@elyracode/ai 0.9.10 → 0.9.12

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.
@@ -217,6 +217,24 @@ export const MODELS = {
217
217
  contextWindow: 1000000,
218
218
  maxTokens: 64000,
219
219
  },
220
+ "anthropic.claude-sonnet-5": {
221
+ id: "anthropic.claude-sonnet-5",
222
+ name: "Claude Sonnet 5",
223
+ api: "bedrock-converse-stream",
224
+ provider: "amazon-bedrock",
225
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
226
+ reasoning: true,
227
+ thinkingType: "adaptive",
228
+ input: ["text", "image"],
229
+ cost: {
230
+ input: 2,
231
+ output: 10,
232
+ cacheRead: 0.2,
233
+ cacheWrite: 2.5,
234
+ },
235
+ contextWindow: 1000000,
236
+ maxTokens: 128000,
237
+ },
220
238
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
221
239
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
222
240
  name: "Claude Haiku 4.5 (AU)",
@@ -309,6 +327,24 @@ export const MODELS = {
309
327
  contextWindow: 1000000,
310
328
  maxTokens: 128000,
311
329
  },
330
+ "au.anthropic.claude-sonnet-5": {
331
+ id: "au.anthropic.claude-sonnet-5",
332
+ name: "Claude Sonnet 5 (AU)",
333
+ api: "bedrock-converse-stream",
334
+ provider: "amazon-bedrock",
335
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
336
+ reasoning: true,
337
+ thinkingType: "adaptive",
338
+ input: ["text", "image"],
339
+ cost: {
340
+ input: 2,
341
+ output: 10,
342
+ cacheRead: 0.2,
343
+ cacheWrite: 2.5,
344
+ },
345
+ contextWindow: 1000000,
346
+ maxTokens: 128000,
347
+ },
312
348
  "deepseek.r1-v1:0": {
313
349
  id: "deepseek.r1-v1:0",
314
350
  name: "DeepSeek-R1",
@@ -389,10 +425,10 @@ export const MODELS = {
389
425
  thinkingType: "budget",
390
426
  input: ["text", "image"],
391
427
  cost: {
392
- input: 1,
393
- output: 5,
394
- cacheRead: 0.1,
395
- cacheWrite: 1.25,
428
+ input: 1.1,
429
+ output: 5.5,
430
+ cacheRead: 0.11,
431
+ cacheWrite: 1.375,
396
432
  },
397
433
  contextWindow: 200000,
398
434
  maxTokens: 64000,
@@ -407,10 +443,10 @@ export const MODELS = {
407
443
  thinkingType: "budget",
408
444
  input: ["text", "image"],
409
445
  cost: {
410
- input: 5,
411
- output: 25,
412
- cacheRead: 0.5,
413
- cacheWrite: 6.25,
446
+ input: 5.5,
447
+ output: 27.5,
448
+ cacheRead: 0.55,
449
+ cacheWrite: 6.875,
414
450
  },
415
451
  contextWindow: 200000,
416
452
  maxTokens: 64000,
@@ -508,6 +544,24 @@ export const MODELS = {
508
544
  contextWindow: 1000000,
509
545
  maxTokens: 64000,
510
546
  },
547
+ "eu.anthropic.claude-sonnet-5": {
548
+ id: "eu.anthropic.claude-sonnet-5",
549
+ name: "Claude Sonnet 5 (EU)",
550
+ api: "bedrock-converse-stream",
551
+ provider: "amazon-bedrock",
552
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
553
+ reasoning: true,
554
+ thinkingType: "adaptive",
555
+ input: ["text", "image"],
556
+ cost: {
557
+ input: 2.2,
558
+ output: 11,
559
+ cacheRead: 0.22,
560
+ cacheWrite: 2.75,
561
+ },
562
+ contextWindow: 1000000,
563
+ maxTokens: 128000,
564
+ },
511
565
  "global.anthropic.claude-fable-5": {
512
566
  id: "global.anthropic.claude-fable-5",
513
567
  name: "Claude Fable 5 (Global)",
@@ -656,6 +710,24 @@ export const MODELS = {
656
710
  contextWindow: 1000000,
657
711
  maxTokens: 64000,
658
712
  },
713
+ "global.anthropic.claude-sonnet-5": {
714
+ id: "global.anthropic.claude-sonnet-5",
715
+ name: "Claude Sonnet 5 (Global)",
716
+ api: "bedrock-converse-stream",
717
+ provider: "amazon-bedrock",
718
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
719
+ reasoning: true,
720
+ thinkingType: "adaptive",
721
+ input: ["text", "image"],
722
+ cost: {
723
+ input: 2,
724
+ output: 10,
725
+ cacheRead: 0.2,
726
+ cacheWrite: 2.5,
727
+ },
728
+ contextWindow: 1000000,
729
+ maxTokens: 128000,
730
+ },
659
731
  "google.gemma-3-27b-it": {
660
732
  id: "google.gemma-3-27b-it",
661
733
  name: "Google Gemma 3 27B Instruct",
@@ -764,6 +836,24 @@ export const MODELS = {
764
836
  contextWindow: 1000000,
765
837
  maxTokens: 64000,
766
838
  },
839
+ "jp.anthropic.claude-sonnet-5": {
840
+ id: "jp.anthropic.claude-sonnet-5",
841
+ name: "Claude Sonnet 5 (JP)",
842
+ api: "bedrock-converse-stream",
843
+ provider: "amazon-bedrock",
844
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
845
+ reasoning: true,
846
+ thinkingType: "adaptive",
847
+ input: ["text", "image"],
848
+ cost: {
849
+ input: 2,
850
+ output: 10,
851
+ cacheRead: 0.2,
852
+ cacheWrite: 2.5,
853
+ },
854
+ contextWindow: 1000000,
855
+ maxTokens: 128000,
856
+ },
767
857
  "meta.llama3-1-70b-instruct-v1:0": {
768
858
  id: "meta.llama3-1-70b-instruct-v1:0",
769
859
  name: "Llama 3.1 70B Instruct",
@@ -1578,6 +1668,24 @@ export const MODELS = {
1578
1668
  contextWindow: 1000000,
1579
1669
  maxTokens: 64000,
1580
1670
  },
1671
+ "us.anthropic.claude-sonnet-5": {
1672
+ id: "us.anthropic.claude-sonnet-5",
1673
+ name: "Claude Sonnet 5 (US)",
1674
+ api: "bedrock-converse-stream",
1675
+ provider: "amazon-bedrock",
1676
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1677
+ reasoning: true,
1678
+ thinkingType: "adaptive",
1679
+ input: ["text", "image"],
1680
+ cost: {
1681
+ input: 2,
1682
+ output: 10,
1683
+ cacheRead: 0.2,
1684
+ cacheWrite: 2.5,
1685
+ },
1686
+ contextWindow: 1000000,
1687
+ maxTokens: 128000,
1688
+ },
1581
1689
  "us.deepseek.r1-v1:0": {
1582
1690
  id: "us.deepseek.r1-v1:0",
1583
1691
  name: "DeepSeek-R1 (US)",
@@ -1663,6 +1771,23 @@ export const MODELS = {
1663
1771
  contextWindow: 1040000,
1664
1772
  maxTokens: 8192,
1665
1773
  },
1774
+ "xai.grok-4.3": {
1775
+ id: "xai.grok-4.3",
1776
+ name: "Grok 4.3",
1777
+ api: "bedrock-converse-stream",
1778
+ provider: "amazon-bedrock",
1779
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1780
+ reasoning: true,
1781
+ input: ["text", "image"],
1782
+ cost: {
1783
+ input: 1.25,
1784
+ output: 2.5,
1785
+ cacheRead: 0.2,
1786
+ cacheWrite: 0,
1787
+ },
1788
+ contextWindow: 1000000,
1789
+ maxTokens: 131072,
1790
+ },
1666
1791
  "zai.glm-4.7": {
1667
1792
  id: "zai.glm-4.7",
1668
1793
  name: "GLM-4.7",
@@ -2129,6 +2254,24 @@ export const MODELS = {
2129
2254
  contextWindow: 1000000,
2130
2255
  maxTokens: 64000,
2131
2256
  },
2257
+ "claude-sonnet-5": {
2258
+ id: "claude-sonnet-5",
2259
+ name: "Claude Sonnet 5",
2260
+ api: "anthropic-messages",
2261
+ provider: "anthropic",
2262
+ baseUrl: "https://api.anthropic.com",
2263
+ reasoning: true,
2264
+ thinkingType: "adaptive",
2265
+ input: ["text", "image"],
2266
+ cost: {
2267
+ input: 2,
2268
+ output: 10,
2269
+ cacheRead: 0.2,
2270
+ cacheWrite: 2.5,
2271
+ },
2272
+ contextWindow: 1000000,
2273
+ maxTokens: 128000,
2274
+ },
2132
2275
  },
2133
2276
  "azure-openai-responses": {
2134
2277
  "gpt-4": {
@@ -3775,7 +3918,7 @@ export const MODELS = {
3775
3918
  cacheRead: 0.26,
3776
3919
  cacheWrite: 0,
3777
3920
  },
3778
- contextWindow: 1048576,
3921
+ contextWindow: 1048575,
3779
3922
  maxTokens: 131072,
3780
3923
  },
3781
3924
  "accounts/fireworks/models/gpt-oss-120b": {
@@ -3914,6 +4057,23 @@ export const MODELS = {
3914
4057
  contextWindow: 202800,
3915
4058
  maxTokens: 131072,
3916
4059
  },
4060
+ "accounts/fireworks/routers/glm-5p2-fast": {
4061
+ id: "accounts/fireworks/routers/glm-5p2-fast",
4062
+ name: "GLM 5.2 Fast",
4063
+ api: "anthropic-messages",
4064
+ provider: "fireworks",
4065
+ baseUrl: "https://api.fireworks.ai/inference",
4066
+ reasoning: true,
4067
+ input: ["text"],
4068
+ cost: {
4069
+ input: 2.1,
4070
+ output: 6.6,
4071
+ cacheRead: 0.21,
4072
+ cacheWrite: 0,
4073
+ },
4074
+ contextWindow: 1048575,
4075
+ maxTokens: 131072,
4076
+ },
3917
4077
  "accounts/fireworks/routers/kimi-k2p6-fast": {
3918
4078
  id: "accounts/fireworks/routers/kimi-k2p6-fast",
3919
4079
  name: "Kimi K2.6 Fast",
@@ -4667,42 +4827,6 @@ export const MODELS = {
4667
4827
  contextWindow: 262144,
4668
4828
  maxTokens: 32768,
4669
4829
  },
4670
- "gemma-4-E2B-it": {
4671
- id: "gemma-4-E2B-it",
4672
- name: "Gemma 4 E2B IT",
4673
- api: "google-generative-ai",
4674
- provider: "google",
4675
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4676
- reasoning: true,
4677
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4678
- input: ["text", "image"],
4679
- cost: {
4680
- input: 0,
4681
- output: 0,
4682
- cacheRead: 0,
4683
- cacheWrite: 0,
4684
- },
4685
- contextWindow: 131072,
4686
- maxTokens: 8192,
4687
- },
4688
- "gemma-4-E4B-it": {
4689
- id: "gemma-4-E4B-it",
4690
- name: "Gemma 4 E4B IT",
4691
- api: "google-generative-ai",
4692
- provider: "google",
4693
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4694
- reasoning: true,
4695
- thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4696
- input: ["text", "image"],
4697
- cost: {
4698
- input: 0,
4699
- output: 0,
4700
- cacheRead: 0,
4701
- cacheWrite: 0,
4702
- },
4703
- contextWindow: 131072,
4704
- maxTokens: 8192,
4705
- },
4706
4830
  },
4707
4831
  "google-vertex": {
4708
4832
  "gemini-1.5-flash": {
@@ -5028,7 +5152,7 @@ export const MODELS = {
