow: 1000000,
1756
1841
  maxTokens: 128000,
1757
1842
  },
1843
+ "claude-opus-4-7": {
1844
+ id: "claude-opus-4-7",
1845
+ name: "Claude Opus 4.7",
1846
+ api: "anthropic-messages",
1847
+ provider: "anthropic",
1848
+ baseUrl: "https://api.anthropic.com",
1849
+ reasoning: true,
1850
+ input: ["text", "image"],
1851
+ cost: {
1852
+ input: 5,
1853
+ output: 25,
1854
+ cacheRead: 0.5,
1855
+ cacheWrite: 6.25,
1856
+ },
1857
+ contextWindow: 1000000,
1858
+ maxTokens: 128000,
1859
+ },
1758
1860
  "claude-sonnet-4-0": {
1759
1861
  id: "claude-sonnet-4-0",
1760
1862
  name: "Claude Sonnet 4 (latest)",
@@ -1842,23 +1944,6 @@ export const MODELS = {
1842
1944
  },
1843
1945
  },
1844
1946
  "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
1947
  "gpt-4": {
1863
1948
  id: "gpt-4",
1864
1949
  name: "GPT-4",
@@ -2665,6 +2750,24 @@ export const MODELS = {
2665
2750
  contextWindow: 1000000,
2666
2751
  maxTokens: 64000,
2667
2752
  },
2753
+ "claude-opus-4.7": {
2754
+ id: "claude-opus-4.7",
2755
+ name: "Claude Opus 4.7",
2756
+ api: "anthropic-messages",
2757
+ provider: "github-copilot",
2758
+ baseUrl: "https://api.individual.githubcopilot.com",
2759
+ 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" },
2760
+ reasoning: true,
2761
+ input: ["text", "image"],
2762
+ cost: {
2763
+ input: 0,
2764
+ output: 0,
2765
+ cacheRead: 0,
2766
+ cacheWrite: 0,
2767
+ },
2768
+ contextWindow: 144000,
2769
+ maxTokens: 64000,
2770
+ },
2668
2771
  "claude-sonnet-4": {
2669
2772
  id: "claude-sonnet-4",
2670
2773
  name: "Claude Sonnet 4",
@@ -3015,7 +3118,7 @@ export const MODELS = {
3015
3118
  },
3016
3119
  "gpt-5.4-mini": {
3017
3120
  id: "gpt-5.4-mini",
3018
- name: "GPT-5.4 mini",
3121
+ name: "GPT-5.4 Mini",
3019
3122
  api: "openai-responses",
3020
3123
  provider: "github-copilot",
3021
3124
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3148,7 +3251,7 @@ export const MODELS = {
3148
3251
  cost: {
3149
3252
  input: 0.3,
3150
3253
  output: 2.5,
3151
- cacheRead: 0.075,
3254
+ cacheRead: 0.03,
3152
3255
  cacheWrite: 0,
3153
3256
  },
3154
3257
  contextWindow: 1048576,
@@ -3267,7 +3370,7 @@ export const MODELS = {
3267
3370
  cost: {
3268
3371
  input: 1.25,
3269
3372
  output: 10,
3270
- cacheRead: 0.31,
3373
+ cacheRead: 0.125,
3271
3374
  cacheWrite: 0,
3272
3375
  },
3273
3376
  contextWindow: 1048576,
@@ -3477,8 +3580,8 @@ export const MODELS = {
3477
3580
  contextWindow: 131072,
3478
3581
  maxTokens: 8192,
3479
3582
  },
3480
- "gemma-4-26b": {
3481
- id: "gemma-4-26b",
3583
+ "gemma-4-26b-a4b-it": {
3584
+ id: "gemma-4-26b-a4b-it",
3482
3585
  name: "Gemma 4 26B",
3483
3586
  api: "google-generative-ai",
3484
3587
  provider: "google",
@@ -3494,8 +3597,8 @@ export const MODELS = {
3494
3597
  contextWindow: 256000,
3495
3598
  maxTokens: 8192,
3496
3599
  },
3497
- "gemma-4-31b": {
3498
- id: "gemma-4-31b",
3600
+ "gemma-4-31b-it": {
3601
+ id: "gemma-4-31b-it",
3499
3602
  name: "Gemma 4 31B",
3500
3603
  api: "google-generative-ai",
3501
3604
  provider: "google",
@@ -3976,6 +4079,23 @@ export const MODELS = {
3976
4079
  contextWindow: 1048576,
3977
4080
  maxTokens: 65536,
3978
4081
  },
4082
+ "gemini-3.1-pro-preview-customtools": {
4083
+ id: "gemini-3.1-pro-preview-customtools",
4084
+ name: "Gemini 3.1 Pro Preview Custom Tools (Vertex)",
4085
+ api: "google-vertex",
4086
+ provider: "google-vertex",
4087
+ baseUrl: "https://{location}-aiplatform.googleapis.com",
4088
+ reasoning: true,
4089
+ input: ["text", "image"],
4090
+ cost: {
4091
+ input: 2,
4092
+ output: 12,
4093
+ cacheRead: 0.2,
4094
+ cacheWrite: 0,
4095
+ },
4096
+ contextWindow: 1048576,
4097
+ maxTokens: 65536,
4098
+ },
3979
4099
  },
3980
4100
  "groq": {
3981
4101
  "deepseek-r1-distill-llama-70b": {
@@ -4322,6 +4442,24 @@ export const MODELS = {
4322
4442
  contextWindow: 204800,
4323
4443
  maxTokens: 131072,
4324
4444
  },
4445
+ "MiniMaxAI/MiniMax-M2.7": {
4446
+ id: "MiniMaxAI/MiniMax-M2.7",
4447
+ name: "MiniMax-M2.7",
4448
+ api: "openai-completions",
4449
+ provider: "huggingface",
4450
+ baseUrl: "https://router.huggingface.co/v1",
4451
+ compat: { "supportsDeveloperRole": false },
4452
+ reasoning: true,
4453
+ input: ["text"],
4454
+ cost: {
4455
+ input: 0.3,
4456
+ output: 1.2,
4457
+ cacheRead: 0.06,
4458
+ cacheWrite: 0,
4459
+ },
4460
+ contextWindow: 204800,
4461
+ maxTokens: 131072,
4462
+ },
4325
4463
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4326
4464
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4327
4465
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4556,6 +4694,24 @@ export const MODELS = {
4556
4694
  contextWindow: 262144,
4557
4695
  maxTokens: 262144,
4558
4696
  },
4697
+ "moonshotai/Kimi-K2.6": {
4698
+ id: "moonshotai/Kimi-K2.6",
4699
+ name: "Kimi-K2.6",
4700
+ api: "openai-completions",
4701
+ provider: "huggingface",
4702
+ baseUrl: "https://router.huggingface.co/v1",
4703
+ compat: { "supportsDeveloperRole": false },
4704
+ reasoning: true,
4705
+ input: ["text", "image"],
4706
+ cost: {
4707
+ input: 0.95,
4708
+ output: 4,
4709
+ cacheRead: 0.16,
4710
+ cacheWrite: 0,
4711
+ },
4712
+ contextWindow: 262144,
4713
+ maxTokens: 262144,
4714
+ },
4559
4715
  "zai-org/GLM-4.7": {
4560
4716
  id: "zai-org/GLM-4.7",
4561
4717
  name: "GLM-4.7",
@@ -4610,6 +4766,24 @@ export const MODELS = {
4610
4766
  contextWindow: 202752,
4611
4767
  maxTokens: 131072,
4612
4768
  },
4769
+ "zai-org/GLM-5.1": {
4770
+ id: "zai-org/GLM-5.1",
4771
+ name: "GLM-5.1",
4772
+ api: "openai-completions",
4773
+ provider: "huggingface",
4774
+ baseUrl: "https://router.huggingface.co/v1",
4775
+ compat: { "supportsDeveloperRole": false },
4776
+ reasoning: true,
4777
+ input: ["text"],
4778
+ cost: {
4779
+ input: 1,
4780
+ output: 3.2,
4781
+ cacheRead: 0.2,
4782
+ cacheWrite: 0,
4783
+ },
4784
+ contextWindow: 202752,
4785
+ maxTokens: 131072,
4786
+ },
4613
4787
  },
4614
4788
  "kimi-coding": {
4615
4789
  "k2p5": {
@@ -4629,14 +4803,14 @@ export const MODELS = {
4629
4803
  contextWindow: 262144,
4630
4804
  maxTokens: 32768,
4631
4805
  },
4632
- "kimi-k2-thinking": {
4633
- id: "kimi-k2-thinking",
4634
- name: "Kimi K2 Thinking",
4806
+ "k2p6": {
4807
+ id: "k2p6",
4808
+ name: "Kimi K2.6",
4635
4809
  api: "anthropic-messages",
4636
4810
  provider: "kimi-coding",
4637
4811
  baseUrl: "https://api.kimi.com/coding",
4638
4812
  reasoning: true,
4639
- input: ["text"],
4813
+ input: ["text", "image"],
4640
4814
  cost: {
4641
4815
  input: 0,
4642
4816
  output: 0,
@@ -4646,62 +4820,28 @@ export const MODELS = {
4646
4820
  contextWindow: 262144,
4647
4821
  maxTokens: 32768,
4648
4822
  },
4649
- },
4650
- "minimax": {
4651
- "MiniMax-M2": {
4652
- id: "MiniMax-M2",
4653
- name: "MiniMax-M2",
4654
- api: "anthropic-messages",
4655
- provider: "minimax",
4656
- baseUrl: "https://api.minimax.io/anthropic",
4657
- reasoning: true,
4658
- input: ["text"],
4659
- cost: {
4660
- input: 0.3,
4661
- output: 1.2,
4662
- cacheRead: 0,
4663
- cacheWrite: 0,
4664
- },
4665
- contextWindow: 204800,
4666
- maxTokens: 131072,
4667
- },
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",
4823
+ "kimi-k2-thinking": {
4824
+ id: "kimi-k2-thinking",
4825
+ name: "Kimi K2 Thinking",
4688
4826
  api: "anthropic-messages",
4689
- provider: "minimax",
4690
- baseUrl: "https://api.minimax.io/anthropic",
4827
+ provider: "kimi-coding",
4828
+ baseUrl: "https://api.kimi.com/coding",
4691
4829
  reasoning: true,
4692
4830
  input: ["text"],
4693
4831
  cost: {
4694
- input: 0.6,
4695
- output: 2.4,
4832
+ input: 0,
4833
+ output: 0,
4696
4834
  cacheRead: 0,
4697
4835
  cacheWrite: 0,
4698
4836
  },
4699
- contextWindow: 204800,
4700
- maxTokens: 131072,
4837
+ contextWindow: 262144,
4838
+ maxTokens: 32768,
4701
4839
  },
4702
- "MiniMax-M2.5": {
4703
- id: "MiniMax-M2.5",
4704
- name: "MiniMax-M2.5",
4840
+ },
4841
+ "minimax": {
4842
+ "MiniMax-M2.7": {
4843
+ id: "MiniMax-M2.7",
4844
+ name: "MiniMax-M2.7",
4705
4845
  api: "anthropic-messages",
4706
4846
  provider: "minimax",
4707
4847
  baseUrl: "https://api.minimax.io/anthropic",
@@ -4710,15 +4850,15 @@ export const MODELS = {
4710
4850
  cost: {
4711
4851
  input: 0.3,
4712
4852
  output: 1.2,
4713
- cacheRead: 0.03,
4853
+ cacheRead: 0.06,
4714
4854
  cacheWrite: 0.375,
4715
4855
  },
4716
4856
  contextWindow: 204800,
4717
4857
  maxTokens: 131072,
4718
4858
  },
4719
- "MiniMax-M2.5-highspeed": {
4720
- id: "MiniMax-M2.5-highspeed",
4721
- name: "MiniMax-M2.5-highspeed",
4859
+ "MiniMax-M2.7-highspeed": {
4860
+ id: "MiniMax-M2.7-highspeed",
4861
+ name: "MiniMax-M2.7-highspeed",
4722
4862