5028
5152
  cost: {
5029
5153
  input: 0.075,
5030
5154
  output: 0.3,
5031
- cacheRead: 0.037,
5155
+ cacheRead: 0,
5032
5156
  cacheWrite: 0,
5033
5157
  },
5034
5158
  contextWindow: 131072,
@@ -5054,6 +5178,24 @@ export const MODELS = {
5054
5178
  },
5055
5179
  },
5056
5180
  "huggingface": {
5181
+ "MiniMaxAI/MiniMax-M2": {
5182
+ id: "MiniMaxAI/MiniMax-M2",
5183
+ name: "MiniMax-M2",
5184
+ api: "openai-completions",
5185
+ provider: "huggingface",
5186
+ baseUrl: "https://router.huggingface.co/v1",
5187
+ compat: { "supportsDeveloperRole": false },
5188
+ reasoning: true,
5189
+ input: ["text"],
5190
+ cost: {
5191
+ input: 0.3,
5192
+ output: 1.2,
5193
+ cacheRead: 0,
5194
+ cacheWrite: 0,
5195
+ },
5196
+ contextWindow: 204800,
5197
+ maxTokens: 128000,
5198
+ },
5057
5199
  "MiniMaxAI/MiniMax-M2.1": {
5058
5200
  id: "MiniMaxAI/MiniMax-M2.1",
5059
5201
  name: "MiniMax-M2.1",
@@ -5108,6 +5250,42 @@ export const MODELS = {
5108
5250
  contextWindow: 204800,
5109
5251
  maxTokens: 131072,
5110
5252
  },
5253
+ "MiniMaxAI/MiniMax-M3": {
5254
+ id: "MiniMaxAI/MiniMax-M3",
5255
+ name: "MiniMax-M3",
5256
+ api: "openai-completions",
5257
+ provider: "huggingface",
5258
+ baseUrl: "https://router.huggingface.co/v1",
5259
+ compat: { "supportsDeveloperRole": false },
5260
+ reasoning: true,
5261
+ input: ["text", "image"],
5262
+ cost: {
5263
+ input: 0.3,
5264
+ output: 1.2,
5265
+ cacheRead: 0,
5266
+ cacheWrite: 0,
5267
+ },
5268
+ contextWindow: 524288,
5269
+ maxTokens: 128000,
5270
+ },
5271
+ "Qwen/Qwen3-235B-A22B": {
5272
+ id: "Qwen/Qwen3-235B-A22B",
5273
+ name: "Qwen3 235B-A22B",
5274
+ api: "openai-completions",
5275
+ provider: "huggingface",
5276
+ baseUrl: "https://router.huggingface.co/v1",
5277
+ compat: { "supportsDeveloperRole": false },
5278
+ reasoning: true,
5279
+ input: ["text"],
5280
+ cost: {
5281
+ input: 0.2,
5282
+ output: 0.8,
5283
+ cacheRead: 0,
5284
+ cacheWrite: 0,
5285
+ },
5286
+ contextWindow: 40960,
5287
+ maxTokens: 16384,
5288
+ },
5111
5289
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5112
5290
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5113
5291
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5126,6 +5304,42 @@ export const MODELS = {
5126
5304
  contextWindow: 262144,
5127
5305
  maxTokens: 131072,
5128
5306
  },
5307
+ "Qwen/Qwen3-32B": {
5308
+ id: "Qwen/Qwen3-32B",
5309
+ name: "Qwen3 32B",
5310
+ api: "openai-completions",
5311
+ provider: "huggingface",
5312
+ baseUrl: "https://router.huggingface.co/v1",
5313
+ compat: { "supportsDeveloperRole": false },
5314
+ reasoning: true,
5315
+ input: ["text"],
5316
+ cost: {
5317
+ input: 0.29,
5318
+ output: 0.59,
5319
+ cacheRead: 0,
5320
+ cacheWrite: 0,
5321
+ },
5322
+ contextWindow: 131072,
5323
+ maxTokens: 16384,
5324
+ },
5325
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
5326
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
5327
+ name: "Qwen3-Coder 30B-A3B Instruct",
5328
+ api: "openai-completions",
5329
+ provider: "huggingface",
5330
+ baseUrl: "https://router.huggingface.co/v1",
5331
+ compat: { "supportsDeveloperRole": false },
5332
+ reasoning: false,
5333
+ input: ["text"],
5334
+ cost: {
5335
+ input: 0.07,
5336
+ output: 0.26,
5337
+ cacheRead: 0,
5338
+ cacheWrite: 0,
5339
+ },
5340
+ contextWindow: 262144,
5341
+ maxTokens: 65536,
5342
+ },
5129
5343
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5130
5344
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5131
5345
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5198,9 +5412,9 @@ export const MODELS = {
5198
5412
  contextWindow: 262144,
5199
5413
  maxTokens: 131072,
5200
5414
  },
5201
- "Qwen/Qwen3.5-397B-A17B": {
5202
- id: "Qwen/Qwen3.5-397B-A17B",
5203
- name: "Qwen3.5-397B-A17B",
5415
+ "Qwen/Qwen3.5-122B-A10B": {
5416
+ id: "Qwen/Qwen3.5-122B-A10B",
5417
+ name: "Qwen3.5 122B-A10B",
5204
5418
  api: "openai-completions",
5205
5419
  provider: "huggingface",
5206
5420
  baseUrl: "https://router.huggingface.co/v1",
@@ -5208,17 +5422,161 @@ export const MODELS = {
5208
5422
  reasoning: true,
5209
5423
  input: ["text", "image"],
5210
5424
  cost: {
5211
- input: 0.6,
5212
- output: 3.6,
5425
+ input: 0.4,
5426
+ output: 3.2,
5213
5427
  cacheRead: 0,
5214
5428
  cacheWrite: 0,
5215
5429
  },
5216
5430
  contextWindow: 262144,
5217
- maxTokens: 32768,
5431
+ maxTokens: 65536,
5218
5432
  },
5219
- "deepseek-ai/DeepSeek-R1-0528": {
5220
- id: "deepseek-ai/DeepSeek-R1-0528",
5221
- name: "DeepSeek-R1-0528",
5433
+ "Qwen/Qwen3.5-27B": {
5434
+ id: "Qwen/Qwen3.5-27B",
5435
+ name: "Qwen3.5 27B",
5436
+ api: "openai-completions",
5437
+ provider: "huggingface",
5438
+ baseUrl: "https://router.huggingface.co/v1",
5439
+ compat: { "supportsDeveloperRole": false },
5440
+ reasoning: true,
5441
+ input: ["text", "image"],
5442
+ cost: {
5443
+ input: 0.3,
5444
+ output: 2.4,
5445
+ cacheRead: 0,
5446
+ cacheWrite: 0,
5447
+ },
5448
+ contextWindow: 262144,
5449
+ maxTokens: 65536,
5450
+ },
5451
+ "Qwen/Qwen3.5-35B-A3B": {
5452
+ id: "Qwen/Qwen3.5-35B-A3B",
5453
+ name: "Qwen3.5 35B-A3B",
5454
+ api: "openai-completions",
5455
+ provider: "huggingface",
5456
+ baseUrl: "https://router.huggingface.co/v1",
5457
+ compat: { "supportsDeveloperRole": false },
5458
+ reasoning: true,
5459
+ input: ["text", "image"],
5460
+ cost: {
5461
+ input: 0.25,
5462
+ output: 2,
5463
+ cacheRead: 0,
5464
+ cacheWrite: 0,
5465
+ },
5466
+ contextWindow: 262144,
5467
+ maxTokens: 65536,
5468
+ },
5469
+ "Qwen/Qwen3.5-397B-A17B": {
5470
+ id: "Qwen/Qwen3.5-397B-A17B",
5471
+ name: "Qwen3.5-397B-A17B",
5472
+ api: "openai-completions",
5473
+ provider: "huggingface",
5474
+ baseUrl: "https://router.huggingface.co/v1",
5475
+ compat: { "supportsDeveloperRole": false },
5476
+ reasoning: true,
5477
+ input: ["text", "image"],
5478
+ cost: {
5479
+ input: 0.6,
5480
+ output: 3.6,
5481
+ cacheRead: 0,
5482
+ cacheWrite: 0,
5483
+ },
5484
+ contextWindow: 262144,
5485
+ maxTokens: 32768,
5486
+ },
5487
+ "Qwen/Qwen3.5-9B": {
5488
+ id: "Qwen/Qwen3.5-9B",
5489
+ name: "Qwen3.5 9B",
5490
+ api: "openai-completions",
5491
+ provider: "huggingface",
5492
+ baseUrl: "https://router.huggingface.co/v1",
5493
+ compat: { "supportsDeveloperRole": false },
5494
+ reasoning: true,
5495
+ input: ["text", "image"],
5496
+ cost: {
5497
+ input: 0.17,
5498
+ output: 0.25,
5499
+ cacheRead: 0,
5500
+ cacheWrite: 0,
5501
+ },
5502
+ contextWindow: 262144,
5503
+ maxTokens: 65536,
5504
+ },
5505
+ "Qwen/Qwen3.6-27B": {
5506
+ id: "Qwen/Qwen3.6-27B",
5507
+ name: "Qwen3.6 27B",
5508
+ api: "openai-completions",
5509
+ provider: "huggingface",
5510
+ baseUrl: "https://router.huggingface.co/v1",
5511
+ compat: { "supportsDeveloperRole": false },
5512
+ reasoning: true,
5513
+ input: ["text", "image"],
5514
+ cost: {
5515
+ input: 0.47,
5516
+ output: 3.19,
5517
+ cacheRead: 0,
5518
+ cacheWrite: 0,
5519
+ },
5520
+ contextWindow: 262144,
5521
+ maxTokens: 65536,
5522
+ },
5523
+ "Qwen/Qwen3.6-35B-A3B": {
5524
+ id: "Qwen/Qwen3.6-35B-A3B",
5525
+ name: "Qwen3.6 35B-A3B",
5526
+ api: "openai-completions",
5527
+ provider: "huggingface",
5528
+ baseUrl: "https://router.huggingface.co/v1",
5529
+ compat: { "supportsDeveloperRole": false },
5530
+ reasoning: true,
5531
+ input: ["text", "image"],
5532
+ cost: {
5533
+ input: 0.15,
5534
+ output: 0.95,
5535
+ cacheRead: 0,
5536
+ cacheWrite: 0,
5537
+ },
5538
+ contextWindow: 262144,
5539
+ maxTokens: 65536,
5540
+ },
5541
+ "XiaomiMiMo/MiMo-V2.5-Pro": {
5542
+ id: "XiaomiMiMo/MiMo-V2.5-Pro",
5543
+ name: "MiMo-V2.5-Pro",
5544
+ api: "openai-completions",
5545
+ provider: "huggingface",
5546
+ baseUrl: "https://router.huggingface.co/v1",
5547
+ compat: { "supportsDeveloperRole": false },
5548
+ reasoning: true,
5549
+ input: ["text"],
5550
+ cost: {
5551
+ input: 1,
5552
+ output: 3,
5553
+ cacheRead: 0,
5554
+ cacheWrite: 0,
5555
+ },
5556
+ contextWindow: 1048576,
5557
+ maxTokens: 131072,
5558
+ },
5559
+ "deepseek-ai/DeepSeek-R1": {
5560
+ id: "deepseek-ai/DeepSeek-R1",
5561
+ name: "DeepSeek-R1",
5562
+ api: "openai-completions",
5563
+ provider: "huggingface",
5564
+ baseUrl: "https://router.huggingface.co/v1",
5565
+ compat: { "supportsDeveloperRole": false },
5566
+ reasoning: true,
5567
+ input: ["text"],
5568
+ cost: {
5569
+ input: 0.7,
5570
+ output: 2.5,
5571
+ cacheRead: 0,
5572
+ cacheWrite: 0,
5573
+ },
5574
+ contextWindow: 64000,
5575
+ maxTokens: 32768,
5576
+ },
5577
+ "deepseek-ai/DeepSeek-R1-0528": {
5578
+ id: "deepseek-ai/DeepSeek-R1-0528",
5579
+ name: "DeepSeek-R1-0528",
5222
5580
  api: "openai-completions",
5223
5581
  provider: "huggingface",
5224
5582
  baseUrl: "https://router.huggingface.co/v1",
@@ -5252,6 +5610,24 @@ export const MODELS = {
5252
5610
  contextWindow: 163840,
5253
5611
  maxTokens: 65536,
5254
5612
  },
5613
+ "deepseek-ai/DeepSeek-V4-Flash": {
5614
+ id: "deepseek-ai/DeepSeek-V4-Flash",
5615
+ name: "DeepSeek V4 Flash",
5616
+ api: "openai-completions",
5617
+ provider: "huggingface",
5618
+ baseUrl: "https://router.huggingface.co/v1",
5619
+ compat: { "supportsDeveloperRole": false },
5620
+ reasoning: true,
5621
+ input: ["text"],
5622
+ cost: {
5623
+ input: 0.14,
5624
+ output: 0.28,
5625
+ cacheRead: 0,
5626
+ cacheWrite: 0,
5627
+ },
5628
+ contextWindow: 1048576,
5629
+ maxTokens: 384000,
5630
+ },
5255
5631
  "deepseek-ai/DeepSeek-V4-Pro": {
5256
5632
  id: "deepseek-ai/DeepSeek-V4-Pro",
5257
5633
  name: "DeepSeek V4 Pro",
@@ -5270,6 +5646,60 @@ export const MODELS = {
5270
5646
  contextWindow: 1048576,
5271
5647
  maxTokens: 393216,
5272
5648
  },
5649
+ "google/gemma-4-26B-A4B-it": {
5650
+ id: "google/gemma-4-26B-A4B-it",
5651
+ name: "Gemma 4 26B A4B IT",
5652
+ api: "openai-completions",
5653
+ provider: "huggingface",
5654
+ baseUrl: "https://router.huggingface.co/v1",
5655
+ compat: { "supportsDeveloperRole": false },
5656
+ reasoning: true,
5657
+ input: ["text", "image"],
5658
+ cost: {
5659
+ input: 0.13,
5660
+ output: 0.4,
5661
+ cacheRead: 0,
5662
+ cacheWrite: 0,
5663
+ },
5664
+ contextWindow: 262144,
5665
+ maxTokens: 32768,
5666
+ },
5667
+ "google/gemma-4-31B-it": {
5668
+ id: "google/gemma-4-31B-it",
5669
+ name: "Gemma 4 31B IT",
5670
+ api: "openai-completions",
5671
+ provider: "huggingface",
5672
+ baseUrl: "https://router.huggingface.co/v1",
5673
+ compat: { "supportsDeveloperRole": false },
5674
+ reasoning: true,
5675
+ input: ["text", "image"],
5676
+ cost: {
5677
+ input: 0.14,
5678
+ output: 0.4,
5679
+ cacheRead: 0,
5680
+ cacheWrite: 0,
5681
+ },
5682
+ contextWindow: 262144,
5683
+ maxTokens: 32768,
5684
+ },
5685
+ "meta-llama/Llama-3.3-70B-Instruct": {
5686
+ id: "meta-llama/Llama-3.3-70B-Instruct",
5687
+ name: "Llama-3.3-70B-Instruct",
5688
+ api: "openai-completions",
5689
+ provider: "huggingface",
5690
+ baseUrl: "https://router.huggingface.co/v1",
5691
+ compat: { "supportsDeveloperRole": false },
5692
+ reasoning: false,
5693
+ input: ["text"],
5694
+ cost: {
5695
+ input: 0.59,
5696
+ output: 0.79,
5697
+ cacheRead: 0,
5698
+ cacheWrite: 0,
5699
+ },
5700
+ contextWindow: 131072,
5701
+ maxTokens: 4096,
5702
+ },
5273
5703
  "moonshotai/Kimi-K2-Instruct": {
5274
5704
  id: "moonshotai/Kimi-K2-Instruct",
5275
5705
  name: "Kimi-K2-Instruct",
@@ -5288,27 +5718,171 @@ export const MODELS = {
5288
5718
  contextWindow: 131072,
5289
5719
  maxTokens: 16384,
5290
5720
  },
5291
- "moonshotai/Kimi-K2-Instruct-0905": {
5292
- id: "moonshotai/Kimi-K2-Instruct-0905",
5293
- name: "Kimi-K2-Instruct-0905",
5721
+ "moonshotai/Kimi-K2-Instruct-0905": {
5722
+ id: "moonshotai/Kimi-K2-Instruct-0905",
5723
+ name: "Kimi-K2-Instruct-0905",
5724
+ api: "openai-completions",
5725
+ provider: "huggingface",
5726
+ baseUrl: "https://router.huggingface.co/v1",
5727
+ compat: { "supportsDeveloperRole": false },
5728
+ reasoning: false,
5729
+ input: ["text"],
5730
+ cost: {
5731
+ input: 1,
5732
+ output: 3,
5733
+ cacheRead: 0,
5734
+ cacheWrite: 0,
5735
+ },
5736
+ contextWindow: 262144,
5737
+ maxTokens: 16384,
5738
+ },
5739
+ "moonshotai/Kimi-K2-Thinking": {
5740
+ id: "moonshotai/Kimi-K2-Thinking",
5741
+ name: "Kimi-K2-Thinking",
5742
+ api: "openai-completions",
5743
+ provider: "huggingface",
5744
+ baseUrl: "https://router.huggingface.co/v1",
5745
+ compat: { "supportsDeveloperRole": false },
5746
+ reasoning: true,
5747
+ input: ["text"],
5748
+ cost: {
5749
+ input: 0.6,
5750
+ output: 2.5,
5751
+ cacheRead: 0.15,
5752
+ cacheWrite: 0,
5753
+ },
5754
+ contextWindow: 262144,
5755
+ maxTokens: 262144,
5756
+ },
5757
+ "moonshotai/Kimi-K2.5": {
5758
+ id: "moonshotai/Kimi-K2.5",
5759
+ name: "Kimi-K2.5",
5760
+ api: "openai-completions",
5761
+ provider: "huggingface",
5762
+ baseUrl: "https://router.huggingface.co/v1",
5763
+ compat: { "supportsDeveloperRole": false },
5764
+ reasoning: true,
5765
+ input: ["text", "image"],
5766
+ cost: {
5767
+ input: 0.6,
5768
+ output: 3,
5769
+ cacheRead: 0.1,
5770
+ cacheWrite: 0,
5771
+ },
5772
+ contextWindow: 262144,
5773
+ maxTokens: 262144,
5774
+ },
5775
+ "moonshotai/Kimi-K2.6": {
5776
+ id: "moonshotai/Kimi-K2.6",
5777
+ name: "Kimi-K2.6",
5778
+ api: "openai-completions",
5779
+ provider: "huggingface",
5780
+ baseUrl: "https://router.huggingface.co/v1",
5781
+ compat: { "supportsDeveloperRole": false },
5782
+ reasoning: true,
5783
+ input: ["text", "image"],
5784
+ cost: {
5785
+ input: 0.95,
5786
+ output: 4,
5787
+ cacheRead: 0.16,
5788
+ cacheWrite: 0,
5789
+ },
5790
+ contextWindow: 262144,
5791
+ maxTokens: 262144,
5792
+ },
5793
+ "moonshotai/Kimi-K2.7-Code": {
5794
+ id: "moonshotai/Kimi-K2.7-Code",
5795
+ name: "Kimi K2.7 Code",
5796
+ api: "openai-completions",
5797
+ provider: "huggingface",
5798
+ baseUrl: "https://router.huggingface.co/v1",
5799
+ compat: { "supportsDeveloperRole": false },
5800
+ reasoning: true,
5801
+ input: ["text", "image"],
5802
+ cost: {
5803
+ input: 0.95,
5804
+ output: 4,
5805
+ cacheRead: 0,
5806
+ cacheWrite: 0,
5807
+ },
5808
+ contextWindow: 262144,
5809
+ maxTokens: 262144,
5810
+ },
5811
+ "openai/gpt-oss-120b": {
5812
+ id: "openai/gpt-oss-120b",
5813
+ name: "GPT OSS 120B",
5814
+ api: "openai-completions",
5815
+ provider: "huggingface",
5816
+ baseUrl: "https://router.huggingface.co/v1",
5817
+ compat: { "supportsDeveloperRole": false },
5818
+ reasoning: true,
5819
+ input: ["text"],
5820
+ cost: {
5821
+ input: 0.25,
5822
+ output: 0.69,
5823
+ cacheRead: 0,
5824
+ cacheWrite: 0,
5825
+ },
5826
+ contextWindow: 131072,
5827
+ maxTokens: 32768,
5828
+ },
5829
+ "stepfun-ai/Step-3.5-Flash": {
5830
+ id: "stepfun-ai/Step-3.5-Flash",
5831
+ name: "Step 3.5 Flash",
5832
+ api: "openai-completions",
5833
+ provider: "huggingface",
5834
+ baseUrl: "https://router.huggingface.co/v1",
5835
+ compat: { "supportsDeveloperRole": false },
5836
+ reasoning: true,
5837
+ input: ["text"],
5838
+ cost: {
5839
+ input: 0.1,
5840
+ output: 0.3,
5841
+ cacheRead: 0,
5842
+ cacheWrite: 0,
5843
+ },
5844
+ contextWindow: 262144,
5845
+ maxTokens: 256000,
5846
+ },
5847
+ "stepfun-ai/Step-3.7-Flash": {
5848
+ id: "stepfun-ai/Step-3.7-Flash",
5849
+ name: "Step 3.7 Flash",
5850
+ api: "openai-completions",
5851
+ provider: "huggingface",
5852
+ baseUrl: "https://router.huggingface.co/v1",
5853
+ compat: { "supportsDeveloperRole": false },
5854
+ reasoning: true,
5855
+ input: ["text", "image"],
5856
+ cost: {
5857
+ input: 0.2,
5858
+ output: 1.15,
5859
+ cacheRead: 0,
5860
+ cacheWrite: 0,
5861
+ },
5862
+ contextWindow: 262144,
5863
+ maxTokens: 256000,
5864
+ },
5865
+ "zai-org/GLM-4.5": {
5866
+ id: "zai-org/GLM-4.5",
5867
+ name: "GLM-4.5",
5294
5868
  api: "openai-completions",
5295
5869
  provider: "huggingface",
5296
5870
  baseUrl: "https://router.huggingface.co/v1",
5297
5871
  compat: { "supportsDeveloperRole": false },
5298
- reasoning: false,
5872
+ reasoning: true,
5299
5873
  input: ["text"],
5300
5874
  cost: {
5301
- input: 1,
5302
- output: 3,
5875
+ input: 0.6,
5876
+ output: 2.2,
5303
5877
  cacheRead: 0,
5304
5878
  cacheWrite: 0,
5305
5879
  },
5306
- contextWindow: 262144,
5307
- maxTokens: 16384,
5880
+ contextWindow: 131072,
5881
+ maxTokens: 98304,
5308
5882
  },
5309
- "moonshotai/Kimi-K2-Thinking": {
5310
- id: "moonshotai/Kimi-K2-Thinking",
5311
- name: "Kimi-K2-Thinking",
5883
+ "zai-org/GLM-4.5-Air": {
5884
+ id: "zai-org/GLM-4.5-Air",
5885
+ name: "GLM-4.5-Air",
5312
5886
  api: "openai-completions",
5313
5887
  provider: "huggingface",
5314
5888
  baseUrl: "https://router.huggingface.co/v1",
@@ -5316,17 +5890,17 @@ export const MODELS = {
5316
5890
  reasoning: true,
5317
5891
  input: ["text"],
5318
5892
  cost: {
5319
- input: 0.6,
5320
- output: 2.5,
5321
- cacheRead: 0.15,
5893
+ input: 0.13,
5894
+ output: 0.85,
5895
+ cacheRead: 0,
5322
5896
  cacheWrite: 0,
5323
5897
  },
5324
- contextWindow: 262144,
5325
- maxTokens: 262144,
5898
+ contextWindow: 131072,
5899
+ maxTokens: 98304,
5326
5900
  },
5327
- "moonshotai/Kimi-K2.5": {
5328
- id: "moonshotai/Kimi-K2.5",
5329
- name: "Kimi-K2.5",
5901
+ "zai-org/GLM-4.5V": {
5902
+ id: "zai-org/GLM-4.5V",
5903
+ name: "GLM-4.5V",
5330
5904
  api: "openai-completions",
5331
5905
  provider: "huggingface",
5332
5906
  baseUrl: "https://router.huggingface.co/v1",
@@ -5335,30 +5909,30 @@ export const MODELS = {
5335
5909
  input: ["text", "image"],
5336
5910
  cost: {
5337
5911
  input: 0.6,
5338
- output: 3,
5339
- cacheRead: 0.1,
5912
+ output: 1.8,
5913
+ cacheRead: 0,
5340
5914
  cacheWrite: 0,
5341
5915
  },
5342
- contextWindow: 262144,
5343
- maxTokens: 262144,
5916
+ contextWindow: 65536,
5917
+ maxTokens: 16384,
5344
5918
  },
5345
- "moonshotai/Kimi-K2.6": {
5346
- id: "moonshotai/Kimi-K2.6",
5347
- name: "Kimi-K2.6",
5919
+ "zai-org/GLM-4.6": {
5920
+ id: "zai-org/GLM-4.6",
5921
+ name: "GLM-4.6",
5348
5922
  api: "openai-completions",
5349
5923
  provider: "huggingface",
5350
5924