  api: "anthropic-messages",
4723
4863
  provider: "minimax",
4724
4864
  baseUrl: "https://api.minimax.io/anthropic",
@@ -4733,12 +4873,14 @@ export const MODELS = {
4733
4873
  contextWindow: 204800,
4734
4874
  maxTokens: 131072,
4735
4875
  },
4876
+ },
4877
+ "minimax-cn": {
4736
4878
  "MiniMax-M2.7": {
4737
4879
  id: "MiniMax-M2.7",
4738
4880
  name: "MiniMax-M2.7",
4739
4881
  api: "anthropic-messages",
4740
- provider: "minimax",
4741
- baseUrl: "https://api.minimax.io/anthropic",
4882
+ provider: "minimax-cn",
4883
+ baseUrl: "https://api.minimaxi.com/anthropic",
4742
4884
  reasoning: true,
4743
4885
  input: ["text"],
4744
4886
  cost: {
@@ -4754,129 +4896,8 @@ export const MODELS = {
4754
4896
  id: "MiniMax-M2.7-highspeed",
4755
4897
  name: "MiniMax-M2.7-highspeed",
4756
4898
  api: "anthropic-messages",
4757
- provider: "minimax",
4758
- baseUrl: "https://api.minimax.io/anthropic",
4759
- reasoning: true,
4760
- input: ["text"],
4761
- cost: {
4762
- input: 0.6,
4763
- output: 2.4,
4764
- cacheRead: 0.06,
4765
- cacheWrite: 0.375,
4766
- },
4767
- contextWindow: 204800,
4768
- maxTokens: 131072,
4769
- },
4770
- },
4771
- "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
- "MiniMax-M2.7": {
4858
- id: "MiniMax-M2.7",
4859
- name: "MiniMax-M2.7",
4860
- api: "anthropic-messages",
4861
- provider: "minimax-cn",
4862
- baseUrl: "https://api.minimaxi.com/anthropic",
4863
- reasoning: true,
4864
- input: ["text"],
4865
- cost: {
4866
- input: 0.3,
4867
- output: 1.2,
4868
- cacheRead: 0.06,
4869
- cacheWrite: 0.375,
4870
- },
4871
- contextWindow: 204800,
4872
- maxTokens: 131072,
4873
- },
4874
- "MiniMax-M2.7-highspeed": {
4875
- id: "MiniMax-M2.7-highspeed",
4876
- name: "MiniMax-M2.7-highspeed",
4877
- api: "anthropic-messages",
4878
- provider: "minimax-cn",
4879
- baseUrl: "https://api.minimaxi.com/anthropic",
4899
+ provider: "minimax-cn",
4900
+ baseUrl: "https://api.minimaxi.com/anthropic",
4880
4901
  reasoning: true,
4881
4902
  input: ["text"],
4882
4903
  cost: {
@@ -4893,9 +4914,9 @@ export const MODELS = {
4893
4914
  "codestral-latest": {
4894
4915
  id: "codestral-latest",
4895
4916
  name: "Codestral (latest)",
4896
- api: "mistral-conversations",
4917
+ api: "openai-completions",
4897
4918
  provider: "mistral",
4898
- baseUrl: "https://api.mistral.ai",
4919
+ baseUrl: "https://api.mistral.ai/v1",
4899
4920
  reasoning: false,
4900
4921
  input: ["text"],
4901
4922
  cost: {
@@ -4910,9 +4931,9 @@ export const MODELS = {
4910
4931
  "devstral-2512": {
4911
4932
  id: "devstral-2512",
4912
4933
  name: "Devstral 2",
4913
- api: "mistral-conversations",
4934
+ api: "openai-completions",
4914
4935
  provider: "mistral",
4915
- baseUrl: "https://api.mistral.ai",
4936
+ baseUrl: "https://api.mistral.ai/v1",
4916
4937
  reasoning: false,
4917
4938
  input: ["text"],
4918
4939
  cost: {
@@ -4927,9 +4948,9 @@ export const MODELS = {
4927
4948
  "devstral-medium-2507": {
4928
4949
  id: "devstral-medium-2507",
4929
4950
  name: "Devstral Medium",
4930
- api: "mistral-conversations",
4951
+ api: "openai-completions",
4931
4952
  provider: "mistral",
4932
- baseUrl: "https://api.mistral.ai",
4953
+ baseUrl: "https://api.mistral.ai/v1",
4933
4954
  reasoning: false,
4934
4955
  input: ["text"],
4935
4956
  cost: {
@@ -4944,9 +4965,9 @@ export const MODELS = {
4944
4965
  "devstral-medium-latest": {
4945
4966
  id: "devstral-medium-latest",
4946
4967
  name: "Devstral 2 (latest)",
4947
- api: "mistral-conversations",
4968
+ api: "openai-completions",
4948
4969
  provider: "mistral",
4949
- baseUrl: "https://api.mistral.ai",
4970
+ baseUrl: "https://api.mistral.ai/v1",
4950
4971
  reasoning: false,
4951
4972
  input: ["text"],
4952
4973
  cost: {
@@ -4961,9 +4982,9 @@ export const MODELS = {
4961
4982
  "devstral-small-2505": {
4962
4983
  id: "devstral-small-2505",
4963
4984
  name: "Devstral Small 2505",
4964
- api: "mistral-conversations",
4985
+ api: "openai-completions",
4965
4986
  provider: "mistral",
4966
- baseUrl: "https://api.mistral.ai",
4987
+ baseUrl: "https://api.mistral.ai/v1",
4967
4988
  reasoning: false,
4968
4989
  input: ["text"],
4969
4990
  cost: {
@@ -4978,9 +4999,9 @@ export const MODELS = {
4978
4999
  "devstral-small-2507": {
4979
5000
  id: "devstral-small-2507",
4980
5001
  name: "Devstral Small",
4981
- api: "mistral-conversations",
5002
+ api: "openai-completions",
4982
5003
  provider: "mistral",
4983
- baseUrl: "https://api.mistral.ai",
5004
+ baseUrl: "https://api.mistral.ai/v1",
4984
5005
  reasoning: false,
4985
5006
  input: ["text"],
4986
5007
  cost: {
@@ -4995,9 +5016,9 @@ export const MODELS = {
4995
5016
  "labs-devstral-small-2512": {
4996
5017
  id: "labs-devstral-small-2512",
4997
5018
  name: "Devstral Small 2",
4998
- api: "mistral-conversations",
5019
+ api: "openai-completions",
4999
5020
  provider: "mistral",
5000
- baseUrl: "https://api.mistral.ai",
5021
+ baseUrl: "https://api.mistral.ai/v1",
5001
5022
  reasoning: false,
5002
5023
  input: ["text", "image"],
5003
5024
  cost: {
@@ -5012,9 +5033,9 @@ export const MODELS = {
5012
5033
  "magistral-medium-latest": {
5013
5034
  id: "magistral-medium-latest",
5014
5035
  name: "Magistral Medium (latest)",
5015
- api: "mistral-conversations",
5036
+ api: "openai-completions",
5016
5037
  provider: "mistral",
5017
- baseUrl: "https://api.mistral.ai",
5038
+ baseUrl: "https://api.mistral.ai/v1",
5018
5039
  reasoning: true,
5019
5040
  input: ["text"],
5020
5041
  cost: {
@@ -5029,9 +5050,9 @@ export const MODELS = {
5029
5050
  "magistral-small": {
5030
5051
  id: "magistral-small",
5031
5052
  name: "Magistral Small",
5032
- api: "mistral-conversations",
5053
+ api: "openai-completions",
5033
5054
  provider: "mistral",
5034
- baseUrl: "https://api.mistral.ai",
5055
+ baseUrl: "https://api.mistral.ai/v1",
5035
5056
  reasoning: true,
5036
5057
  input: ["text"],
5037
5058
  cost: {
@@ -5046,9 +5067,9 @@ export const MODELS = {
5046
5067
  "ministral-3b-latest": {
5047
5068
  id: "ministral-3b-latest",
5048
5069
  name: "Ministral 3B (latest)",
5049
- api: "mistral-conversations",
5070
+ api: "openai-completions",
5050
5071
  provider: "mistral",
5051
- baseUrl: "https://api.mistral.ai",
5072
+ baseUrl: "https://api.mistral.ai/v1",
5052
5073
  reasoning: false,
5053
5074
  input: ["text"],
5054
5075
  cost: {
@@ -5063,9 +5084,9 @@ export const MODELS = {
5063
5084
  "ministral-8b-latest": {
5064
5085
  id: "ministral-8b-latest",
5065
5086
  name: "Ministral 8B (latest)",
5066
- api: "mistral-conversations",
5087
+ api: "openai-completions",
5067
5088
  provider: "mistral",
5068
- baseUrl: "https://api.mistral.ai",
5089
+ baseUrl: "https://api.mistral.ai/v1",
5069
5090
  reasoning: false,
5070
5091
  input: ["text"],
5071
5092
  cost: {
@@ -5080,9 +5101,9 @@ export const MODELS = {
5080
5101
  "mistral-large-2411": {
5081
5102
  id: "mistral-large-2411",
5082
5103
  name: "Mistral Large 2.1",
5083
- api: "mistral-conversations",
5104
+ api: "openai-completions",
5084
5105
  provider: "mistral",
5085
- baseUrl: "https://api.mistral.ai",
5106
+ baseUrl: "https://api.mistral.ai/v1",
5086
5107
  reasoning: false,
5087
5108
  input: ["text"],
5088
5109
  cost: {
@@ -5097,9 +5118,9 @@ export const MODELS = {
5097
5118
  "mistral-large-2512": {
5098
5119
  id: "mistral-large-2512",
5099
5120
  name: "Mistral Large 3",
5100
- api: "mistral-conversations",
5121
+ api: "openai-completions",
5101
5122
  provider: "mistral",
5102
- baseUrl: "https://api.mistral.ai",
5123
+ baseUrl: "https://api.mistral.ai/v1",
5103
5124
  reasoning: false,
5104
5125
  input: ["text", "image"],
5105
5126
  cost: {
@@ -5114,9 +5135,9 @@ export const MODELS = {
5114
5135
  "mistral-large-latest": {
5115
5136
  id: "mistral-large-latest",
5116
5137
  name: "Mistral Large (latest)",
5117
- api: "mistral-conversations",
5138
+ api: "openai-completions",
5118
5139
  provider: "mistral",
5119
- baseUrl: "https://api.mistral.ai",
5140
+ baseUrl: "https://api.mistral.ai/v1",
5120
5141
  reasoning: false,
5121
5142
  input: ["text", "image"],
5122
5143
  cost: {
@@ -5131,9 +5152,9 @@ export const MODELS = {
5131
5152
  "mistral-medium-2505": {
5132
5153
  id: "mistral-medium-2505",
5133
5154
  name: "Mistral Medium 3",
5134
- api: "mistral-conversations",
5155
+ api: "openai-completions",
5135
5156
  provider: "mistral",
5136
- baseUrl: "https://api.mistral.ai",
5157
+ baseUrl: "https://api.mistral.ai/v1",
5137
5158
  reasoning: false,
5138
5159
  input: ["text", "image"],
5139
5160
  cost: {
@@ -5148,9 +5169,9 @@ export const MODELS = {
5148
5169
  "mistral-medium-2508": {
5149
5170
  id: "mistral-medium-2508",
5150
5171
  name: "Mistral Medium 3.1",
5151
- api: "mistral-conversations",
5172
+ api: "openai-completions",
5152
5173
  provider: "mistral",
5153
- baseUrl: "https://api.mistral.ai",
5174
+ baseUrl: "https://api.mistral.ai/v1",
5154
5175
  reasoning: false,
5155
5176
  input: ["text", "image"],
5156
5177
  cost: {
@@ -5165,9 +5186,9 @@ export const MODELS = {
5165
5186
  "mistral-medium-latest": {
5166
5187
  id: "mistral-medium-latest",
5167
5188
  name: "Mistral Medium (latest)",
5168
- api: "mistral-conversations",
5189
+ api: "openai-completions",
5169
5190
  provider: "mistral",
5170
- baseUrl: "https://api.mistral.ai",
5191
+ baseUrl: "https://api.mistral.ai/v1",
5171
5192
  reasoning: false,
5172
5193
  input: ["text", "image"],
5173
5194
  cost: {
@@ -5182,9 +5203,9 @@ export const MODELS = {
5182
5203
  "mistral-nemo": {
5183
5204
  id: "mistral-nemo",
5184
5205
  name: "Mistral Nemo",
5185
- api: "mistral-conversations",
5206
+ api: "openai-completions",
5186
5207
  provider: "mistral",
5187
- baseUrl: "https://api.mistral.ai",
5208
+ baseUrl: "https://api.mistral.ai/v1",
5188
5209
  reasoning: false,
5189
5210
  input: ["text"],
5190
5211
  cost: {
@@ -5199,9 +5220,9 @@ export const MODELS = {
5199
5220
  "mistral-small-2506": {
5200
5221
  id: "mistral-small-2506",
5201
5222
  name: "Mistral Small 3.2",
5202
- api: "mistral-conversations",
5223
+ api: "openai-completions",
5203
5224
  provider: "mistral",
5204
- baseUrl: "https://api.mistral.ai",
5225
+ baseUrl: "https://api.mistral.ai/v1",
5205
5226
  reasoning: false,
5206
5227
  input: ["text", "image"],
5207
5228
  cost: {
@@ -5216,9 +5237,9 @@ export const MODELS = {
5216
5237
  "mistral-small-2603": {
5217
5238
  id: "mistral-small-2603",
5218
5239
  name: "Mistral Small 4",
5219
- api: "mistral-conversations",
5240
+ api: "openai-completions",
5220
5241
  provider: "mistral",
5221
- baseUrl: "https://api.mistral.ai",
5242
+ baseUrl: "https://api.mistral.ai/v1",
5222
5243
  reasoning: true,
5223
5244
  input: ["text", "image"],
5224
5245
  cost: {
@@ -5233,9 +5254,9 @@ export const MODELS = {
5233
5254
  "mistral-small-latest": {
5234
5255
  id: "mistral-small-latest",
5235
5256
  name: "Mistral Small (latest)",
5236
- api: "mistral-conversations",
5257
+ api: "openai-completions",
5237
5258
  provider: "mistral",
5238
- baseUrl: "https://api.mistral.ai",
5259
+ baseUrl: "https://api.mistral.ai/v1",
5239
5260
  reasoning: true,
5240
5261
  input: ["text", "image"],
5241
5262
  cost: {
@@ -5250,9 +5271,9 @@ export const MODELS = {
5250
5271
  "open-mistral-7b": {
5251
5272
  id: "open-mistral-7b",
5252
5273
  name: "Mistral 7B",
5253
- api: "mistral-conversations",
5274
+ api: "openai-completions",
5254
5275
  provider: "mistral",
5255
- baseUrl: "https://api.mistral.ai",
5276
+ baseUrl: "https://api.mistral.ai/v1",
5256
5277
  reasoning: false,
5257
5278
  input: ["text"],
5258
5279
  cost: {
@@ -5267,9 +5288,9 @@ export const MODELS = {
5267
5288
  "open-mixtral-8x22b": {
5268
5289
  id: "open-mixtral-8x22b",
5269
5290
  name: "Mixtral 8x22B",
5270
- api: "mistral-conversations",
5291
+ api: "openai-completions",
5271
5292
  provider: "mistral",
5272
- baseUrl: "https://api.mistral.ai",
5293
+ baseUrl: "https://api.mistral.ai/v1",
5273
5294
  reasoning: false,
5274
5295
  input: ["text"],
5275
5296
  cost: {
@@ -5284,9 +5305,9 @@ export const MODELS = {
5284
5305
  "open-mixtral-8x7b": {
5285
5306
  id: "open-mixtral-8x7b",
5286
5307
  name: "Mixtral 8x7B",
5287
- api: "mistral-conversations",
5308
+ api: "openai-completions",
5288
5309
  provider: "mistral",
5289
- baseUrl: "https://api.mistral.ai",
5310
+ baseUrl: "https://api.mistral.ai/v1",
5290
5311
  reasoning: false,
5291
5312
  input: ["text"],
5292
5313
  cost: {
@@ -5301,9 +5322,9 @@ export const MODELS = {
5301
5322
  "pixtral-12b": {
5302
5323
  id: "pixtral-12b",
5303
5324
  name: "Pixtral 12B",
5304
- api: "mistral-conversations",
5325
+ api: "openai-completions",
5305
5326
  provider: "mistral",
5306
- baseUrl: "https://api.mistral.ai",
5327
+ baseUrl: "https://api.mistral.ai/v1",
5307
5328
  reasoning: false,
5308
5329
  input: ["text", "image"],
5309
5330
  cost: {
@@ -5318,9 +5339,9 @@ export const MODELS = {
5318
5339
  "pixtral-large-latest": {
5319
5340
  id: "pixtral-large-latest",
5320
5341
  name: "Pixtral Large (latest)",
5321
- api: "mistral-conversations",
5342
+ api: "openai-completions",
5322
5343
  provider: "mistral",
5323
- baseUrl: "https://api.mistral.ai",
5344
+ baseUrl: "https://api.mistral.ai/v1",
5324
5345
  reasoning: false,
5325
5346
  input: ["text", "image"],
5326
5347
  cost: {
@@ -5334,23 +5355,6 @@ export const MODELS = {
5334
5355
  },
5335
5356
  },
5336
5357
  "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
5358
  "gpt-4": {
5355
5359
  id: "gpt-4",
5356
5360
  name: "GPT-4",
@@ -6290,6 +6294,23 @@ export const MODELS = {
6290
6294
  contextWindow: 1000000,
6291
6295
  maxTokens: 128000,
6292
6296
  },
6297
+ "claude-opus-4-7": {
6298
+ id: "claude-opus-4-7",
6299
+ name: "Claude Opus 4.7",
6300
+ api: "anthropic-messages",
6301
+ provider: "opencode",
6302
+ baseUrl: "https://opencode.ai/zen",
6303
+ reasoning: true,
6304
+ input: ["text", "image"],
6305
+ cost: {
6306
+ input: 5,
6307
+ output: 25,
6308
+ cacheRead: 0.5,
6309
+ cacheWrite: 6.25,
6310
+ },
6311
+ contextWindow: 1000000,
6312
+ maxTokens: 128000,
6313
+ },
6293
6314
  "claude-sonnet-4": {
6294
6315
  id: "claude-sonnet-4",
6295
6316
  name: "Claude Sonnet 4",
@@ -6392,6 +6413,23 @@ export const MODELS = {
6392
6413
  contextWindow: 204800,
6393
6414
  maxTokens: 131072,
6394
6415
  },
6416
+ "glm-5.1": {
6417
+ id: "glm-5.1",
6418
+ name: "GLM-5.1",
6419
+ api: "openai-completions",
6420
+ provider: "opencode",
6421
+ baseUrl: "https://opencode.ai/zen/v1",
6422
+ reasoning: true,
6423
+ input: ["text"],
6424
+ cost: {
6425
+ input: 1.4,
6426
+ output: 4.4,
6427
+ cacheRead: 0.26,
6428
+ cacheWrite: 0,
6429
+ },
6430
+ contextWindow: 204800,
6431
+ maxTokens: 131072,
6432
+ },
6395
6433
  "gpt-5": {
6396
6434
  id: "gpt-5",
6397
6435
  name: "GPT-5",
@@ -6630,6 +6668,23 @@ export const MODELS = {
6630
6668
  contextWindow: 1050000,
6631
6669
  maxTokens: 128000,
6632
6670
  },
6671
+ "hy3-preview-free": {
6672
+ id: "hy3-preview-free",
6673
+ name: "Hy3 Preview Free",
6674
+ api: "openai-completions",
6675
+ provider: "opencode",
6676
+ baseUrl: "https://opencode.ai/zen/v1",
6677
+ reasoning: true,
6678
+ input: ["text"],
6679
+ cost: {
6680
+ input: 0,
6681
+ output: 0,
6682
+ cacheRead: 0,
6683
+ cacheWrite: 0,
6684
+ },
6685
+ contextWindow: 256000,
6686
+ maxTokens: 64000,
6687
+ },
6633
6688
  "kimi-k2.5": {
6634
6689
  id: "kimi-k2.5",
6635
6690
  name: "Kimi K2.5",
@@ -6647,6 +6702,40 @@ export const MODELS = {
6647
6702
  contextWindow: 262144,
6648
6703
  maxTokens: 65536,
6649
6704
  },
6705
+ "kimi-k2.6": {
6706
+ id: "kimi-k2.6",
6707
+ name: "Kimi K2.6",
6708
+ api: "openai-completions",
6709
+ provider: "opencode",
6710
+ baseUrl: "https://opencode.ai/zen/v1",
6711
+ reasoning: true,
6712
+ input: ["text", "image"],
6713
+ cost: {
6714
+ input: 0.95,
6715
+ output: 4,
6716
+ cacheRead: 0.16,
6717
+ cacheWrite: 0,
6718
+ },
6719
+ contextWindow: 262144,
6720
+ maxTokens: 65536,
6721
+ },
6722
+ "ling-2.6-flash-free": {
6723
+ id: "ling-2.6-flash-free",
6724
+ name: "Ling 2.6 Flash Free",
6725
+ api: "openai-completions",
6726
+ provider: "opencode",
6727
+ baseUrl: "https://opencode.ai/zen/v1",
6728
+ reasoning: false,
6729
+ input: ["text"],
6730
+ cost: {
6731
+ input: 0,
6732
+ output: 0,
6733
+ cacheRead: 0,
6734
+ cacheWrite: 0,
6735
+ },
6736
+ contextWindow: 262100,
6737
+ maxTokens: 32800,
6738
+ },
6650
6739
  "minimax-m2.5": {
6651
6740
  id: "minimax-m2.5",
6652
6741
  name: "MiniMax M2.5",
@@ -6681,26 +6770,26 @@ export const MODELS = {
6681
6770
  contextWindow: 204800,
6682
6771
  maxTokens: 131072,
6683
6772
  },
6684
- "nemotron-3-super-free": {
6685
- id: "nemotron-3-super-free",
6686
- name: "Nemotron 3 Super Free",
6773
+ "minimax-m2.7": {
6774
+ id: "minimax-m2.7",
6775
+ name: "MiniMax M2.7",
6687
6776
  api: "openai-completions",
6688
6777
  provider: "opencode",
6689
6778
  baseUrl: "https://opencode.ai/zen/v1",
6690
6779
  reasoning: true,
6691
6780
  input: ["text"],
6692
6781
  cost: {
6693
- input: 0,
6694
- output: 0,
6695
- cacheRead: 0,
6782
+ input: 0.3,
6783
+ output: 1.2,
6784
+ cacheRead: 0.06,
6696
6785
  cacheWrite: 0,
6697
6786
  },
6698
6787
  contextWindow: 204800,
6699
- maxTokens: 128000,
6788
+ maxTokens: 131072,
6700
6789
  },
6701
- "qwen3.6-plus-free": {
6702
- id: "qwen3.6-plus-free",