  baseUrl: "https://router.huggingface.co/v1",
5351
5925
  compat: { "supportsDeveloperRole": false },
5352
5926
  reasoning: true,
5353
- input: ["text", "image"],
5927
+ input: ["text"],
5354
5928
  cost: {
5355
- input: 0.95,
5356
- output: 4,
5357
- cacheRead: 0.16,
5929
+ input: 0.55,
5930
+ output: 2.2,
5931
+ cacheRead: 0,
5358
5932
  cacheWrite: 0,
5359
5933
  },
5360
- contextWindow: 262144,
5361
- maxTokens: 262144,
5934
+ contextWindow: 204800,
5935
+ maxTokens: 131072,
5362
5936
  },
5363
5937
  "zai-org/GLM-4.7": {
5364
5938
  id: "zai-org/GLM-4.7",
@@ -5432,6 +6006,24 @@ export const MODELS = {
5432
6006
  contextWindow: 202752,
5433
6007
  maxTokens: 131072,
5434
6008
  },
6009
+ "zai-org/GLM-5.2": {
6010
+ id: "zai-org/GLM-5.2",
6011
+ name: "GLM-5.2",
6012
+ api: "openai-completions",
6013
+ provider: "huggingface",
6014
+ baseUrl: "https://router.huggingface.co/v1",
6015
+ compat: { "supportsDeveloperRole": false },
6016
+ reasoning: true,
6017
+ input: ["text"],
6018
+ cost: {
6019
+ input: 1.4,
6020
+ output: 4.4,
6021
+ cacheRead: 0,
6022
+ cacheWrite: 0,
6023
+ },
6024
+ contextWindow: 262144,
6025
+ maxTokens: 131072,
6026
+ },
5435
6027
  },
5436
6028
  "kimi-coding": {
5437
6029
  "k2p7": {
@@ -5533,12 +6125,12 @@ export const MODELS = {
5533
6125
  reasoning: true,
5534
6126
  input: ["text", "image"],
5535
6127
  cost: {
5536
- input: 0.6,
5537
- output: 2.4,
5538
- cacheRead: 0.12,
6128
+ input: 0.3,
6129
+ output: 1.2,
6130
+ cacheRead: 0.06,
5539
6131
  cacheWrite: 0,
5540
6132
  },
5541
- contextWindow: 512000,
6133
+ contextWindow: 1000000,
5542
6134
  maxTokens: 128000,
5543
6135
  },
5544
6136
  },
@@ -5586,12 +6178,12 @@ export const MODELS = {
5586
6178
  reasoning: true,
5587
6179
  input: ["text", "image"],
5588
6180
  cost: {
5589
- input: 0.6,
5590
- output: 2.4,
5591
- cacheRead: 0.12,
6181
+ input: 0.3,
6182
+ output: 1.2,
6183
+ cacheRead: 0.06,
5592
6184
  cacheWrite: 0,
5593
6185
  },
5594
- contextWindow: 512000,
6186
+ contextWindow: 1000000,
5595
6187
  maxTokens: 128000,
5596
6188
  },
5597
6189
  },
@@ -7606,6 +8198,23 @@ export const MODELS = {
7606
8198
  contextWindow: 204800,
7607
8199
  maxTokens: 131072,
7608
8200
  },
8201
+ "glm-5.2": {
8202
+ id: "glm-5.2",
8203
+ name: "GLM-5.2",
8204
+ api: "openai-completions",
8205
+ provider: "opencode",
8206
+ baseUrl: "https://opencode.ai/zen/v1",
8207
+ reasoning: true,
8208
+ input: ["text"],
8209
+ cost: {
8210
+ input: 1.4,
8211
+ output: 4.4,
8212
+ cacheRead: 0.26,
8213
+ cacheWrite: 0,
8214
+ },
8215
+ contextWindow: 1000000,
8216
+ maxTokens: 131072,
8217
+ },
7609
8218
  "gpt-5": {
7610
8219
  id: "gpt-5",
7611
8220
  name: "GPT-5",
@@ -8232,12 +8841,12 @@ export const MODELS = {
8232
8841
  reasoning: true,
8233
8842
  input: ["text", "image"],
8234
8843
  cost: {
8235
- input: 0.1,
8236
- output: 0.4,
8237
- cacheRead: 0.02,
8844
+ input: 0.3,
8845
+ output: 1.2,
8846
+ cacheRead: 0.06,
8238
8847
  cacheWrite: 0,
8239
8848
  },
8240
- contextWindow: 512000,
8849
+ contextWindow: 1000000,
8241
8850
  maxTokens: 131072,
8242
8851
  },
8243
8852
  "qwen3.6-plus": {
@@ -8517,24 +9126,6 @@ export const MODELS = {
8517
9126
  contextWindow: 1000000,
8518
9127
  maxTokens: 128000,
8519
9128
  },
8520
- "anthropic/claude-opus-4.6-fast": {
8521
- id: "anthropic/claude-opus-4.6-fast",
8522
- name: "Anthropic: Claude Opus 4.6 (Fast)",
8523
- api: "openai-completions",
8524
- provider: "openrouter",
8525
- baseUrl: "https://openrouter.ai/api/v1",
8526
- reasoning: true,
8527
- thinkingLevelMap: { "xhigh": "max" },
8528
- input: ["text", "image"],
8529
- cost: {
8530
- input: 30,
8531
- output: 150,
8532
- cacheRead: 3,
8533
- cacheWrite: 37.5,
8534
- },
8535
- contextWindow: 1000000,
8536
- maxTokens: 128000,
8537
- },
8538
9129
  "anthropic/claude-opus-4.7": {
8539
9130
  id: "anthropic/claude-opus-4.7",
8540
9131
  name: "Anthropic: Claude Opus 4.7",
@@ -8658,6 +9249,23 @@ export const MODELS = {
8658
9249
  contextWindow: 1000000,
8659
9250
  maxTokens: 128000,
8660
9251
  },
9252
+ "anthropic/claude-sonnet-5": {
9253
+ id: "anthropic/claude-sonnet-5",
9254
+ name: "Anthropic: Claude Sonnet 5",
9255
+ api: "openai-completions",
9256
+ provider: "openrouter",
9257
+ baseUrl: "https://openrouter.ai/api/v1",
9258
+ reasoning: true,
9259
+ input: ["text", "image"],
9260
+ cost: {
9261
+ input: 2,
9262
+ output: 10,
9263
+ cacheRead: 0.19999999999999998,
9264
+ cacheWrite: 2.5,
9265
+ },
9266
+ contextWindow: 1000000,
9267
+ maxTokens: 128000,
9268
+ },
8661
9269
  "arcee-ai/trinity-large-thinking": {
8662
9270
  id: "arcee-ai/trinity-large-thinking",
8663
9271
  name: "Arcee AI: Trinity Large Thinking",
@@ -8667,13 +9275,13 @@ export const MODELS = {
8667
9275
  reasoning: true,
8668
9276
  input: ["text"],
8669
9277
  cost: {
8670
- input: 0.22,
8671
- output: 0.85,
9278
+ input: 0.25,
9279
+ output: 0.7999999999999999,
8672
9280
  cacheRead: 0.06,
8673
9281
  cacheWrite: 0,
8674
9282
  },
8675
9283
  contextWindow: 262144,
8676
- maxTokens: 262144,
9284
+ maxTokens: 80000,
8677
9285
  },
8678
9286
  "arcee-ai/trinity-mini": {
8679
9287
  id: "arcee-ai/trinity-mini",
@@ -8958,7 +9566,7 @@ export const MODELS = {
8958
9566
  cost: {
8959
9567
  input: 0.2288,
8960
9568
  output: 0.3432,
8961
- cacheRead: 0,
9569
+ cacheRead: 0.02288,
8962
9570
  cacheWrite: 0,
8963
9571
  },
8964
9572
  contextWindow: 131072,
@@ -8992,13 +9600,13 @@ export const MODELS = {
8992
9600
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8993
9601
  input: ["text"],
8994
9602
  cost: {
8995
- input: 0.09,
8996
- output: 0.18,
9603
+ input: 0.098,
9604
+ output: 0.196,
8997
9605
  cacheRead: 0.02,
8998
9606
  cacheWrite: 0,
8999
9607
  },
9000
9608
  contextWindow: 1048576,
9001
- maxTokens: 65536,
9609
+ maxTokens: 4096,
9002
9610
  },
9003
9611
  "deepseek/deepseek-v4-pro": {
9004
9612
  id: "deepseek/deepseek-v4-pro",
@@ -9019,23 +9627,6 @@ export const MODELS = {
9019
9627
  contextWindow: 1048576,
9020
9628
  maxTokens: 384000,
9021
9629
  },
9022
- "essentialai/rnj-1-instruct": {
9023
- id: "essentialai/rnj-1-instruct",
9024
- name: "EssentialAI: Rnj 1 Instruct",
9025
- api: "openai-completions",
9026
- provider: "openrouter",
9027
- baseUrl: "https://openrouter.ai/api/v1",
9028
- reasoning: false,
9029
- input: ["text"],
9030
- cost: {
9031
- input: 0.15,
9032
- output: 0.15,
9033
- cacheRead: 0,
9034
- cacheWrite: 0,
9035
- },
9036
- contextWindow: 32768,
9037
- maxTokens: 4096,
9038
- },
9039
9630
  "google/gemini-2.5-flash": {
9040
9631
  id: "google/gemini-2.5-flash",
9041
9632
  name: "Google: Gemini 2.5 Flash",
@@ -9640,9 +10231,9 @@ export const MODELS = {
9640
10231
  reasoning: true,
9641
10232
  input: ["text"],
9642
10233
  cost: {
9643
- input: 0.15,
9644
- output: 0.8999999999999999,
9645
- cacheRead: 0.049999999999999996,
10234
+ input: 0.12,
10235
+ output: 0.48,
10236
+ cacheRead: 0,
9646
10237
  cacheWrite: 0,
9647
10238
  },
9648
10239
  contextWindow: 204800,
@@ -9657,13 +10248,13 @@ export const MODELS = {
9657
10248
  reasoning: true,
9658
10249
  input: ["text"],
9659
10250
  cost: {
9660
- input: 0.25,
9661
- output: 1,
9662
- cacheRead: 0.049999999999999996,
10251
+ input: 0.18,
10252
+ output: 0.72,
10253
+ cacheRead: 0,
9663
10254
  cacheWrite: 0,
9664
10255
  },
9665
10256
  contextWindow: 204800,
9666
- maxTokens: 131072,
10257
+ maxTokens: 196608,
9667
10258
  },
9668
10259
  "minimax/minimax-m3": {
9669
10260
  id: "minimax/minimax-m3",
@@ -9986,7 +10577,7 @@ export const MODELS = {
9986
10577
  cacheWrite: 0,
9987
10578
  },
9988
10579
  contextWindow: 131072,
9989
- maxTokens: 32768,
10580
+ maxTokens: 100352,
9990
10581
  },
9991
10582
  "moonshotai/kimi-k2-0905": {
9992
10583
  id: "moonshotai/kimi-k2-0905",
@@ -10003,7 +10594,7 @@ export const MODELS = {
10003
10594
  cacheWrite: 0,
10004
10595
  },
10005
10596
  contextWindow: 262144,
10006
- maxTokens: 262144,
10597
+ maxTokens: 100352,
10007
10598
  },
10008
10599
  "moonshotai/kimi-k2-thinking": {
10009
10600
  id: "moonshotai/kimi-k2-thinking",
@@ -10016,7 +10607,7 @@ export const MODELS = {
10016
10607
  cost: {
10017
10608
  input: 0.6,
10018
10609
  output: 2.5,
10019
- cacheRead: 0,
10610
+ cacheRead: 0.6,
10020
10611
  cacheWrite: 0,
10021
10612
  },
10022
10613
  contextWindow: 262144,
@@ -10048,9 +10639,9 @@ export const MODELS = {
10048
10639
  reasoning: true,
10049
10640
  input: ["text", "image"],
10050
10641
  cost: {
10051
- input: 0.67,
10052
- output: 3.5,
10053
- cacheRead: 0.19999999999999998,
10642
+ input: 0.55,
10643
+ output: 3.1999999999999997,
10644
+ cacheRead: 0.11,
10054
10645
  cacheWrite: 0,
10055
10646
  },
10056
10647
  contextWindow: 262144,
@@ -10073,23 +10664,6 @@ export const MODELS = {
10073
10664
  contextWindow: 262144,
10074
10665
  maxTokens: 16384,
10075
10666
  },
10076
- "nex-agi/nex-n2-pro:free": {
10077
- id: "nex-agi/nex-n2-pro:free",
10078
- name: "Nex AGI: Nex-N2-Pro (free)",
10079