6703
- name: "Qwen3.6 Plus Free",
6790
+ "nemotron-3-super-free": {
6791
+ id: "nemotron-3-super-free",
6792
+ name: "Nemotron 3 Super Free",
6704
6793
  api: "openai-completions",
6705
6794
  provider: "opencode",
6706
6795
  baseUrl: "https://opencode.ai/zen/v1",
@@ -6712,8 +6801,42 @@ export const MODELS = {
6712
6801
  cacheRead: 0,
6713
6802
  cacheWrite: 0,
6714
6803
  },
6715
- contextWindow: 1048576,
6716
- maxTokens: 64000,
6804
+ contextWindow: 204800,
6805
+ maxTokens: 128000,
6806
+ },
6807
+ "qwen3.5-plus": {
6808
+ id: "qwen3.5-plus",
6809
+ name: "Qwen3.5 Plus",
6810
+ api: "anthropic-messages",
6811
+ provider: "opencode",
6812
+ baseUrl: "https://opencode.ai/zen",
6813
+ reasoning: true,
6814
+ input: ["text", "image"],
6815
+ cost: {
6816
+ input: 0.2,
6817
+ output: 1.2,
6818
+ cacheRead: 0.02,
6819
+ cacheWrite: 0.25,
6820
+ },
6821
+ contextWindow: 262144,
6822
+ maxTokens: 65536,
6823
+ },
6824
+ "qwen3.6-plus": {
6825
+ id: "qwen3.6-plus",
6826
+ name: "Qwen3.6 Plus",
6827
+ api: "anthropic-messages",
6828
+ provider: "opencode",
6829
+ baseUrl: "https://opencode.ai/zen",
6830
+ reasoning: true,
6831
+ input: ["text", "image"],
6832
+ cost: {
6833
+ input: 0.5,
6834
+ output: 3,
6835
+ cacheRead: 0.05,
6836
+ cacheWrite: 0.625,
6837
+ },
6838
+ contextWindow: 262144,
6839
+ maxTokens: 65536,
6717
6840
  },
6718
6841
  },
6719
6842
  "opencode-go": {
@@ -6734,6 +6857,23 @@ export const MODELS = {
6734
6857
  contextWindow: 204800,
6735
6858
  maxTokens: 131072,
6736
6859
  },
6860
+ "glm-5.1": {
6861
+ id: "glm-5.1",
6862
+ name: "GLM-5.1",
6863
+ api: "openai-completions",
6864
+ provider: "opencode-go",
6865
+ baseUrl: "https://opencode.ai/zen/go/v1",
6866
+ reasoning: true,
6867
+ input: ["text"],
6868
+ cost: {
6869
+ input: 1.4,
6870
+ output: 4.4,
6871
+ cacheRead: 0.26,
6872
+ cacheWrite: 0,
6873
+ },
6874
+ contextWindow: 204800,
6875
+ maxTokens: 131072,
6876
+ },
6737
6877
  "kimi-k2.5": {
6738
6878
  id: "kimi-k2.5",
6739
6879
  name: "Kimi K2.5",
@@ -6751,6 +6891,23 @@ export const MODELS = {
6751
6891
  contextWindow: 262144,
6752
6892
  maxTokens: 65536,
6753
6893
  },
6894
+ "kimi-k2.6": {
6895
+ id: "kimi-k2.6",
6896
+ name: "Kimi K2.6 (3x limits)",
6897
+ api: "openai-completions",
6898
+ provider: "opencode-go",
6899
+ baseUrl: "https://opencode.ai/zen/go/v1",
6900
+ reasoning: true,
6901
+ input: ["text", "image"],
6902
+ cost: {
6903
+ input: 0.32,
6904
+ output: 1.34,
6905
+ cacheRead: 0.054,
6906
+ cacheWrite: 0,
6907
+ },
6908
+ contextWindow: 262144,
6909
+ maxTokens: 65536,
6910
+ },
6754
6911
  "mimo-v2-omni": {
6755
6912
  id: "mimo-v2-omni",
6756
6913
  name: "MiMo V2 Omni",
@@ -6766,7 +6923,7 @@ export const MODELS = {
6766
6923
  cacheWrite: 0,
6767
6924
  },
6768
6925
  contextWindow: 262144,
6769
- maxTokens: 64000,
6926
+ maxTokens: 128000,
6770
6927
  },
6771
6928
  "mimo-v2-pro": {
6772
6929
  id: "mimo-v2-pro",
@@ -6783,7 +6940,41 @@ export const MODELS = {
6783
6940
  cacheWrite: 0,
6784
6941
  },
6785
6942
  contextWindow: 1048576,
6786
- maxTokens: 64000,
6943
+ maxTokens: 128000,
6944
+ },
6945
+ "mimo-v2.5": {
6946
+ id: "mimo-v2.5",
6947
+ name: "MiMo V2.5",
6948
+ api: "openai-completions",
6949
+ provider: "opencode-go",
6950
+ baseUrl: "https://opencode.ai/zen/go/v1",
6951
+ reasoning: true,
6952
+ input: ["text", "image"],
6953
+ cost: {
6954
+ input: 0.4,
6955
+ output: 2,
6956
+ cacheRead: 0.08,
6957
+ cacheWrite: 0,
6958
+ },
6959
+ contextWindow: 262144,
6960
+ maxTokens: 128000,
6961
+ },
6962
+ "mimo-v2.5-pro": {
6963
+ id: "mimo-v2.5-pro",
6964
+ name: "MiMo V2.5 Pro",
6965
+ api: "openai-completions",
6966
+ provider: "opencode-go",
6967
+ baseUrl: "https://opencode.ai/zen/go/v1",
6968
+ reasoning: true,
6969
+ input: ["text"],
6970
+ cost: {
6971
+ input: 1,
6972
+ output: 3,
6973
+ cacheRead: 0.2,
6974
+ cacheWrite: 0,
6975
+ },
6976
+ contextWindow: 1048576,
6977
+ maxTokens: 128000,
6787
6978
  },
6788
6979
  "minimax-m2.5": {
6789
6980
  id: "minimax-m2.5",
@@ -6800,7 +6991,7 @@ export const MODELS = {
6800
6991
  cacheWrite: 0,
6801
6992
  },
6802
6993
  contextWindow: 204800,
6803
- maxTokens: 131072,
6994
+ maxTokens: 65536,
6804
6995
  },
6805
6996
  "minimax-m2.7": {
6806
6997
  id: "minimax-m2.7",
@@ -6819,6 +7010,40 @@ export const MODELS = {
6819
7010
  contextWindow: 204800,
6820
7011
  maxTokens: 131072,
6821
7012
  },
7013
+ "qwen3.5-plus": {
7014
+ id: "qwen3.5-plus",
7015
+ name: "Qwen3.5 Plus",
7016
+ api: "anthropic-messages",
7017
+ provider: "opencode-go",
7018
+ baseUrl: "https://opencode.ai/zen/go",
7019
+ reasoning: true,
7020
+ input: ["text", "image"],
7021
+ cost: {
7022
+ input: 0.2,
7023
+ output: 1.2,
7024
+ cacheRead: 0.02,
7025
+ cacheWrite: 0.25,
7026
+ },
7027
+ contextWindow: 262144,
7028
+ maxTokens: 65536,
7029
+ },
7030
+ "qwen3.6-plus": {
7031
+ id: "qwen3.6-plus",
7032
+ name: "Qwen3.6 Plus",
7033
+ api: "anthropic-messages",
7034
+ provider: "opencode-go",
7035
+ baseUrl: "https://opencode.ai/zen/go",
7036
+ reasoning: true,
7037
+ input: ["text", "image"],
7038
+ cost: {
7039
+ input: 0.5,
7040
+ output: 3,
7041
+ cacheRead: 0.05,
7042
+ cacheWrite: 0.625,
7043
+ },
7044
+ contextWindow: 262144,
7045
+ maxTokens: 65536,
7046
+ },
6822
7047
  },
6823
7048
  "openrouter": {
6824
7049
  "ai21/jamba-large-1.7": {
@@ -7006,7 +7231,7 @@ export const MODELS = {
7006
7231
  cacheWrite: 3.75,
7007
7232
  },
7008
7233
  contextWindow: 200000,
7009
- maxTokens: 64000,
7234
+ maxTokens: 128000,
7010
7235
  },
7011
7236
  "anthropic/claude-3.7-sonnet:thinking": {
7012
7237
  id: "anthropic/claude-3.7-sonnet:thinking",
@@ -7110,6 +7335,40 @@ export const MODELS = {
7110
7335
  contextWindow: 1000000,
7111
7336
  maxTokens: 128000,
7112
7337
  },
7338
+ "anthropic/claude-opus-4.6-fast": {
7339
+ id: "anthropic/claude-opus-4.6-fast",
7340
+ name: "Anthropic: Claude Opus 4.6 (Fast)",
7341
+ api: "openai-completions",
7342
+ provider: "openrouter",
7343
+ baseUrl: "https://openrouter.ai/api/v1",
7344
+ reasoning: true,
7345
+ input: ["text", "image"],
7346
+ cost: {
7347
+ input: 30,
7348
+ output: 150,
7349
+ cacheRead: 3,
7350
+ cacheWrite: 37.5,
7351
+ },
7352
+ contextWindow: 1000000,
7353
+ maxTokens: 128000,
7354
+ },
7355
+ "anthropic/claude-opus-4.7": {
7356
+ id: "anthropic/claude-opus-4.7",
7357
+ name: "Anthropic: Claude Opus 4.7",
7358
+ api: "openai-completions",
7359
+ provider: "openrouter",
7360
+ baseUrl: "https://openrouter.ai/api/v1",
7361
+ reasoning: true,
7362
+ input: ["text", "image"],
7363
+ cost: {
7364
+ input: 5,
7365
+ output: 25,
7366
+ cacheRead: 0.5,
7367
+ cacheWrite: 6.25,
7368
+ },
7369
+ contextWindow: 1000000,
7370
+ maxTokens: 128000,
7371
+ },
7113
7372
  "anthropic/claude-sonnet-4": {
7114
7373
  id: "anthropic/claude-sonnet-4",
7115
7374
  name: "Anthropic: Claude Sonnet 4",
@@ -7124,7 +7383,7 @@ export const MODELS = {
7124
7383
  cacheRead: 0.3,
7125
7384
  cacheWrite: 3.75,
7126
7385
  },
7127
- contextWindow: 200000,
7386
+ contextWindow: 1000000,
7128
7387
  maxTokens: 64000,
7129
7388
  },
7130
7389
  "anthropic/claude-sonnet-4.5": {
@@ -7161,17 +7420,17 @@ export const MODELS = {
7161
7420
  contextWindow: 1000000,
7162
7421
  maxTokens: 128000,
7163
7422
  },
7164
- "arcee-ai/trinity-large-preview:free": {
7165
- id: "arcee-ai/trinity-large-preview:free",
7166
- name: "Arcee AI: Trinity Large Preview (free)",
7423
+ "arcee-ai/trinity-large-preview": {
7424
+ id: "arcee-ai/trinity-large-preview",
7425
+ name: "Arcee AI: Trinity Large Preview",
7167
7426
  api: "openai-completions",
7168
7427
  provider: "openrouter",
7169
7428
  baseUrl: "https://openrouter.ai/api/v1",
7170
7429
  reasoning: false,
7171
7430
  input: ["text"],
7172
7431
  cost: {
7173
- input: 0,
7174
- output: 0,
7432
+ input: 0.15,
7433
+ output: 0.44999999999999996,
7175
7434
  cacheRead: 0,
7176
7435
  cacheWrite: 0,
7177
7436
  },
@@ -7189,7 +7448,7 @@ export const MODELS = {
7189
7448
  cost: {
7190
7449
  input: 0.22,
7191
7450
  output: 0.85,
7192
- cacheRead: 0,
7451
+ cacheRead: 0.06,
7193
7452
  cacheWrite: 0,
7194
7453
  },
7195
7454
  contextWindow: 262144,
@@ -7212,23 +7471,6 @@ export const MODELS = {
7212
7471
  contextWindow: 131072,
7213
7472
  maxTokens: 131072,
7214
7473
  },
7215
- "arcee-ai/trinity-mini:free": {
7216
- id: "arcee-ai/trinity-mini:free",
7217
- name: "Arcee AI: Trinity Mini (free)",
7218
- api: "openai-completions",
7219
- provider: "openrouter",
7220
- baseUrl: "https://openrouter.ai/api/v1",
7221
- reasoning: true,
7222
- input: ["text"],
7223
- cost: {
7224
- input: 0,
7225
- output: 0,
7226
- cacheRead: 0,
7227