- api: "openai-completions",
10080
- provider: "openrouter",
10081
- baseUrl: "https://openrouter.ai/api/v1",
10082
- reasoning: true,
10083
- input: ["text", "image"],
10084
- cost: {
10085
- input: 0,
10086
- output: 0,
10087
- cacheRead: 0,
10088
- cacheWrite: 0,
10089
- },
10090
- contextWindow: 262144,
10091
- maxTokens: 262144,
10092
- },
10093
10667
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
10094
10668
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
10095
10669
  name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
@@ -10167,13 +10741,13 @@ export const MODELS = {
10167
10741
  reasoning: true,
10168
10742
  input: ["text"],
10169
10743
  cost: {
10170
- input: 0.09,
10171
- output: 0.44999999999999996,
10744
+ input: 0.08499999999999999,
10745
+ output: 0.39999999999999997,
10172
10746
  cacheRead: 0,
10173
10747
  cacheWrite: 0,
10174
10748
  },
10175
10749
  contextWindow: 1000000,
10176
- maxTokens: 4096,
10750
+ maxTokens: 16384,
10177
10751
  },
10178
10752
  "nvidia/nemotron-3-super-120b-a12b:free": {
10179
10753
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -10859,13 +11433,13 @@ export const MODELS = {
10859
11433
  reasoning: true,
10860
11434
  input: ["text"],
10861
11435
  cost: {
10862
- input: 0.039,
10863
- output: 0.18,
11436
+ input: 0.03,
11437
+ output: 0.15,
10864
11438
  cacheRead: 0,
10865
11439
  cacheWrite: 0,
10866
11440
  },
10867
11441
  contextWindow: 131072,
10868
- maxTokens: 4096,
11442
+ maxTokens: 131072,
10869
11443
  },
10870
11444
  "openai/gpt-oss-120b:free": {
10871
11445
  id: "openai/gpt-oss-120b:free",
@@ -11122,23 +11696,6 @@ export const MODELS = {
11122
11696
  contextWindow: 200000,
11123
11697
  maxTokens: 4096,
11124
11698
  },
11125
- "openrouter/owl-alpha": {
11126
- id: "openrouter/owl-alpha",
11127
- name: "Owl Alpha",
11128
- api: "openai-completions",
11129
- provider: "openrouter",
11130
- baseUrl: "https://openrouter.ai/api/v1",
11131
- reasoning: false,
11132
- input: ["text"],
11133
- cost: {
11134
- input: 0,
11135
- output: 0,
11136
- cacheRead: 0,
11137
- cacheWrite: 0,
11138
- },
11139
- contextWindow: 1048756,
11140
- maxTokens: 262144,
11141
- },
11142
11699
  "poolside/laguna-m.1": {
11143
11700
  id: "poolside/laguna-m.1",
11144
11701
  name: "Poolside: Laguna M.1",
@@ -11207,23 +11764,6 @@ export const MODELS = {
11207
11764
  contextWindow: 262144,
11208
11765
  maxTokens: 32768,
11209
11766
  },
11210
- "prime-intellect/intellect-3": {
11211
- id: "prime-intellect/intellect-3",
11212
- name: "Prime Intellect: INTELLECT-3",
11213
- api: "openai-completions",
11214
- provider: "openrouter",
11215
- baseUrl: "https://openrouter.ai/api/v1",
11216
- reasoning: true,
11217
- input: ["text"],
11218
- cost: {
11219
- input: 0.19999999999999998,
11220
- output: 1.1,
11221
- cacheRead: 0,
11222
- cacheWrite: 0,
11223
- },
11224
- contextWindow: 131072,
11225
- maxTokens: 131072,
11226
- },
11227
11767
  "qwen/qwen-2.5-72b-instruct": {
11228
11768
  id: "qwen/qwen-2.5-72b-instruct",
11229
11769
  name: "Qwen2.5 72B Instruct",
@@ -11369,13 +11909,13 @@ export const MODELS = {
11369
11909
  reasoning: true,
11370
11910
  input: ["text"],
11371
11911
  cost: {
11372
- input: 0.09999999999999999,
11373
- output: 0.09999999999999999,
11374
- cacheRead: 0.09999999999999999,
11912
+ input: 0.14950000000000002,
11913
+ output: 1.495,
11914
+ cacheRead: 0,
11375
11915
  cacheWrite: 0,
11376
11916
  },
11377
11917
  contextWindow: 262144,
11378
- maxTokens: 262144,
11918
+ maxTokens: 4096,
11379
11919
  },
11380
11920
  "qwen/qwen3-30b-a3b": {
11381
11921
  id: "qwen/qwen3-30b-a3b",
@@ -11813,11 +12353,11 @@ export const MODELS = {
11813
12353
  cost: {
11814
12354
  input: 0.14,
11815
12355
  output: 1,
11816
- cacheRead: 0,
12356
+ cacheRead: 0.049999999999999996,
11817
12357
  cacheWrite: 0,
11818
12358
  },
11819
12359
  contextWindow: 262144,
11820
- maxTokens: 262144,
12360
+ maxTokens: 81920,
11821
12361
  },
11822
12362
  "qwen/qwen3.5-397b-a17b": {
11823
12363
  id: "qwen/qwen3.5-397b-a17b",
@@ -11913,8 +12453,8 @@ export const MODELS = {
11913
12453
  reasoning: true,
11914
12454
  input: ["text", "image"],
11915
12455
  cost: {
11916
- input: 0.28850000000000003,
11917
- output: 3.17,
12456
+ input: 0.28500000000000003,
12457
+ output: 2.4,
11918
12458
  cacheRead: 0,
11919
12459
  cacheWrite: 0,
11920
12460
  },
@@ -12057,6 +12597,23 @@ export const MODELS = {
12057
12597
  contextWindow: 256000,
12058
12598
  maxTokens: 128000,
12059
12599
  },
12600
+ "sakana/fugu-ultra": {
12601
+ id: "sakana/fugu-ultra",
12602
+ name: "Sakana: Fugu Ultra",
12603
+ api: "openai-completions",
12604
+ provider: "openrouter",
12605
+ baseUrl: "https://openrouter.ai/api/v1",
12606
+ reasoning: true,
12607
+ input: ["text", "image"],
12608
+ cost: {
12609
+ input: 5,
12610
+ output: 30,
12611
+ cacheRead: 0.5,
12612
+ cacheWrite: 0,
12613
+ },
12614
+ contextWindow: 1000000,
12615
+ maxTokens: 128000,
12616
+ },
12060
12617
  "sao10k/l3.1-euryale-70b": {
12061
12618
  id: "sao10k/l3.1-euryale-70b",
12062
12619
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -12083,13 +12640,13 @@ export const MODELS = {
12083
12640
  reasoning: true,
12084
12641
  input: ["text"],
12085
12642
  cost: {
12086
- input: 0.09,
12643
+ input: 0.09999999999999999,
12087
12644
  output: 0.3,
12088
- cacheRead: 0.02,
12645
+ cacheRead: 0,
12089
12646
  cacheWrite: 0,
12090
12647
  },
12091
12648
  contextWindow: 262144,
12092
- maxTokens: 16384,
12649
+ maxTokens: 65536,
12093
12650
  },
12094
12651
  "stepfun/step-3.7-flash": {
12095
12652
  id: "stepfun/step-3.7-flash",
@@ -12108,23 +12665,6 @@ export const MODELS = {
12108
12665
  contextWindow: 256000,
12109
12666
  maxTokens: 256000,
12110
12667
  },
12111
- "thedrummer/rocinante-12b": {
12112
- id: "thedrummer/rocinante-12b",
12113
- name: "TheDrummer: Rocinante 12B",
12114
- api: "openai-completions",
12115
- provider: "openrouter",
12116
- baseUrl: "https://openrouter.ai/api/v1",
12117
- reasoning: false,
12118
- input: ["text"],
12119
- cost: {
12120
- input: 0.16999999999999998,
12121
- output: 0.43,
12122
- cacheRead: 0,
12123
- cacheWrite: 0,
12124
- },
12125
- contextWindow: 32768,
12126
- maxTokens: 32768,
12127
- },
12128
12668
  "thedrummer/unslopnemo-12b": {
12129
12669
  id: "thedrummer/unslopnemo-12b",
12130
12670
  name: "TheDrummer: UnslopNemo 12B",
@@ -12219,13 +12759,13 @@ export const MODELS = {
12219
12759
  reasoning: true,
12220
12760
  input: ["text", "image"],
12221
12761
  cost: {
12222
- input: 0.14,
12762
+ input: 0.105,
12223
12763
  output: 0.28,
12224
- cacheRead: 0.0028,
12764
+ cacheRead: 0,
12225
12765
  cacheWrite: 0,
12226
12766
  },
12227
12767
  contextWindow: 1048576,
12228
- maxTokens: 131072,
12768
+ maxTokens: 4096,
12229
12769
  },
12230
12770
  "xiaomi/mimo-v2.5-pro": {
12231
12771
  id: "xiaomi/mimo-v2.5-pro",
@@ -12406,9 +12946,9 @@ export const MODELS = {
12406
12946
  reasoning: true,
12407
12947
  input: ["text"],
12408
12948
  cost: {
12409
- input: 0.98,
12410
- output: 3.08,
12411
- cacheRead: 0.182,
12949
+ input: 0.975,
12950
+ output: 4.300000000000001,
12951
+ cacheRead: 0,
12412
12952
  cacheWrite: 0,
12413
12953
  },
12414
12954
  contextWindow: 202752,
@@ -12423,12 +12963,29 @@ export const MODELS = {
12423
12963
  reasoning: true,
12424
12964
  input: ["text"],
12425
12965
  cost: {
12426
- input: 1.2,
12427
- output: 4.1,
12428
- cacheRead: 0.19999999999999998,
12966
+ input: 0.9299999999999999,
12967
+ output: 3,
12968
+ cacheRead: 0.18,
12429
12969
  cacheWrite: 0,
12430
12970
  },
12431
12971
  contextWindow: 1048576,
12972
+ maxTokens: 32768,
12973
+ },
12974
+ "z-ai/glm-5v-turbo": {
12975
+ id: "z-ai/glm-5v-turbo",
12976
+ name: "Z.ai: GLM 5V Turbo",
12977
+ api: "openai-completions",
12978
+ provider: "openrouter",
12979
+ baseUrl: "https://openrouter.ai/api/v1",
12980
+ reasoning: true,
12981
+ input: ["text", "image"],
12982
+ cost: {
12983
+ input: 1.2,
12984
+ output: 4,
12985
+ cacheRead: 0.24,
12986
+ cacheWrite: 0,
12987
+ },
12988
+ contextWindow: 202752,
12432
12989
  maxTokens: 131072,
12433
12990
  },
12434
12991
  "~anthropic/claude-fable-latest": {
@@ -12492,10 +13049,10 @@ export const MODELS = {
12492
13049
  reasoning: true,
12493
13050
  input: ["text", "image"],
12494
13051
  cost: {
12495
- input: 3,
12496
- output: 15,
12497
- cacheRead: 0.3,
12498
- cacheWrite: 3.75,
13052
+ input: 2,
13053
+ output: 10,
13054
+ cacheRead: 0.19999999999999998,
13055
+ cacheWrite: 2.5,
12499
13056
  },
12500
13057
  contextWindow: 1000000,
12501
13058
  maxTokens: 128000,
@@ -12543,9 +13100,9 @@ export const MODELS = {
12543
13100
  reasoning: true,
12544
13101
  input: ["text", "image"],
12545
13102
  cost: {
12546
- input: 0.67,
12547
- output: 3.5,
12548
- cacheRead: 0.19999999999999998,
13103
+ input: 0.55,
13104
+ output: 3.1999999999999997,
13105
+ cacheRead: 0.11,
12549
13106
  cacheWrite: 0,
12550
13107
  },