- cacheWrite: 0,
7228
- },
7229
- contextWindow: 131072,
7230
- maxTokens: 4096,
7231
- },
7232
7474
  "arcee-ai/virtuoso-large": {
7233
7475
  id: "arcee-ai/virtuoso-large",
7234
7476
  name: "Arcee AI: Virtuoso Large",
@@ -7422,7 +7664,7 @@ export const MODELS = {
7422
7664
  api: "openai-completions",
7423
7665
  provider: "openrouter",
7424
7666
  baseUrl: "https://openrouter.ai/api/v1",
7425
- reasoning: true,
7667
+ reasoning: false,
7426
7668
  input: ["text"],
7427
7669
  cost: {
7428
7670
  input: 0.19999999999999998,
@@ -7476,13 +7718,13 @@ export const MODELS = {
7476
7718
  reasoning: true,
7477
7719
  input: ["text"],
7478
7720
  cost: {
7479
- input: 0.44999999999999996,
7721
+ input: 0.5,
7480
7722
  output: 2.1500000000000004,
7481
- cacheRead: 0.22499999999999998,
7723
+ cacheRead: 0.35,
7482
7724
  cacheWrite: 0,
7483
7725
  },
7484
7726
  contextWindow: 163840,
7485
- maxTokens: 65536,
7727
+ maxTokens: 4096,
7486
7728
  },
7487
7729
  "deepseek/deepseek-v3.1-terminus": {
7488
7730
  id: "deepseek/deepseek-v3.1-terminus",
@@ -7510,13 +7752,13 @@ export const MODELS = {
7510
7752
  reasoning: true,
7511
7753
  input: ["text"],
7512
7754
  cost: {
7513
- input: 0.26,
7514
- output: 0.38,
7515
- cacheRead: 0.13,
7755
+ input: 0.252,
7756
+ output: 0.378,
7757
+ cacheRead: 0.0252,
7516
7758
  cacheWrite: 0,
7517
7759
  },
7518
- contextWindow: 163840,
7519
- maxTokens: 4096,
7760
+ contextWindow: 131072,
7761
+ maxTokens: 65536,
7520
7762
  },
7521
7763
  "deepseek/deepseek-v3.2-exp": {
7522
7764
  id: "deepseek/deepseek-v3.2-exp",
@@ -7566,7 +7808,7 @@ export const MODELS = {
7566
7808
  cacheRead: 0.024999999999999998,
7567
7809
  cacheWrite: 0.08333333333333334,
7568
7810
  },
7569
- contextWindow: 1048576,
7811
+ contextWindow: 1000000,
7570
7812
  maxTokens: 8192,
7571
7813
  },
7572
7814
  "google/gemini-2.0-flash-lite-001": {
@@ -7765,13 +8007,30 @@ export const MODELS = {
7765
8007
  reasoning: true,
7766
8008
  input: ["text", "image"],
7767
8009
  cost: {
7768
- input: 0.13,
7769
- output: 0.39999999999999997,
8010
+ input: 0.06,
8011
+ output: 0.33,
7770
8012
  cacheRead: 0,
7771
8013
  cacheWrite: 0,
7772
8014
  },
7773
8015
  contextWindow: 262144,
7774
- maxTokens: 262144,
8016
+ maxTokens: 4096,
8017
+ },
8018
+ "google/gemma-4-26b-a4b-it:free": {
8019
+ id: "google/gemma-4-26b-a4b-it:free",
8020
+ name: "Google: Gemma 4 26B A4B (free)",
8021
+ api: "openai-completions",
8022
+ provider: "openrouter",
8023
+ baseUrl: "https://openrouter.ai/api/v1",
8024
+ reasoning: true,
8025
+ input: ["text", "image"],
8026
+ cost: {
8027
+ input: 0,
8028
+ output: 0,
8029
+ cacheRead: 0,
8030
+ cacheWrite: 0,
8031
+ },
8032
+ contextWindow: 262144,
8033
+ maxTokens: 32768,
7775
8034
  },
7776
8035
  "google/gemma-4-31b-it": {
7777
8036
  id: "google/gemma-4-31b-it",
@@ -7782,30 +8041,30 @@ export const MODELS = {
7782
8041
  reasoning: true,
7783
8042
  input: ["text", "image"],
7784
8043
  cost: {
7785
- input: 0.14,
7786
- output: 0.39999999999999997,
8044
+ input: 0.13,
8045
+ output: 0.38,
7787
8046
  cacheRead: 0,
7788
8047
  cacheWrite: 0,
7789
8048
  },
7790
8049
  contextWindow: 262144,
7791
- maxTokens: 131072,
8050
+ maxTokens: 4096,
7792
8051
  },
7793
- "inception/mercury": {
7794
- id: "inception/mercury",
7795
- name: "Inception: Mercury",
8052
+ "google/gemma-4-31b-it:free": {
8053
+ id: "google/gemma-4-31b-it:free",
8054
+ name: "Google: Gemma 4 31B (free)",
7796
8055
  api: "openai-completions",
7797
8056
  provider: "openrouter",
7798
8057
  baseUrl: "https://openrouter.ai/api/v1",
7799
- reasoning: false,
7800
- input: ["text"],
8058
+ reasoning: true,
8059
+ input: ["text", "image"],
7801
8060
  cost: {
7802
- input: 0.25,
7803
- output: 0.75,
7804
- cacheRead: 0.024999999999999998,
8061
+ input: 0,
8062
+ output: 0,
8063
+ cacheRead: 0,
7805
8064
  cacheWrite: 0,
7806
8065
  },
7807
- contextWindow: 128000,
7808
- maxTokens: 32000,
8066
+ contextWindow: 262144,
8067
+ maxTokens: 32768,
7809
8068
  },
7810
8069
  "inception/mercury-2": {
7811
8070
  id: "inception/mercury-2",
@@ -7824,22 +8083,22 @@ export const MODELS = {
7824
8083
  contextWindow: 128000,
7825
8084
  maxTokens: 50000,
7826
8085
  },
7827
- "inception/mercury-coder": {
7828
- id: "inception/mercury-coder",
7829
- name: "Inception: Mercury Coder",
8086
+ "inclusionai/ling-2.6-flash:free": {
8087
+ id: "inclusionai/ling-2.6-flash:free",
8088
+ name: "inclusionAI: Ling-2.6-flash (free)",
7830
8089
  api: "openai-completions",
7831
8090
  provider: "openrouter",
7832
8091
  baseUrl: "https://openrouter.ai/api/v1",
7833
8092
  reasoning: false,
7834
8093
  input: ["text"],
7835
8094
  cost: {
7836
- input: 0.25,
7837
- output: 0.75,
7838
- cacheRead: 0.024999999999999998,
8095
+ input: 0,
8096
+ output: 0,
8097
+ cacheRead: 0,
7839
8098
  cacheWrite: 0,
7840
8099
  },
7841
- contextWindow: 128000,
7842
- maxTokens: 32000,
8100
+ contextWindow: 262144,
8101
+ maxTokens: 32768,
7843
8102
  },
7844
8103
  "kwaipilot/kat-coder-pro-v2": {
7845
8104
  id: "kwaipilot/kat-coder-pro-v2",
@@ -7858,23 +8117,6 @@ export const MODELS = {
7858
8117
  contextWindow: 256000,
7859
8118
  maxTokens: 80000,
7860
8119
  },
7861
- "meituan/longcat-flash-chat": {
7862
- id: "meituan/longcat-flash-chat",
7863
- name: "Meituan: LongCat Flash Chat",
7864
- api: "openai-completions",
7865
- provider: "openrouter",
7866
- baseUrl: "https://openrouter.ai/api/v1",
7867
- reasoning: false,
7868
- input: ["text"],
7869
- cost: {
7870
- input: 0.19999999999999998,
7871
- output: 0.7999999999999999,
7872
- cacheRead: 0.19999999999999998,
7873
- cacheWrite: 0,
7874
- },
7875
- contextWindow: 131072,
7876
- maxTokens: 131072,
7877
- },
7878
8120
  "meta-llama/llama-3-8b-instruct": {
7879
8121
  id: "meta-llama/llama-3-8b-instruct",
7880
8122
  name: "Meta: Llama 3 8B Instruct",
@@ -7960,23 +8202,6 @@ export const MODELS = {
7960
8202
  contextWindow: 65536,
7961
8203
  maxTokens: 4096,
7962
8204
  },
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
8205
  "meta-llama/llama-4-scout": {
7981
8206
  id: "meta-llama/llama-4-scout",
7982
8207
  name: "Meta: Llama 4 Scout",
@@ -8037,13 +8262,13 @@ export const MODELS = {
8037
8262
  reasoning: true,
8038
8263
  input: ["text"],
8039
8264
  cost: {
8040
- input: 0.27,
8265
+ input: 0.29,
8041
8266
  output: 0.95,
8042
- cacheRead: 0.0290000007,
8267
+ cacheRead: 0.03,
8043
8268
  cacheWrite: 0,
8044
8269
  },
8045
8270
  contextWindow: 196608,
8046
- maxTokens: 4096,
8271
+ maxTokens: 196608,
8047
8272
  },
8048
8273
  "minimax/minimax-m2.5": {
8049
8274
  id: "minimax/minimax-m2.5",
@@ -8054,9 +8279,9 @@ export const MODELS = {
8054
8279
  reasoning: true,
8055
8280
  input: ["text"],
8056
8281
  cost: {
8057
- input: 0.118,
8058
- output: 0.9900000000000001,
8059
- cacheRead: 0.059,
8282
+ input: 0.15,
8283
+ output: 1.2,
8284
+ cacheRead: 0.075,
8060
8285
  cacheWrite: 0,
8061
8286
  },
8062
8287
  contextWindow: 196608,
@@ -8077,7 +8302,7 @@ export const MODELS = {
8077
8302
  cacheWrite: 0,
8078
8303
  },
8079
8304
  contextWindow: 196608,
8080
- maxTokens: 196608,
8305
+ maxTokens: 8192,
8081
8306
  },
8082
8307
  "minimax/minimax-m2.7": {
8083
8308
  id: "minimax/minimax-m2.7",
@@ -8090,11 +8315,11 @@ export const MODELS = {
8090
8315
  cost: {
8091
8316
  input: 0.3,
8092
8317
  output: 1.2,
8093
- cacheRead: 0.06,
8318
+ cacheRead: 0.059,
8094
8319
  cacheWrite: 0,
8095
8320
  },
8096
- contextWindow: 204800,
8097
- maxTokens: 131072,
8321
+ contextWindow: 196608,
8322
+ maxTokens: 4096,
8098
8323
  },
8099
8324
  "mistralai/codestral-2508": {
8100
8325
  id: "mistralai/codestral-2508",
@@ -8326,13 +8551,13 @@ export const MODELS = {
8326
8551
  reasoning: false,
8327
8552
  input: ["text"],
8328
8553
  cost: {
8329
- input: 0.02,
8330
- output: 0.04,
8554
+ input: 0.01,
8555
+ output: 0.03,
8331
8556
  cacheRead: 0,
8332
8557
  cacheWrite: 0,
8333
8558
  },
8334
8559
  contextWindow: 131072,
8335
- maxTokens: 16384,
8560
+ maxTokens: 4096,
8336
8561
  },
8337
8562
  "mistralai/mistral-saba": {
8338
8563
  id: "mistralai/mistral-saba",
@@ -8485,7 +8710,7 @@ export const MODELS = {
8485
8710
  cacheWrite: 0,
8486
8711
  },
8487
8712
  contextWindow: 131072,
8488
- maxTokens: 131072,
8713
+ maxTokens: 32768,
8489
8714
  },
8490
8715
  "moonshotai/kimi-k2-0905": {
8491
8716
  id: "moonshotai/kimi-k2-0905",
@@ -8498,11 +8723,11 @@ export const MODELS = {
8498
8723
  cost: {
8499
8724
  input: 0.39999999999999997,
8500
8725
  output: 2,
8501
- cacheRead: 0.15,
8726
+ cacheRead: 0,
8502
8727
  cacheWrite: 0,
8503
8728
  },
8504
- contextWindow: 131072,
8505
- maxTokens: 4096,