12551
13108
  contextWindow: 262144,
@@ -12943,6 +13500,25 @@ export const MODELS = {
12943
13500
  contextWindow: 202752,
12944
13501
  maxTokens: 131072,
12945
13502
  },
13503
+ "zai-org/GLM-5.2": {
13504
+ id: "zai-org/GLM-5.2",
13505
+ name: "GLM-5.2",
13506
+ api: "openai-completions",
13507
+ provider: "together",
13508
+ baseUrl: "https://api.together.ai/v1",
13509
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13510
+ reasoning: true,
13511
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13512
+ input: ["text"],
13513
+ cost: {
13514
+ input: 1.4,
13515
+ output: 4.4,
13516
+ cacheRead: 0.26,
13517
+ cacheWrite: 0,
13518
+ },
13519
+ contextWindow: 262144,
13520
+ maxTokens: 164000,
13521
+ },
12946
13522
  },
12947
13523
  "vercel-ai-gateway": {
12948
13524
  "alibaba/qwen-3-14b": {
@@ -13053,7 +13629,7 @@ export const MODELS = {
13053
13629
  api: "anthropic-messages",
13054
13630
  provider: "vercel-ai-gateway",
13055
13631
  baseUrl: "https://ai-gateway.vercel.sh",
13056
- reasoning: true,
13632
+ reasoning: false,
13057
13633
  input: ["text"],
13058
13634
  cost: {
13059
13635
  input: 1.5,
@@ -13070,7 +13646,7 @@ export const MODELS = {
13070
13646
  api: "anthropic-messages",
13071
13647
  provider: "vercel-ai-gateway",
13072
13648
  baseUrl: "https://ai-gateway.vercel.sh",
13073
- reasoning: true,
13649
+ reasoning: false,
13074
13650
  input: ["text"],
13075
13651
  cost: {
13076
13652
  input: 0.15,
@@ -13087,7 +13663,7 @@ export const MODELS = {
13087
13663
  api: "anthropic-messages",
13088
13664
  provider: "vercel-ai-gateway",
13089
13665
  baseUrl: "https://ai-gateway.vercel.sh",
13090
- reasoning: true,
13666
+ reasoning: false,
13091
13667
  input: ["text"],
13092
13668
  cost: {
13093
13669
  input: 0.5,
@@ -13200,6 +13776,40 @@ export const MODELS = {
13200
13776
  contextWindow: 131072,
13201
13777
  maxTokens: 32768,
13202
13778
  },
13779
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
13780
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
13781
+ name: "Qwen3 VL 235B A22B Instruct",
13782
+ api: "anthropic-messages",
13783
+ provider: "vercel-ai-gateway",
13784
+ baseUrl: "https://ai-gateway.vercel.sh",
13785
+ reasoning: false,
13786
+ input: ["text", "image"],
13787
+ cost: {
13788
+ input: 0.39999999999999997,
13789
+ output: 1.5999999999999999,
13790
+ cacheRead: 0,
13791
+ cacheWrite: 0,
13792
+ },
13793
+ contextWindow: 131072,
13794
+ maxTokens: 129024,
13795
+ },
13796
+ "alibaba/qwen3-vl-instruct": {
13797
+ id: "alibaba/qwen3-vl-instruct",
13798
+ name: "Qwen3 VL 235B A22B Instruct",
13799
+ api: "anthropic-messages",
13800
+ provider: "vercel-ai-gateway",
13801
+ baseUrl: "https://ai-gateway.vercel.sh",
13802
+ reasoning: false,
13803
+ input: ["text", "image"],
13804
+ cost: {
13805
+ input: 0.39999999999999997,
13806
+ output: 1.5999999999999999,
13807
+ cacheRead: 0,
13808
+ cacheWrite: 0,
13809
+ },
13810
+ contextWindow: 131072,
13811
+ maxTokens: 129024,
13812
+ },
13203
13813
  "alibaba/qwen3-vl-thinking": {
13204
13814
  id: "alibaba/qwen3-vl-thinking",
13205
13815
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13319,6 +13929,74 @@ export const MODELS = {
13319
13929
  contextWindow: 1000000,
13320
13930
  maxTokens: 64000,
13321
13931
  },
13932
+ "amazon/nova-2-lite": {
13933
+ id: "amazon/nova-2-lite",
13934
+ name: "Nova 2 Lite",
13935
+ api: "anthropic-messages",
13936
+ provider: "vercel-ai-gateway",
13937
+ baseUrl: "https://ai-gateway.vercel.sh",
13938
+ reasoning: true,
13939
+ input: ["text", "image"],
13940
+ cost: {
13941
+ input: 0.3,
13942
+ output: 2.5,
13943
+ cacheRead: 0.075,
13944
+ cacheWrite: 0,
13945
+ },
13946
+ contextWindow: 1000000,
13947
+ maxTokens: 1000000,
13948
+ },
13949
+ "amazon/nova-lite": {
13950
+ id: "amazon/nova-lite",
13951
+ name: "Nova Lite",
13952
+ api: "anthropic-messages",
13953
+ provider: "vercel-ai-gateway",
13954
+ baseUrl: "https://ai-gateway.vercel.sh",
13955
+ reasoning: false,
13956
+ input: ["text", "image"],
13957
+ cost: {
13958
+ input: 0.06,
13959
+ output: 0.24,
13960
+ cacheRead: 0,
13961
+ cacheWrite: 0,
13962
+ },
13963
+ contextWindow: 300000,
13964
+ maxTokens: 8192,
13965
+ },
13966
+ "amazon/nova-micro": {
13967
+ id: "amazon/nova-micro",
13968
+ name: "Nova Micro",
13969
+ api: "anthropic-messages",
13970
+ provider: "vercel-ai-gateway",
13971
+ baseUrl: "https://ai-gateway.vercel.sh",
13972
+ reasoning: false,
13973
+ input: ["text"],
13974
+ cost: {
13975
+ input: 0.035,
13976
+ output: 0.14,
13977
+ cacheRead: 0,
13978
+ cacheWrite: 0,
13979
+ },
13980
+ contextWindow: 128000,
13981
+ maxTokens: 8192,
13982
+ },
13983
+ "amazon/nova-pro": {
13984
+ id: "amazon/nova-pro",
13985
+ name: "Nova Pro",
13986
+ api: "anthropic-messages",
13987
+ provider: "vercel-ai-gateway",
13988
+ baseUrl: "https://ai-gateway.vercel.sh",
13989
+ reasoning: false,
13990
+ input: ["text", "image"],
13991
+ cost: {
13992
+ input: 0.7999999999999999,
13993
+ output: 3.1999999999999997,
13994
+ cacheRead: 0,
13995
+ cacheWrite: 0,
13996
+ },
13997
+ contextWindow: 300000,
13998
+ maxTokens: 8192,
13999
+ },
13322
14000
  "anthropic/claude-3-haiku": {
13323
14001
  id: "anthropic/claude-3-haiku",
13324
14002
  name: "Claude 3 Haiku",
@@ -13509,6 +14187,23 @@ export const MODELS = {
13509
14187
  contextWindow: 1000000,
13510
14188
  maxTokens: 128000,
13511
14189
  },
14190
+ "anthropic/claude-sonnet-5": {
14191
+ id: "anthropic/claude-sonnet-5",
14192
+ name: "Claude Sonnet 5",
14193
+ api: "anthropic-messages",
14194
+ provider: "vercel-ai-gateway",
14195
+ baseUrl: "https://ai-gateway.vercel.sh",
14196
+ reasoning: true,
14197
+ input: ["text", "image"],
14198
+ cost: {
14199
+ input: 2,
14200
+ output: 10,
14201
+ cacheRead: 0.19999999999999998,
14202
+ cacheWrite: 2.5,
14203
+ },
14204
+ contextWindow: 1000000,
14205
+ maxTokens: 128000,
14206
+ },
13512
14207
  "arcee-ai/trinity-large-preview": {
13513
14208
  id: "arcee-ai/trinity-large-preview",
13514
14209
  name: "Trinity Large Preview",
@@ -13543,6 +14238,23 @@ export const MODELS = {
13543
14238
  contextWindow: 262100,
13544
14239
  maxTokens: 80000,
13545
14240
  },
14241
+ "arcee-ai/trinity-mini": {
14242
+ id: "arcee-ai/trinity-mini",
14243
+ name: "Trinity Mini",
14244
+ api: "anthropic-messages",
14245
+ provider: "vercel-ai-gateway",
14246
+ baseUrl: "https://ai-gateway.vercel.sh",
14247
+ reasoning: false,
14248
+ input: ["text"],
14249
+ cost: {
14250
+ input: 0.045,
14251
+ output: 0.15,
14252
+ cacheRead: 0,
14253
+ cacheWrite: 0,
14254
+ },
14255
+ contextWindow: 131072,
14256
+ maxTokens: 131072,
14257
+ },
13546
14258
  "bytedance/seed-1.6": {
13547
14259
  id: "bytedance/seed-1.6",
13548
14260
  name: "Seed 1.6",
@@ -13550,7 +14262,7 @@ export const MODELS = {
13550
14262
  provider: "vercel-ai-gateway",
13551
14263
  baseUrl: "https://ai-gateway.vercel.sh",
13552
14264
  reasoning: true,
13553
- input: ["text"],
14265
+ input: ["text", "image"],
13554
14266
  cost: {
13555
14267
  input: 0.25,
13556
14268
  output: 2,
@@ -13560,6 +14272,23 @@ export const MODELS = {
13560
14272
  contextWindow: 256000,
13561
14273
  maxTokens: 32000,
13562
14274
  },
14275
+ "bytedance/seed-1.8": {
14276
+ id: "bytedance/seed-1.8",
14277
+ name: "Bytedance Seed 1.8",
14278
+ api: "anthropic-messages",
14279
+ provider: "vercel-ai-gateway",
14280
+ baseUrl: "https://ai-gateway.vercel.sh",
14281
+ reasoning: true,
14282
+ input: ["text", "image"],
14283
+ cost: {
14284
+ input: 0.25,
14285
+ output: 2,
14286
+ cacheRead: 0.049999999999999996,
14287
+ cacheWrite: 0,
14288
+ },
14289
+ contextWindow: 256000,
14290
+ maxTokens: 64000,
14291
+ },
13563
14292
  "cohere/command-a": {
13564
14293
  id: "cohere/command-a",
13565
14294
  name: "Command A",
@@ -13620,13 +14349,13 @@ export const MODELS = {
13620
14349
  reasoning: true,
13621
14350
  input: ["text"],
13622
14351
  cost: {
13623
- input: 0.56,
13624
- output: 1.68,
13625
- cacheRead: 0.28,
14352
+ input: 0.6,
14353
+ output: 1.7,
14354
+ cacheRead: 0,
13626
14355
  cacheWrite: 0,
13627
14356
  },
13628
- contextWindow: 163840,
13629
- maxTokens: 8192,
14357
+ contextWindow: 128000,
14358
+ maxTokens: 128000,
13630
14359
  },
13631
14360
  "deepseek/deepseek-v3.1-terminus": {
13632
14361
  id: "deepseek/deepseek-v3.1-terminus",
@@ -13651,8 +14380,8 @@ export const MODELS = {
13651
14380
  api: "anthropic-messages",
13652
14381
  provider: "vercel-ai-gateway",
13653
14382
  baseUrl: "https://ai-gateway.vercel.sh",
13654
- reasoning: true,
13655
- input: ["text", "image"],
14383
+ reasoning: false,
14384
+ input: ["text"],
13656
14385
  cost: {
13657
14386
  input: 0.28,
13658
14387
  output: 0.42,
@@ -13669,7 +14398,7 @@ export const MODELS = {
13669
14398
  provider: "vercel-ai-gateway",
13670
14399
  baseUrl: "https://ai-gateway.vercel.sh",
13671
14400
  reasoning: true,
13672
- input: ["text", "image"],
14401
+ input: ["text"],
13673
14402
  cost: {
13674