8729
+ contextWindow: 262144,
8730
+ maxTokens: 262144,
8506
8731
  },
8507
8732
  "moonshotai/kimi-k2-thinking": {
8508
8733
  id: "moonshotai/kimi-k2-thinking",
@@ -8513,13 +8738,13 @@ export const MODELS = {
8513
8738
  reasoning: true,
8514
8739
  input: ["text"],
8515
8740
  cost: {
8516
- input: 0.47,
8517
- output: 2,
8518
- cacheRead: 0.14100000000000001,
8741
+ input: 0.6,
8742
+ output: 2.5,
8743
+ cacheRead: 0.15,
8519
8744
  cacheWrite: 0,
8520
8745
  },
8521
- contextWindow: 131072,
8522
- maxTokens: 4096,
8746
+ contextWindow: 262144,
8747
+ maxTokens: 262144,
8523
8748
  },
8524
8749
  "moonshotai/kimi-k2.5": {
8525
8750
  id: "moonshotai/kimi-k2.5",
@@ -8538,6 +8763,23 @@ export const MODELS = {
8538
8763
  contextWindow: 262144,
8539
8764
  maxTokens: 4096,
8540
8765
  },
8766
+ "moonshotai/kimi-k2.6": {
8767
+ id: "moonshotai/kimi-k2.6",
8768
+ name: "MoonshotAI: Kimi K2.6",
8769
+ api: "openai-completions",
8770
+ provider: "openrouter",
8771
+ baseUrl: "https://openrouter.ai/api/v1",
8772
+ reasoning: true,
8773
+ input: ["text", "image"],
8774
+ cost: {
8775
+ input: 0.56,
8776
+ output: 3.5,
8777
+ cacheRead: 0.11,
8778
+ cacheWrite: 0,
8779
+ },
8780
+ contextWindow: 256000,
8781
+ maxTokens: 65536,
8782
+ },
8541
8783
  "nex-agi/deepseek-v3.1-nex-n1": {
8542
8784
  id: "nex-agi/deepseek-v3.1-nex-n1",
8543
8785
  name: "Nex AGI: DeepSeek V3.1 Nex N1",
@@ -8632,9 +8874,9 @@ export const MODELS = {
8632
8874
  reasoning: true,
8633
8875
  input: ["text"],
8634
8876
  cost: {
8635
- input: 0.09999999999999999,
8636
- output: 0.5,
8637
- cacheRead: 0.09999999999999999,
8877
+ input: 0.09,
8878
+ output: 0.44999999999999996,
8879
+ cacheRead: 0,
8638
8880
  cacheWrite: 0,
8639
8881
  },
8640
8882
  contextWindow: 262144,
@@ -8859,7 +9101,7 @@ export const MODELS = {
8859
9101
  cacheWrite: 0,
8860
9102
  },
8861
9103
  contextWindow: 1047576,
8862
- maxTokens: 32768,
9104
+ maxTokens: 4096,
8863
9105
  },
8864
9106
  "openai/gpt-4.1-mini": {
8865
9107
  id: "openai/gpt-4.1-mini",
@@ -8906,7 +9148,7 @@ export const MODELS = {
8906
9148
  cost: {
8907
9149
  input: 2.5,
8908
9150
  output: 10,
8909
- cacheRead: 1.25,
9151
+ cacheRead: 0,
8910
9152
  cacheWrite: 0,
8911
9153
  },
8912
9154
  contextWindow: 128000,
@@ -9014,23 +9256,6 @@ export const MODELS = {
9014
9256
  contextWindow: 128000,
9015
9257
  maxTokens: 16384,
9016
9258
  },
9017
- "openai/gpt-4o:extended": {
9018
- id: "openai/gpt-4o:extended",
9019
- name: "OpenAI: GPT-4o (extended)",
9020
- api: "openai-completions",
9021
- provider: "openrouter",
9022
- baseUrl: "https://openrouter.ai/api/v1",
9023
- reasoning: false,
9024
- input: ["text", "image"],
9025
- cost: {
9026
- input: 6,
9027
- output: 18,
9028
- cacheRead: 0,
9029
- cacheWrite: 0,
9030
- },
9031
- contextWindow: 128000,
9032
- maxTokens: 64000,
9033
- },
9034
9259
  "openai/gpt-5": {
9035
9260
  id: "openai/gpt-5",
9036
9261
  name: "OpenAI: GPT-5",
@@ -9065,40 +9290,6 @@ export const MODELS = {
9065
9290
  contextWindow: 400000,
9066
9291
  maxTokens: 128000,
9067
9292
  },
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
9293
  "openai/gpt-5-mini": {
9103
9294
  id: "openai/gpt-5-mini",
9104
9295
  name: "OpenAI: GPT-5 Mini",
@@ -9127,11 +9318,11 @@ export const MODELS = {
9127
9318
  cost: {
9128
9319
  input: 0.049999999999999996,
9129
9320
  output: 0.39999999999999997,
9130
- cacheRead: 0.005,
9321
+ cacheRead: 0.01,
9131
9322
  cacheWrite: 0,
9132
9323
  },
9133
9324
  contextWindow: 400000,
9134
- maxTokens: 128000,
9325
+ maxTokens: 4096,
9135
9326
  },
9136
9327
  "openai/gpt-5-pro": {
9137
9328
  id: "openai/gpt-5-pro",
@@ -9161,7 +9352,7 @@ export const MODELS = {
9161
9352
  cost: {
9162
9353
  input: 1.25,
9163
9354
  output: 10,
9164
- cacheRead: 0.125,
9355
+ cacheRead: 0.13,
9165
9356
  cacheWrite: 0,
9166
9357
  },
9167
9358
  contextWindow: 400000,
@@ -9229,11 +9420,11 @@ export const MODELS = {
9229
9420
  cost: {
9230
9421
  input: 0.25,
9231
9422
  output: 2,
9232
- cacheRead: 0.024999999999999998,
9423
+ cacheRead: 0.03,
9233
9424
  cacheWrite: 0,
9234
9425
  },
9235
9426
  contextWindow: 400000,
9236
- maxTokens: 100000,
9427
+ maxTokens: 128000,
9237
9428
  },
9238
9429
  "openai/gpt-5.2": {
9239
9430
  id: "openai/gpt-5.2",
@@ -9267,7 +9458,7 @@ export const MODELS = {
9267
9458
  cacheWrite: 0,
9268
9459
  },
9269
9460
  contextWindow: 128000,
9270
- maxTokens: 16384,
9461
+ maxTokens: 32000,
9271
9462
  },
9272
9463
  "openai/gpt-5.2-codex": {
9273
9464
  id: "openai/gpt-5.2-codex",
@@ -9483,12 +9674,12 @@ export const MODELS = {
9483
9674
  input: ["text"],
9484
9675
  cost: {
9485
9676
  input: 0.03,
9486
- output: 0.11,
9487
- cacheRead: 0.015,
9677
+ output: 0.14,
9678
+ cacheRead: 0,
9488
9679
  cacheWrite: 0,
9489
9680
  },
9490
9681
  contextWindow: 131072,
9491
- maxTokens: 131072,
9682
+ maxTokens: 4096,
9492
9683
  },
9493
9684
  "openai/gpt-oss-20b:free": {
9494
9685
  id: "openai/gpt-oss-20b:free",
@@ -9505,7 +9696,7 @@ export const MODELS = {
9505
9696
  cacheWrite: 0,
9506
9697
  },
9507
9698
  contextWindow: 131072,
9508
- maxTokens: 131072,
9699
+ maxTokens: 8192,
9509
9700
  },
9510
9701
  "openai/gpt-oss-safeguard-20b": {
9511
9702
  id: "openai/gpt-oss-safeguard-20b",
@@ -9791,7 +9982,7 @@ export const MODELS = {
9791
9982
  input: 0.26,
9792
9983
  output: 0.78,
9793
9984
  cacheRead: 0.052000000000000005,
9794
- cacheWrite: 0,
9985
+ cacheWrite: 0.325,
9795
9986
  },
9796
9987
  contextWindow: 1000000,
9797
9988
  maxTokens: 32768,
@@ -9808,7 +9999,7 @@ export const MODELS = {
9808
9999
  input: 0.26,
9809
10000
  output: 0.78,
9810
10001
  cacheRead: 0,
9811
- cacheWrite: 0,
10002
+ cacheWrite: 0.325,
9812
10003
  },
9813
10004
  contextWindow: 1000000,
9814
10005
  maxTokens: 32768,
@@ -9825,7 +10016,7 @@ export const MODELS = {
9825
10016
  input: 0.26,
9826
10017
  output: 0.78,
9827
10018
  cacheRead: 0,
9828
- cacheWrite: 0,
10019
+ cacheWrite: 0.325,
9829
10020
  },
9830
10021
  contextWindow: 1000000,
9831
10022
  maxTokens: 32768,
@@ -9904,7 +10095,7 @@ export const MODELS = {
9904
10095
  api: "openai-completions",
9905
10096
  provider: "openrouter",
9906
10097
  baseUrl: "https://openrouter.ai/api/v1",
9907
- reasoning: true,
10098
+ reasoning: false,
9908
10099
  input: ["text"],
9909
10100
  cost: {
9910
10101
  input: 0.071,
@@ -10063,7 +10254,7 @@ export const MODELS = {
10063
10254
  input: 0.195,
10064
10255
  output: 0.975,
10065
10256
  cacheRead: 0.039,
10066
- cacheWrite: 0,
10257
+ cacheWrite: 0.24375,
10067
10258
  },
10068
10259
  contextWindow: 1000000,
10069
10260
  maxTokens: 65536,
@@ -10077,13 +10268,13 @@ export const MODELS = {
10077
10268
  reasoning: false,
10078
10269
  input: ["text"],
10079
10270
  cost: {
10080
- input: 0.12,
10081
- output: 0.75,
10082
- cacheRead: 0.06,
10271
+ input: 0.15,
10272
+ output: 0.7999999999999999,
10273
+ cacheRead: 0.11,
10083
10274
  cacheWrite: 0,
10084
10275
  },
10085
10276
  contextWindow: 262144,
10086
- maxTokens: 65536,
10277
+ maxTokens: 262144,
10087
10278
  },
10088
10279
  "qwen/qwen3-coder-plus": {
10089
10280
  id: "qwen/qwen3-coder-plus",
@@ -10097,7 +10288,7 @@ export const MODELS = {
10097
10288
  input: 0.65,
10098
10289
  output: 3.25,
10099
10290
  cacheRead: 0.13,
10100
- cacheWrite: 0,
10291
+ cacheWrite: 0.8125,
10101
10292
  },
10102
10293
  contextWindow: 1000000,
10103
10294
  maxTokens: 65536,
@@ -10131,7 +10322,7 @@ export const MODELS = {
10131
10322
  input: 0.78,
10132
10323
  output: 3.9,
10133
10324
  cacheRead: 0.156,
10134
- cacheWrite: 0,
10325
+ cacheWrite: 0.975,
10135
10326
  },
10136
10327
  contextWindow: 262144,
10137
10328
  maxTokens: 32768,
@@ -10385,7 +10576,7 @@ export const MODELS = {
10385
10576
  cost: {
10386
10577
  input: 0.39,
10387
10578
  output: 2.34,
10388
- cacheRead: 0,
10579
+ cacheRead: 0.195,
10389
10580
  cacheWrite: 0,
10390
10581
  },
10391
10582
  contextWindow: 262144,
@@ -10400,13 +10591,13 @@ export const MODELS = {
10400
10591
  reasoning: true,
10401
10592
  input: ["text", "image"],
10402
10593
  cost: {
10403
- input: 0.049999999999999996,
10594
+ input: 0.09999999999999999,
10404
10595
  output: 0.15,
10405
10596
  cacheRead: 0,
10406
10597
  cacheWrite: 0,
10407
10598
  },
10408
- contextWindow: 256000,
10409
- maxTokens: 32768,
10599
+ contextWindow: 262144,
10600
+ maxTokens: 4096,
10410
10601
  },
10411
10602
  "qwen/qwen3.5-flash-02-23": {
10412
10603
  id: "qwen/qwen3.5-flash-02-23",
@@ -10420,7 +10611,7 @@ export const MODELS = {
10420
10611
  input: 0.065,
10421
10612
  output: 0.26,
10422
10613
  cacheRead: 0,
10423