14403
  input: 0.62,
13675
14404
  output: 1.85,
@@ -13686,7 +14415,7 @@ export const MODELS = {
13686
14415
  provider: "vercel-ai-gateway",
13687
14416
  baseUrl: "https://ai-gateway.vercel.sh",
13688
14417
  reasoning: true,
13689
- input: ["text", "image"],
14418
+ input: ["text"],
13690
14419
  cost: {
13691
14420
  input: 0.14,
13692
14421
  output: 0.28,
@@ -13889,7 +14618,7 @@ export const MODELS = {
13889
14618
  api: "anthropic-messages",
13890
14619
  provider: "vercel-ai-gateway",
13891
14620
  baseUrl: "https://ai-gateway.vercel.sh",
13892
- reasoning: false,
14621
+ reasoning: true,
13893
14622
  input: ["text", "image"],
13894
14623
  cost: {
13895
14624
  input: 0.14,
@@ -13914,25 +14643,59 @@ export const MODELS = {
13914
14643
  cacheRead: 0.024999999999999998,
13915
14644
  cacheWrite: 0,
13916
14645
  },
13917
- contextWindow: 128000,
13918
- maxTokens: 128000,
14646
+ contextWindow: 128000,
14647
+ maxTokens: 128000,
14648
+ },
14649
+ "inception/mercury-coder-small": {
14650
+ id: "inception/mercury-coder-small",
14651
+ name: "Mercury Coder Small Beta",
14652
+ api: "anthropic-messages",
14653
+ provider: "vercel-ai-gateway",
14654
+ baseUrl: "https://ai-gateway.vercel.sh",
14655
+ reasoning: false,
14656
+ input: ["text"],
14657
+ cost: {
14658
+ input: 0.25,
14659
+ output: 1,
14660
+ cacheRead: 0,
14661
+ cacheWrite: 0,
14662
+ },
14663
+ contextWindow: 32000,
14664
+ maxTokens: 16384,
14665
+ },
14666
+ "interfaze/interfaze-beta": {
14667
+ id: "interfaze/interfaze-beta",
14668
+ name: "Interfaze Beta",
14669
+ api: "anthropic-messages",
14670
+ provider: "vercel-ai-gateway",
14671
+ baseUrl: "https://ai-gateway.vercel.sh",
14672
+ reasoning: true,
14673
+ input: ["text", "image"],
14674
+ cost: {
14675
+ input: 1.5,
14676
+ output: 3.5,
14677
+ cacheRead: 0,
14678
+ cacheWrite: 0,
14679
+ },
14680
+ contextWindow: 1000000,
14681
+ maxTokens: 32000,
13919
14682
  },
13920
- "inception/mercury-coder-small": {
13921
- id: "inception/mercury-coder-small",
13922
- name: "Mercury Coder Small Beta",
14683
+ "kwaipilot/kat-coder-pro-v1": {
14684
+ id: "kwaipilot/kat-coder-pro-v1",
14685
+ name: "KAT-Coder-Pro V1",
13923
14686
  api: "anthropic-messages",
13924
14687
  provider: "vercel-ai-gateway",
13925
14688
  baseUrl: "https://ai-gateway.vercel.sh",
13926
14689
  reasoning: false,
13927
14690
  input: ["text"],
13928
14691
  cost: {
13929
- input: 0.25,
13930
- output: 1,
13931
- cacheRead: 0,
14692
+ input: 0.3,
14693
+ output: 1.2,
14694
+ cacheRead: 0.06,
13932
14695
  cacheWrite: 0,
13933
14696
  },
13934
- contextWindow: 32000,
13935
- maxTokens: 16384,
14697
+ contextWindow: 256000,
14698
+ maxTokens: 32000,
13936
14699
  },
13937
14700
  "kwaipilot/kat-coder-pro-v2": {
13938
14701
  id: "kwaipilot/kat-coder-pro-v2",
@@ -13968,6 +14731,23 @@ export const MODELS = {
13968
14731
  contextWindow: 128000,
13969
14732
  maxTokens: 100000,
13970
14733
  },
14734
+ "meituan/longcat-flash-thinking-2601": {
14735
+ id: "meituan/longcat-flash-thinking-2601",
14736
+ name: "LongCat Flash Thinking 2601",
14737
+ api: "anthropic-messages",
14738
+ provider: "vercel-ai-gateway",
14739
+ baseUrl: "https://ai-gateway.vercel.sh",
14740
+ reasoning: true,
14741
+ input: ["text"],
14742
+ cost: {
14743
+ input: 0,
14744
+ output: 0,
14745
+ cacheRead: 0,
14746
+ cacheWrite: 0,
14747
+ },
14748
+ contextWindow: 32768,
14749
+ maxTokens: 32768,
14750
+ },
13971
14751
  "meta/llama-3.1-70b": {
13972
14752
  id: "meta/llama-3.1-70b",
13973
14753
  name: "Llama 3.1 70B Instruct",
@@ -14281,7 +15061,7 @@ export const MODELS = {
14281
15061
  provider: "vercel-ai-gateway",
14282
15062
  baseUrl: "https://ai-gateway.vercel.sh",
14283
15063
  reasoning: false,
14284
- input: ["text"],
15064
+ input: ["text", "image"],
14285
15065
  cost: {
14286
15066
  input: 0.09999999999999999,
14287
15067
  output: 0.3,
@@ -14291,6 +15071,57 @@ export const MODELS = {
14291
15071
  contextWindow: 256000,
14292
15072
  maxTokens: 256000,
14293
15073
  },
15074
+ "mistral/magistral-medium": {
15075
+ id: "mistral/magistral-medium",
15076
+ name: "Magistral Medium 2509",
15077
+ api: "anthropic-messages",
15078
+ provider: "vercel-ai-gateway",
15079
+ baseUrl: "https://ai-gateway.vercel.sh",
15080
+ reasoning: true,
15081
+ input: ["text", "image"],
15082
+ cost: {
15083
+ input: 2,
15084
+ output: 5,
15085
+ cacheRead: 0,
15086
+ cacheWrite: 0,
15087
+ },
15088
+ contextWindow: 128000,
15089
+ maxTokens: 64000,
15090
+ },
15091
+ "mistral/magistral-small": {
15092
+ id: "mistral/magistral-small",
15093
+ name: "Magistral Small 2509",
15094
+ api: "anthropic-messages",
15095
+ provider: "vercel-ai-gateway",
15096
+ baseUrl: "https://ai-gateway.vercel.sh",
15097
+ reasoning: true,
15098
+ input: ["text", "image"],
15099
+ cost: {
15100
+ input: 0.5,
15101
+ output: 1.5,
15102
+ cacheRead: 0,
15103
+ cacheWrite: 0,
15104
+ },
15105
+ contextWindow: 128000,
15106
+ maxTokens: 64000,
15107
+ },
15108
+ "mistral/ministral-14b": {
15109
+ id: "mistral/ministral-14b",
15110
+ name: "Ministral 14B",
15111
+ api: "anthropic-messages",
15112
+ provider: "vercel-ai-gateway",
15113
+ baseUrl: "https://ai-gateway.vercel.sh",
15114
+ reasoning: false,
15115
+ input: ["text", "image"],
15116
+ cost: {
15117
+ input: 0.19999999999999998,
15118
+ output: 0.19999999999999998,
15119
+ cacheRead: 0,
15120
+ cacheWrite: 0,
15121
+ },
15122
+ contextWindow: 256000,
15123
+ maxTokens: 256000,
15124
+ },
14294
15125
  "mistral/ministral-3b": {
14295
15126
  id: "mistral/ministral-3b",
14296
15127
  name: "Ministral 3B",
@@ -14325,6 +15156,23 @@ export const MODELS = {
14325
15156
  contextWindow: 128000,
14326
15157
  maxTokens: 4000,
14327
15158
  },
15159
+ "mistral/mistral-large-3": {
15160
+ id: "mistral/mistral-large-3",
15161
+ name: "Mistral Large 3",
15162
+ api: "anthropic-messages",
15163
+ provider: "vercel-ai-gateway",
15164
+ baseUrl: "https://ai-gateway.vercel.sh",
15165
+ reasoning: false,
15166
+ input: ["text", "image"],
15167
+ cost: {
15168
+ input: 0.5,
15169
+ output: 1.5,
15170
+ cacheRead: 0,
15171
+ cacheWrite: 0,
15172
+ },
15173
+ contextWindow: 256000,
15174
+ maxTokens: 256000,
15175
+ },
14328
15176
  "mistral/mistral-medium": {
14329
15177
  id: "mistral/mistral-medium",
14330
15178
  name: "Mistral Medium 3.1",
@@ -14349,7 +15197,7 @@ export const MODELS = {
14349
15197
  provider: "vercel-ai-gateway",
14350
15198
  baseUrl: "https://ai-gateway.vercel.sh",
14351
15199
  reasoning: true,
14352
- input: ["text"],
15200
+ input: ["text", "image"],
14353
15201
  cost: {
14354
15202
  input: 1.5,
14355
15203
  output: 7.5,
@@ -14453,13 +15301,13 @@ export const MODELS = {
14453
15301
  reasoning: true,
14454
15302
  input: ["text"],
14455
15303
  cost: {
14456
- input: 0.6,
14457
- output: 2.5,
14458
- cacheRead: 0.15,
15304
+ input: 0.47,
15305
+ output: 2,
15306
+ cacheRead: 0.14100000000000001,
14459
15307
  cacheWrite: 0,
14460
15308
  },
14461
- contextWindow: 262114,
14462
- maxTokens: 262114,
15309
+ contextWindow: 216144,
15310
+ maxTokens: 216144,
14463
15311
  },
14464
15312
  "moonshotai/kimi-k2.5": {
14465
15313
  id: "moonshotai/kimi-k2.5",
@@ -14529,6 +15377,23 @@ export const MODELS = {
14529
15377
  contextWindow: 262144,
14530
15378
  maxTokens: 32768,
14531
15379
  },
15380
+ "nvidia/nemotron-3-nano-30b-a3b": {
15381
+ id: "nvidia/nemotron-3-nano-30b-a3b",
15382
+ name: "Nemotron 3 Nano 30B A3B",
15383
+ api: "anthropic-messages",
15384
+ provider: "vercel-ai-gateway",
15385
+ baseUrl: "https://ai-gateway.vercel.sh",
15386
+ reasoning: true,
15387
+ input: ["text"],
15388
+ cost: {
15389
+ input: 0.049999999999999996,
15390
+ output: 0.24,
15391
+ cacheRead: 0,
15392
+ cacheWrite: 0,
15393
+ },
15394
+ contextWindow: 262144,
15395
+ maxTokens: 262144,
15396
+ },
14532
15397
  "nvidia/nemotron-3-super-120b-a12b": {
14533
15398
  id: "nvidia/nemotron-3-super-120b-a12b",
14534
15399
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -14597,6 +15462,23 @@ export const MODELS = {
14597
15462
  contextWindow: 131072,
14598
15463
  maxTokens: 131072,
14599
15464
  },
15465
+ "openai/gpt-3.5-turbo": {
15466
+ id: "openai/gpt-3.5-turbo",
15467
+ name: "GPT-3.5 Turbo",
15468
+ api: "anthropic-messages",
15469
+ provider: "vercel-ai-gateway",
15470
+ baseUrl: "https://ai-gateway.vercel.sh",
15471
+ reasoning: false,
15472
+ input: ["text"],
15473
+ cost: {
15474
+ input: 0.5,
15475
+ output: 1.5,
15476
+ cacheRead: 0,
15477
+ cacheWrite: 0,
15478
+ },
15479
+ contextWindow: 16385,
15480
+ maxTokens: 4096,
15481
+ },
14600
15482
  "openai/gpt-4.1": {
14601
15483
  id: "openai/gpt-4.1",
14602
15484
  name: "GPT-4.1",
@@ -14705,7 +15587,7 @@ export const MODELS = {
14705
15587
  api: "anthropic-messages",
14706
15588
  provider: "vercel-ai-gateway",
14707
15589
  baseUrl: "https://ai-gateway.vercel.sh",
14708
- reasoning: true,
15590
+ reasoning: false,
14709
15591
  input: ["text", "image"],
14710
15592
  cost: {
14711
15593
  input: 1.25,
@@ -14841,7 +15723,7 @@ export const MODELS = {
14841
15723
  api: "anthropic-messages",
14842
15724
  provider: "vercel-ai-gateway",
14843
15725
  baseUrl: "https://ai-gateway.vercel.sh",
14844
- reasoning: true,
15726
+ reasoning: false,
14845
15727
  input: ["text", "image"],
14846
15728
  cost: {
14847
15729
  input: 1.25,
@@ -14893,7 +15775,7 @@ export const MODELS = {
14893
15775
  api: "anthropic-messages",
14894
15776
  provider: "vercel-ai-gateway",
14895
15777
  baseUrl: "https://ai-gateway.vercel.sh",
14896
- reasoning: true,
15778
+ reasoning: false,
14897
15779
  thinkingLevelMap: { "xhigh": "xhigh" },
14898
15780
  input: ["text", "image"],
14899
15781
  cost: {
@@ -14947,7 +15829,7 @@ export const MODELS = {
14947
15829
  api: "anthropic-messages",
14948
15830
  provider: "vercel-ai-gateway",
14949
15831
  baseUrl: "https://ai-gateway.vercel.sh",
14950
- reasoning: true,
15832
+ reasoning: false,
14951
15833
  thinkingLevelMap: { "xhigh": "xhigh" },
14952
15834
  input: ["text", "image"],
14953
15835
  cost: {
@@ -15094,13 +15976,13 @@ export const MODELS = {
15094
15976
  reasoning: true,
15095
15977
  input: ["text"],
15096
15978
  cost: {
15097
- input: 0.35,
15098
- output: 0.75,
15099
- cacheRead: 0.25,
15979
+ input: 0.09999999999999999,
15980
+ output: 0.5,
15981
+ cacheRead: 0,
15100
15982
  cacheWrite: 0,
15101
15983
  },
15102
15984
  contextWindow: 131072,
15103
- maxTokens: 131000,
15985
+ maxTokens: 131072,
15104
15986
  },
15105
15987
  "openai/gpt-oss-20b": {
15106
15988
  id: "openai/gpt-oss-20b",
@@ -15238,39 +16120,22 @@ export const MODELS = {
15238
16120
  contextWindow: 200000,
15239
16121
  maxTokens: 100000,
15240
16122
  },
15241
- "perplexity/sonar": {
15242
- id: "perplexity/sonar",
15243
- name: "Sonar",
15244
- api: "anthropic-messages",
15245
- provider: "vercel-ai-gateway",
15246
- baseUrl: "https://ai-gateway.vercel.sh",
15247
- reasoning: false,
15248
- input: ["text", "image"],
15249
- cost: {
15250
- input: 0,
15251
- output: 0,
15252
- cacheRead: 0,
15253
- cacheWrite: 0,
15254
- },
15255
- contextWindow: 127000,
15256
- maxTokens: 8000,
15257
- },
15258
- "perplexity/sonar-pro": {
15259
- id: "perplexity/sonar-pro",
15260
- name: "Sonar Pro",
16123
+ "sakana/fugu-ultra": {
16124
+ id: "sakana/fugu-ultra",
16125
+ name: "Fugu Ultra",
15261
16126
  api: "anthropic-messages",
15262
16127
  provider: "vercel-ai-gateway",
15263
16128
  baseUrl: "https://ai-gateway.vercel.sh",
15264
- reasoning: false,
16129
+ reasoning: true,
15265
16130
  input: ["text", "image"],
15266
16131
  cost: {
15267
- input: 0,
15268
- output: 0,
15269
- cacheRead: 0,
16132
+ input: 5,
16133
+ output: 30,
16134
+ cacheRead: 0.5,
15270
16135
  cacheWrite: 0,
15271
16136
  },
15272
- contextWindow: 200000,
15273
- maxTokens: 8000,
16137
+ contextWindow: 1000000,
16138
+ maxTokens: 1000000,
15274
16139
  },
15275
16140
  "stepfun/step-3.5-flash": {
15276
16141
  id: "stepfun/step-3.5-flash",
@@ -15283,8 +16148,8 @@ export const MODELS = {
15283
16148
  cost: {
15284
16149
  input: 0.09,
15285
16150
  output: 0.3,
15286
- cacheRead: 0,
15287
- cacheWrite: 0.02,
16151
+ cacheRead: 0.02,
16152
+ cacheWrite: 0,
15288
16153
  },
15289
16154
  contextWindow: 262114,
15290
16155
  maxTokens: 262114,
@@ -15500,7 +16365,7 @@ export const MODELS = {
15500
16365
  provider: "vercel-ai-gateway",
15501
16366
  baseUrl: "https://ai-gateway.vercel.sh",
15502
16367
  reasoning: true,
15503
- input: ["text", "image"],
16368
+ input: ["text"],
15504
16369
  cost: {
15505
16370
  input: 0.435,
15506
16371
  output: 0.87,
@@ -15512,7 +16377,7 @@ export const MODELS = {
15512
16377
  },
15513
16378
  "zai/glm-4.5": {
15514
16379
  id: "zai/glm-4.5",
15515
- name: "GLM-4.5",
16380
+ name: "GLM 4.5",
15516
16381
  api: "anthropic-messages",
15517
16382
  provider: "vercel-ai-gateway",
15518
16383
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -15621,13 +16486,13 @@ export const MODELS = {
15621
16486
  reasoning: true,
15622
16487
  input: ["text"],
15623
16488
  cost: {
15624
- input: 2.25,
15625
- output: 2.75,
15626
- cacheRead: 2.25,
16489
+ input: 0.6,
16490
+ output: 2.2,
16491
+ cacheRead: 0.12,
15627
16492
  cacheWrite: 0,
15628
16493
  },
15629
- contextWindow: 131000,
15630
- maxTokens: 40000,
16494
+ contextWindow: 200000,
16495
+ maxTokens: 120000,
15631
16496
  },
15632
16497
  "zai/glm-4.7-flash": {
15633
16498
  id: "zai/glm-4.7-flash",
@@ -15672,8 +16537,8 @@ export const MODELS = {
15672
16537
  reasoning: true,
15673
16538
  input: ["text"],
15674
16539
  cost: {
15675
- input: 1,
15676
- output: 3.1999999999999997,
16540
+ input: 0.95,
16541
+ output: 3.15,
15677
16542
  cacheRead: 0.19999999999999998,
15678
16543
  cacheWrite: 0,
15679
16544
  },
@@ -15704,15 +16569,15 @@ export const MODELS = {
15704
16569
  provider: "vercel-ai-gateway",
15705
16570
  baseUrl: "https://ai-gateway.vercel.sh",
15706
16571
  reasoning: true,
15707
- input: ["text", "image"],
16572
+ input: ["text"],
15708
16573
  cost: {
15709
- input: 1.4,
15710
- output: 4.4,
16574
+ input: 1.3,
16575
+ output: 4.300000000000001,
15711
16576
  cacheRead: 0.26,
15712
16577
  cacheWrite: 0,
15713
16578
  },
15714
- contextWindow: 202800,
15715
- maxTokens: 64000,
16579
+ contextWindow: 202000,
16580
+ maxTokens: 202000,
15716
16581
  },
15717
16582
  "zai/glm-5.2": {
15718
16583
  id: "zai/glm-5.2",
@@ -15731,6 +16596,23 @@ export const MODELS = {
15731
16596
  contextWindow: 1000000,
15732
16597
  maxTokens: 128000,
15733
16598
  },
16599
+ "zai/glm-5.2-fast": {
16600
+ id: "zai/glm-5.2-fast",
16601
+ name: "GLM 5.2 Fast",
16602
+ api: "anthropic-messages",
16603
+ provider: "vercel-ai-gateway",
16604
+ baseUrl: "https://ai-gateway.vercel.sh",
16605
+ reasoning: true,
16606
+ input: ["text"],
16607
+ cost: {
16608
+ input: 3,
16609
+ output: 10.25,
16610
+ cacheRead: 0.5,
16611
+ cacheWrite: 0,
16612
+ },
16613
+ contextWindow: 1000000,
16614
+ maxTokens: 128000,
16615
+ },
15734
16616
  "zai/glm-5v-turbo": {
15735
16617
  id: "zai/glm-5v-turbo",
15736
16618
  name: "GLM 5V Turbo",
@@ -15846,9 +16728,9 @@ export const MODELS = {
15846
16728
  reasoning: true,
15847
16729
  input: ["text", "image"],
15848
16730
  cost: {
15849
- input: 0.4,
15850
- output: 2,
15851
- cacheRead: 0.08,
16731
+ input: 0.14,
16732
+ output: 0.28,
16733
+ cacheRead: 0.0028,
15852
16734
  cacheWrite: 0,
15853
16735
  },
15854
16736
  contextWindow: 1048576,
@@ -15863,9 +16745,9 @@ export const MODELS = {
15863
16745
  reasoning: true,
15864
16746
  input: ["text"],
15865
16747
  cost: {
15866
- input: 1,
15867
- output: 3,
15868
- cacheRead: 0.2,
16748
+ input: 0.435,
16749
+ output: 0.87,
16750
+ cacheRead: 0.0036,
15869
16751
  cacheWrite: 0,
15870
16752
  },
15871
16753
  contextWindow: 1048576,
@@ -15899,9 +16781,9 @@ export const MODELS = {
15899
16781
  reasoning: true,
15900
16782
  input: ["text", "image"],
15901
16783
  cost: {
15902
- input: 0.4,
15903
- output: 2,
15904
- cacheRead: 0.08,
16784
+ input: 0.14,
16785
+ output: 0.28,
16786
+ cacheRead: 0.0028,
15905
16787
  cacheWrite: 0,
15906
16788
  },
15907
16789
  contextWindow: 1048576,
@@ -15916,9 +16798,9 @@ export const MODELS = {
15916
16798
  reasoning: true,
15917
16799
  input: ["text"],
15918
16800
  cost: {
15919
- input: 1,
15920
- output: 3,
15921
- cacheRead: 0.2,
16801
+ input: 0.435,
16802
+ output: 0.87,
16803
+ cacheRead: 0.0036,
15922
16804
  cacheWrite: 0,
15923
16805
  },
15924
16806
  contextWindow: 1048576,
@@ -15952,9 +16834,9 @@ export const MODELS = {
15952
16834
  reasoning: true,
15953
16835
  input: ["text", "image"],
15954
16836
  cost: {
15955
- input: 0.4,
15956
- output: 2,
15957
- cacheRead: 0.08,
16837
+ input: 0.14,
16838
+ output: 0.28,
16839
+ cacheRead: 0.0028,
15958
16840
  cacheWrite: 0,
15959
16841
  },
15960
16842
  contextWindow: 1048576,
@@ -15969,9 +16851,9 @@ export const MODELS = {
15969
16851
  reasoning: true,
15970
16852
  input: ["text"],
15971
16853
  cost: {
15972
- input: 1,
15973
- output: 3,
15974
- cacheRead: 0.2,
16854
+ input: 0.435,
16855
+ output: 0.87,
16856
+ cacheRead: 0.0036,
15975
16857
  cacheWrite: 0,
15976
16858
  },
15977
16859
  contextWindow: 1048576,
@@ -16005,9 +16887,9 @@ export const MODELS = {
16005
16887
  reasoning: true,
16006
16888
  input: ["text", "image"],
16007
16889
  cost: {
16008
- input: 0.4,
16009
- output: 2,
16010
- cacheRead: 0.08,
16890
+ input: 0.14,
16891
+ output: 0.28,
16892
+ cacheRead: 0.0028,
16011
16893
  cacheWrite: 0,
16012
16894
  },
16013
16895
  contextWindow: 1048576,
@@ -16022,9 +16904,9 @@ export const MODELS = {
16022
16904
  reasoning: true,
16023
16905
  input: ["text"],
16024
16906
  cost: {
16025
- input: 1,
16026
- output: 3,
16027
- cacheRead: 0.2,
16907
+ input: 0.435,
16908
+ output: 0.87,
16909
+ cacheRead: 0.0036,
16028
16910
  cacheWrite: 0,
16029
16911
  },
16030
16912
  contextWindow: 1048576,