- cacheWrite: 0,
10614
+ cacheWrite: 0.08125,
10424
10615
  },
10425
10616
  contextWindow: 1000000,
10426
10617
  maxTokens: 65536,
@@ -10437,24 +10628,24 @@ export const MODELS = {
10437
10628
  input: 0.26,
10438
10629
  output: 1.56,
10439
10630
  cacheRead: 0,
10440
- cacheWrite: 0,
10631
+ cacheWrite: 0.325,
10441
10632
  },
10442
10633
  contextWindow: 1000000,
10443
10634
  maxTokens: 65536,
10444
10635
  },
10445
- "qwen/qwen3.6-plus:free": {
10446
- id: "qwen/qwen3.6-plus:free",
10447
- name: "Qwen: Qwen3.6 Plus (free)",
10636
+ "qwen/qwen3.6-plus": {
10637
+ id: "qwen/qwen3.6-plus",
10638
+ name: "Qwen: Qwen3.6 Plus",
10448
10639
  api: "openai-completions",
10449
10640
  provider: "openrouter",
10450
10641
  baseUrl: "https://openrouter.ai/api/v1",
10451
10642
  reasoning: true,
10452
10643
  input: ["text", "image"],
10453
10644
  cost: {
10454
- input: 0,
10455
- output: 0,
10645
+ input: 0.325,
10646
+ output: 1.95,
10456
10647
  cacheRead: 0,
10457
- cacheWrite: 0,
10648
+ cacheWrite: 0.40625,
10458
10649
  },
10459
10650
  contextWindow: 1000000,
10460
10651
  maxTokens: 65536,
@@ -10561,9 +10752,9 @@ export const MODELS = {
10561
10752
  contextWindow: 262144,
10562
10753
  maxTokens: 65536,
10563
10754
  },
10564
- "stepfun/step-3.5-flash:free": {
10565
- id: "stepfun/step-3.5-flash:free",
10566
- name: "StepFun: Step 3.5 Flash (free)",
10755
+ "tencent/hy3-preview:free": {
10756
+ id: "tencent/hy3-preview:free",
10757
+ name: "Tencent: Hy3 preview (free)",
10567
10758
  api: "openai-completions",
10568
10759
  provider: "openrouter",
10569
10760
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10575,8 +10766,8 @@ export const MODELS = {
10575
10766
  cacheRead: 0,
10576
10767
  cacheWrite: 0,
10577
10768
  },
10578
- contextWindow: 256000,
10579
- maxTokens: 256000,
10769
+ contextWindow: 262144,
10770
+ maxTokens: 262144,
10580
10771
  },
10581
10772
  "thedrummer/rocinante-12b": {
10582
10773
  id: "thedrummer/rocinante-12b",
@@ -10850,6 +11041,40 @@ export const MODELS = {
10850
11041
  contextWindow: 1048576,
10851
11042
  maxTokens: 131072,
10852
11043
  },
11044
+ "xiaomi/mimo-v2.5": {
11045
+ id: "xiaomi/mimo-v2.5",
11046
+ name: "Xiaomi: MiMo-V2.5",
11047
+ api: "openai-completions",
11048
+ provider: "openrouter",
11049
+ baseUrl: "https://openrouter.ai/api/v1",
11050
+ reasoning: true,
11051
+ input: ["text", "image"],
11052
+ cost: {
11053
+ input: 0.39999999999999997,
11054
+ output: 2,
11055
+ cacheRead: 0.08,
11056
+ cacheWrite: 0,
11057
+ },
11058
+ contextWindow: 1048576,
11059
+ maxTokens: 131072,
11060
+ },
11061
+ "xiaomi/mimo-v2.5-pro": {
11062
+ id: "xiaomi/mimo-v2.5-pro",
11063
+ name: "Xiaomi: MiMo-V2.5-Pro",
11064
+ api: "openai-completions",
11065
+ provider: "openrouter",
11066
+ baseUrl: "https://openrouter.ai/api/v1",
11067
+ reasoning: true,
11068
+ input: ["text"],
11069
+ cost: {
11070
+ input: 1,
11071
+ output: 3,
11072
+ cacheRead: 0.19999999999999998,
11073
+ cacheWrite: 0,
11074
+ },
11075
+ contextWindow: 1048576,
11076
+ maxTokens: 131072,
11077
+ },
10853
11078
  "z-ai/glm-4-32b": {
10854
11079
  id: "z-ai/glm-4-32b",
10855
11080
  name: "Z.ai: GLM 4 32B ",
@@ -10978,13 +11203,13 @@ export const MODELS = {
10978
11203
  reasoning: true,
10979
11204
  input: ["text"],
10980
11205
  cost: {
10981
- input: 0.39,
10982
- output: 1.75,
10983
- cacheRead: 0.195,
11206
+ input: 0.38,
11207
+ output: 1.74,
11208
+ cacheRead: 0,
10984
11209
  cacheWrite: 0,
10985
11210
  },
10986
11211
  contextWindow: 202752,
10987
- maxTokens: 65535,
11212
+ maxTokens: 4096,
10988
11213
  },
10989
11214
  "z-ai/glm-4.7-flash": {
10990
11215
  id: "z-ai/glm-4.7-flash",
@@ -11017,8 +11242,8 @@ export const MODELS = {
11017
11242
  cacheRead: 0.119,
11018
11243
  cacheWrite: 0,
11019
11244
  },
11020
- contextWindow: 80000,
11021
- maxTokens: 131072,
11245
+ contextWindow: 202752,
11246
+ maxTokens: 4096,
11022
11247
  },
11023
11248
  "z-ai/glm-5-turbo": {
11024
11249
  id: "z-ai/glm-5-turbo",
@@ -11037,6 +11262,23 @@ export const MODELS = {
11037
11262
  contextWindow: 202752,
11038
11263
  maxTokens: 131072,
11039
11264
  },
11265
+ "z-ai/glm-5.1": {
11266
+ id: "z-ai/glm-5.1",
11267
+ name: "Z.ai: GLM 5.1",
11268
+ api: "openai-completions",
11269
+ provider: "openrouter",
11270
+ baseUrl: "https://openrouter.ai/api/v1",
11271
+ reasoning: true,
11272
+ input: ["text"],
11273
+ cost: {
11274
+ input: 1.0499999999999998,
11275
+ output: 3.5,
11276
+ cacheRead: 0.5249999999999999,
11277
+ cacheWrite: 0,
11278
+ },
11279
+ contextWindow: 202752,
11280
+ maxTokens: 65535,
11281
+ },
11040
11282
  "z-ai/glm-5v-turbo": {
11041
11283
  id: "z-ai/glm-5v-turbo",
11042
11284
  name: "Z.ai: GLM 5V Turbo",
@@ -11054,6 +11296,23 @@ export const MODELS = {
11054
11296
  contextWindow: 202752,
11055
11297
  maxTokens: 131072,
11056
11298
  },
11299
+ "~anthropic/claude-opus-latest": {
11300
+ id: "~anthropic/claude-opus-latest",
11301
+ name: "Anthropic: Claude Opus Latest",
11302
+ api: "openai-completions",
11303
+ provider: "openrouter",
11304
+ baseUrl: "https://openrouter.ai/api/v1",
11305
+ reasoning: true,
11306
+ input: ["text", "image"],
11307
+ cost: {
11308
+ input: 5,
11309
+ output: 25,
11310
+ cacheRead: 0.5,
11311
+ cacheWrite: 6.25,
11312
+ },
11313
+ contextWindow: 1000000,
11314
+ maxTokens: 128000,
11315
+ },
11057
11316
  },
11058
11317
  "vercel-ai-gateway": {
11059
11318
  "alibaba/qwen-3-14b": {
@@ -11082,13 +11341,13 @@ export const MODELS = {
11082
11341
  reasoning: false,
11083
11342
  input: ["text"],
11084
11343
  cost: {
11085
- input: 0.22,
11086
- output: 0.88,
11087
- cacheRead: 0.11,
11344
+ input: 0.6,
11345
+ output: 1.2,
11346
+ cacheRead: 0.6,
11088
11347
  cacheWrite: 0,
11089
11348
  },
11090
- contextWindow: 32768,
11091
- maxTokens: 16384,
11349
+ contextWindow: 131000,
11350
+ maxTokens: 40000,
11092
11351
  },
11093
11352
  "alibaba/qwen-3-30b": {
11094
11353
  id: "alibaba/qwen-3-30b",
@@ -11124,6 +11383,23 @@ export const MODELS = {
11124
11383
  contextWindow: 128000,
11125
11384
  maxTokens: 8192,
11126
11385
  },
11386
+ "alibaba/qwen-3.6-max-preview": {
11387
+ id: "alibaba/qwen-3.6-max-preview",
11388
+ name: "Qwen 3.6 Max Preview",
11389
+ api: "anthropic-messages",
11390
+ provider: "vercel-ai-gateway",
11391
+ baseUrl: "https://ai-gateway.vercel.sh",
11392
+ reasoning: true,
11393
+ input: ["text", "image"],
11394
+ cost: {
11395
+ input: 1.3,
11396
+ output: 7.8,
11397
+ cacheRead: 0.26,
11398
+ cacheWrite: 1.625,
11399
+ },
11400
+ contextWindow: 240000,
11401
+ maxTokens: 64000,
11402
+ },
11127
11403
  "alibaba/qwen3-235b-a22b-thinking": {
11128
11404
  id: "alibaba/qwen3-235b-a22b-thinking",
11129
11405
  name: "Qwen3 235B A22B Thinking 2507",
@@ -11464,6 +11740,23 @@ export const MODELS = {
11464
11740
  contextWindow: 1000000,
11465
11741
  maxTokens: 128000,
11466
11742
  },
11743
+ "anthropic/claude-opus-4.7": {
11744
+ id: "anthropic/claude-opus-4.7",
11745
+ name: "Claude Opus 4.7",
11746
+ api: "anthropic-messages",
11747
+ provider: "vercel-ai-gateway",
11748
+ baseUrl: "https://ai-gateway.vercel.sh",
11749
+ reasoning: true,
11750
+ input: ["text", "image"],
11751
+ cost: {
11752
+ input: 5,
11753
+ output: 25,
11754
+ cacheRead: 0.5,
11755
+ cacheWrite: 6.25,
11756
+ },
11757
+ contextWindow: 1000000,
11758
+ maxTokens: 128000,
11759
+ },
11467
11760
  "anthropic/claude-sonnet-4": {
11468
11761
  id: "anthropic/claude-sonnet-4",
11469
11762
  name: "Claude Sonnet 4",
@@ -11815,7 +12108,7 @@ export const MODELS = {
11815
12108
  cost: {
11816
12109
  input: 0.25,
11817
12110
  output: 1.5,
11818
- cacheRead: 0,
12111
+ cacheRead: 0.03,
11819
12112
  cacheWrite: 0,
11820
12113
  },
11821
12114
  contextWindow: 1000000,
@@ -11938,7 +12231,7 @@ export const MODELS = {
11938
12231
  cacheWrite: 0,
11939
12232
  },
11940
12233
  contextWindow: 128000,
11941
- maxTokens: 8192,
12234
+ maxTokens: 100000,
11942
12235
  },
11943
12236
  "meta/llama-3.1-70b": {
11944
12237
  id: "meta/llama-3.1-70b",
@@ -12450,6 +12743,23 @@ export const MODELS = {
12450
12743
  contextWindow: 262114,
12451
12744
  maxTokens: 262114,
12452
12745
  },
12746
+ "moonshotai/kimi-k2.6": {
12747
+ id: "moonshotai/kimi-k2.6",
12748
+ name: "Kimi K2.6",
12749
+ api: "anthropic-messages",
12750
+ provider: "vercel-ai-gateway",
12751
+ baseUrl: "https://ai-gateway.vercel.sh",
12752
+ reasoning: true,
12753
+ input: ["text", "image"],
12754
+ cost: {
12755
+ input: 0.95,
12756
+ output: 4,
12757
+ cacheRead: 0.16,
12758
+ cacheWrite: 0,
12759
+ },
12760
+ contextWindow: 262000,
12761
+ maxTokens: 262000,
12762
+ },
12453
12763
  "nvidia/nemotron-nano-12b-v2-vl": {
12454
12764
  id: "nvidia/nemotron-nano-12b-v2-vl",
12455
12765
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -12943,26 +13253,9 @@ export const MODELS = {
12943
13253
  contextWindow: 1050000,
12944
13254
  maxTokens: 128000,
12945
13255
  },
12946
- "openai/gpt-oss-120b": {
12947
- id: "openai/gpt-oss-120b",
12948
- name: "gpt-oss-120b",
12949
- api: "anthropic-messages",
12950
- provider: "vercel-ai-gateway",
12951
- baseUrl: "https://ai-gateway.vercel.sh",
12952
- reasoning: true,
12953
- input: ["text"],
12954
- cost: {
12955
- input: 0.15,
12956
- output: 0.6,
12957
- cacheRead: 0,
12958
- cacheWrite: 0,
12959
- },
12960
- contextWindow: 128000,
12961
- maxTokens: 8192,
12962
- },
12963
13256
  "openai/gpt-oss-20b": {
12964
13257
  id: "openai/gpt-oss-20b",
12965
- name: "gpt-oss-20b",
13258
+ name: "GPT OSS 120B",
12966
13259
  api: "anthropic-messages",
12967
13260
  provider: "vercel-ai-gateway",
12968
13261
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -12979,7 +13272,7 @@ export const MODELS = {
12979
13272
  },
12980
13273
  "openai/gpt-oss-safeguard-20b": {
12981
13274
  id: "openai/gpt-oss-safeguard-20b",
12982
- name: "gpt-oss-safeguard-20b",
13275
+ name: "GPT OSS Safeguard 20B",
12983
13276
  api: "anthropic-messages",
12984
13277
  provider: "vercel-ai-gateway",
12985
13278
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13147,23 +13440,6 @@ export const MODELS = {
13147
13440
  contextWindow: 131072,
13148
13441
  maxTokens: 131072,
13149
13442
  },
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
13443
  "xai/grok-3": {
13168
13444
  id: "xai/grok-3",
13169
13445
  name: "Grok 3 Beta",
@@ -13256,7 +13532,7 @@ export const MODELS = {
13256
13532
  provider: "vercel-ai-gateway",
13257
13533
  baseUrl: "https://ai-gateway.vercel.sh",
13258
13534
  reasoning: false,
13259
- input: ["text"],
13535
+ input: ["text", "image"],
13260
13536
  cost: {
13261
13537
  input: 0.19999999999999998,
13262
13538
  output: 0.5,
@@ -13273,7 +13549,7 @@ export const MODELS = {
13273
13549
  provider: "vercel-ai-gateway",
13274
13550
  baseUrl: "https://ai-gateway.vercel.sh",
13275
13551
  reasoning: true,
13276
- input: ["text"],
13552
+ input: ["text", "image"],
13277
13553
  cost: {
13278
13554
  input: 0.19999999999999998,
13279
13555
  output: 0.5,
@@ -13290,7 +13566,7 @@ export const MODELS = {
13290
13566
  provider: "vercel-ai-gateway",
13291
13567
  baseUrl: "https://ai-gateway.vercel.sh",
13292
13568
  reasoning: false,
13293
- input: ["text"],
13569
+ input: ["text", "image"],
13294
13570
  cost: {
13295
13571
  input: 0.19999999999999998,
13296
13572
  output: 0.5,
@@ -13307,7 +13583,7 @@ export const MODELS = {
13307
13583
  provider: "vercel-ai-gateway",
13308
13584
  baseUrl: "https://ai-gateway.vercel.sh",
13309
13585
  reasoning: true,
13310
- input: ["text"],
13586
+ input: ["text", "image"],
13311
13587
  cost: {
13312
13588
  input: 0.19999999999999998,
13313
13589
  output: 0.5,
@@ -13324,7 +13600,7 @@ export const MODELS = {
13324
13600
  provider: "vercel-ai-gateway",
13325
13601
  baseUrl: "https://ai-gateway.vercel.sh",
13326
13602
  reasoning: true,
13327
- input: ["text"],
13603
+ input: ["text", "image"],
13328
13604
  cost: {
13329
13605
  input: 2,
13330
13606
  output: 6,
@@ -13341,7 +13617,7 @@ export const MODELS = {
13341
13617
  provider: "vercel-ai-gateway",
13342
13618
  baseUrl: "https://ai-gateway.vercel.sh",
13343
13619
  reasoning: true,
13344
- input: ["text"],
13620
+ input: ["text", "image"],
13345
13621
  cost: {
13346
13622
  input: 2,
13347
13623
  output: 6,
@@ -13445,9 +13721,9 @@ export const MODELS = {
13445
13721
  reasoning: true,
13446
13722
  input: ["text"],
13447
13723
  cost: {
13448
- input: 0.09,
13449
- output: 0.29,
13450
- cacheRead: 0.045,
13724
+ input: 0.09999999999999999,
13725
+ output: 0.3,
13726
+ cacheRead: 0.01,
13451
13727
  cacheWrite: 0,
13452
13728
  },
13453
13729
  contextWindow: 262144,
@@ -13581,13 +13857,13 @@ export const MODELS = {
13581
13857
  reasoning: true,
13582
13858
  input: ["text"],
13583
13859
  cost: {
13584
- input: 0.6,
13585
- output: 2.2,
13586
- cacheRead: 0.11,
13860
+ input: 2.25,
13861
+ output: 2.75,
13862
+ cacheRead: 2.25,
13587
13863
  cacheWrite: 0,
13588
13864
  },
13589
- contextWindow: 200000,
13590
- maxTokens: 120000,
13865
+ contextWindow: 131000,
13866
+ maxTokens: 40000,
13591
13867
  },
13592
13868
  "zai/glm-4.7-flash": {
13593
13869
  id: "zai/glm-4.7-flash",
@@ -13657,6 +13933,23 @@ export const MODELS = {
13657
13933
  contextWindow: 202800,
13658
13934
  maxTokens: 131100,
13659
13935
  },
13936
+ "zai/glm-5.1": {
13937
+ id: "zai/glm-5.1",
13938
+ name: "GLM 5.1",
13939
+ api: "anthropic-messages",
13940
+ provider: "vercel-ai-gateway",
13941
+ baseUrl: "https://ai-gateway.vercel.sh",
13942
+ reasoning: true,
13943
+ input: ["text"],
13944
+ cost: {
13945
+ input: 1.4,
13946
+ output: 4.4,
13947
+ cacheRead: 0.26,
13948
+ cacheWrite: 0,
13949
+ },
13950
+ contextWindow: 202800,
13951
+ maxTokens: 64000,
13952
+ },
13660
13953
  "zai/glm-5v-turbo": {
13661
13954
  id: "zai/glm-5v-turbo",
13662
13955
  name: "GLM 5V Turbo",
@@ -14096,9 +14389,9 @@ export const MODELS = {
14096
14389
  reasoning: true,
14097
14390
  input: ["text"],
14098
14391
  cost: {
14099
- input: 0.6,
14100
- output: 2.2,
14101
- cacheRead: 0.11,
14392
+ input: 0,
14393
+ output: 0,
14394
+ cacheRead: 0,
14102
14395
  cacheWrite: 0,
14103
14396
  },
14104
14397
  contextWindow: 131072,
@@ -14114,9 +14407,9 @@ export const MODELS = {
14114
14407
  reasoning: true,
14115
14408
  input: ["text"],
14116
14409
  cost: {
14117
- input: 0.2,
14118
- output: 1.1,
14119
- cacheRead: 0.03,
14410
+ input: 0,
14411
+ output: 0,
14412
+ cacheRead: 0,
14120
14413
  cacheWrite: 0,
14121
14414
  },
14122
14415
  contextWindow: 131072,
@@ -14150,8 +14443,8 @@ export const MODELS = {
14150
14443
  reasoning: true,
14151
14444
  input: ["text", "image"],
14152
14445
  cost: {
14153
- input: 0.6,
14154
- output: 1.8,
14446
+ input: 0,
14447
+ output: 0,
14155
14448
  cacheRead: 0,
14156
14449
  cacheWrite: 0,
14157
14450
  },
@@ -14168,9 +14461,9 @@ export const MODELS = {
14168
14461
  reasoning: true,
14169
14462
  input: ["text"],
14170
14463
  cost: {
14171
- input: 0.6,
14172
- output: 2.2,
14173
- cacheRead: 0.11,
14464
+ input: 0,
14465
+ output: 0,
14466
+ cacheRead: 0,
14174
14467
  cacheWrite: 0,
14175
14468
  },
14176
14469
  contextWindow: 204800,
@@ -14186,8 +14479,8 @@ export const MODELS = {
14186
14479
  reasoning: true,
14187
14480
  input: ["text", "image"],
14188
14481
  cost: {
14189
- input: 0.3,
14190
- output: 0.9,
14482
+ input: 0,
14483
+ output: 0,
14191
14484
  cacheRead: 0,
14192
14485
  cacheWrite: 0,
14193
14486
  },
@@ -14204,9 +14497,9 @@ export const MODELS = {
14204
14497
  reasoning: true,
14205
14498
  input: ["text"],
14206
14499
  cost: {
14207
- input: 0.6,
14208
- output: 2.2,
14209
- cacheRead: 0.11,
14500
+ input: 0,
14501
+ output: 0,
14502
+ cacheRead: 0,
14210
14503
  cacheWrite: 0,
14211
14504
  },
14212
14505
  contextWindow: 204800,
@@ -14258,9 +14551,9 @@ export const MODELS = {
14258
14551
  reasoning: true,
14259
14552
  input: ["text"],
14260
14553
  cost: {
14261
- input: 1,
14262
- output: 3.2,
14263
- cacheRead: 0.2,
14554
+ input: 0,
14555
+ output: 0,
14556
+ cacheRead: 0,
14264
14557
  cacheWrite: 0,
14265
14558
  },
14266
14559
  contextWindow: 204800,
@@ -14276,9 +14569,27 @@ export const MODELS = {
14276
14569
  reasoning: true,
14277
14570
  input: ["text"],
14278
14571
  cost: {
14279
- input: 1.2,
14280
- output: 4,
14281
- cacheRead: 0.24,
14572
+ input: 0,
14573
+ output: 0,
14574
+ cacheRead: 0,
14575
+ cacheWrite: 0,
14576
+ },
14577
+ contextWindow: 200000,
14578
+ maxTokens: 131072,
14579
+ },
14580
+ "glm-5.1": {
14581
+ id: "glm-5.1",
14582
+ name: "GLM-5.1",
14583
+ api: "openai-completions",
14584
+ provider: "zai",
14585
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
14586
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
14587
+ reasoning: true,
14588
+ input: ["text"],
14589
+ cost: {
14590
+ input: 0,
14591
+ output: 0,
14592
+ cacheRead: 0,
14282
14593
  cacheWrite: 0,
14283
14594
  },
14284
14595
  contextWindow: 200000,
@@ -14294,9 +14605,9 @@ export const MODELS = {
14294
14605
  reasoning: true,
14295
14606
  input: ["text", "image"],
14296
14607
  cost: {
14297
- input: 1.2,
14298
- output: 4,
14299
- cacheRead: 0.24,
14608
+ input: 0,
14609
+ output: 0,
14610
+ cacheRead: 0,
14300
14611
  cacheWrite: 0,
14301
14612
  },
14302
14613
  contextWindow: 200000,