@elyracode/ai 0.9.10 → 0.9.11

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.
@@ -3775,7 +3775,7 @@ export const MODELS = {
3775
3775
  cacheRead: 0.26,
3776
3776
  cacheWrite: 0,
3777
3777
  },
3778
- contextWindow: 1048576,
3778
+ contextWindow: 1048575,
3779
3779
  maxTokens: 131072,
3780
3780
  },
3781
3781
  "accounts/fireworks/models/gpt-oss-120b": {
@@ -3914,6 +3914,23 @@ export const MODELS = {
3914
3914
  contextWindow: 202800,
3915
3915
  maxTokens: 131072,
3916
3916
  },
3917
+ "accounts/fireworks/routers/glm-5p2-fast": {
3918
+ id: "accounts/fireworks/routers/glm-5p2-fast",
3919
+ name: "GLM 5.2 Fast",
3920
+ api: "anthropic-messages",
3921
+ provider: "fireworks",
3922
+ baseUrl: "https://api.fireworks.ai/inference",
3923
+ reasoning: true,
3924
+ input: ["text"],
3925
+ cost: {
3926
+ input: 2.1,
3927
+ output: 6.6,
3928
+ cacheRead: 0.21,
3929
+ cacheWrite: 0,
3930
+ },
3931
+ contextWindow: 1048575,
3932
+ maxTokens: 131072,
3933
+ },
3917
3934
  "accounts/fireworks/routers/kimi-k2p6-fast": {
3918
3935
  id: "accounts/fireworks/routers/kimi-k2p6-fast",
3919
3936
  name: "Kimi K2.6 Fast",
@@ -4667,42 +4684,6 @@ export const MODELS = {
4667
4684
  contextWindow: 262144,
4668
4685
  maxTokens: 32768,
4669
4686
  },
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
4687
  },
4707
4688
  "google-vertex": {
4708
4689
  "gemini-1.5-flash": {
@@ -5054,6 +5035,24 @@ export const MODELS = {
5054
5035
  },
5055
5036
  },
5056
5037
  "huggingface": {
5038
+ "MiniMaxAI/MiniMax-M2": {
5039
+ id: "MiniMaxAI/MiniMax-M2",
5040
+ name: "MiniMax-M2",
5041
+ api: "openai-completions",
5042
+ provider: "huggingface",
5043
+ baseUrl: "https://router.huggingface.co/v1",
5044
+ compat: { "supportsDeveloperRole": false },
5045
+ reasoning: true,
5046
+ input: ["text"],
5047
+ cost: {
5048
+ input: 0.3,
5049
+ output: 1.2,
5050
+ cacheRead: 0,
5051
+ cacheWrite: 0,
5052
+ },
5053
+ contextWindow: 204800,
5054
+ maxTokens: 128000,
5055
+ },
5057
5056
  "MiniMaxAI/MiniMax-M2.1": {
5058
5057
  id: "MiniMaxAI/MiniMax-M2.1",
5059
5058
  name: "MiniMax-M2.1",
@@ -5108,6 +5107,42 @@ export const MODELS = {
5108
5107
  contextWindow: 204800,
5109
5108
  maxTokens: 131072,
5110
5109
  },
5110
+ "MiniMaxAI/MiniMax-M3": {
5111
+ id: "MiniMaxAI/MiniMax-M3",
5112
+ name: "MiniMax-M3",
5113
+ api: "openai-completions",
5114
+ provider: "huggingface",
5115
+ baseUrl: "https://router.huggingface.co/v1",
5116
+ compat: { "supportsDeveloperRole": false },
5117
+ reasoning: true,
5118
+ input: ["text", "image"],
5119
+ cost: {
5120
+ input: 0.3,
5121
+ output: 1.2,
5122
+ cacheRead: 0,
5123
+ cacheWrite: 0,
5124
+ },
5125
+ contextWindow: 524288,
5126
+ maxTokens: 128000,
5127
+ },
5128
+ "Qwen/Qwen3-235B-A22B": {
5129
+ id: "Qwen/Qwen3-235B-A22B",
5130
+ name: "Qwen3 235B-A22B",
5131
+ api: "openai-completions",
5132
+ provider: "huggingface",
5133
+ baseUrl: "https://router.huggingface.co/v1",
5134
+ compat: { "supportsDeveloperRole": false },
5135
+ reasoning: true,
5136
+ input: ["text"],
5137
+ cost: {
5138
+ input: 0.2,
5139
+ output: 0.8,
5140
+ cacheRead: 0,
5141
+ cacheWrite: 0,
5142
+ },
5143
+ contextWindow: 40960,
5144
+ maxTokens: 16384,
5145
+ },
5111
5146
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5112
5147
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5113
5148
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5126,6 +5161,42 @@ export const MODELS = {
5126
5161
  contextWindow: 262144,
5127
5162
  maxTokens: 131072,
5128
5163
  },
5164
+ "Qwen/Qwen3-32B": {
5165
+ id: "Qwen/Qwen3-32B",
5166
+ name: "Qwen3 32B",
5167
+ api: "openai-completions",
5168
+ provider: "huggingface",
5169
+ baseUrl: "https://router.huggingface.co/v1",
5170
+ compat: { "supportsDeveloperRole": false },
5171
+ reasoning: true,
5172
+ input: ["text"],
5173
+ cost: {
5174
+ input: 0.29,
5175
+ output: 0.59,
5176
+ cacheRead: 0,
5177
+ cacheWrite: 0,
5178
+ },
5179
+ contextWindow: 131072,
5180
+ maxTokens: 16384,
5181
+ },
5182
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
5183
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
5184
+ name: "Qwen3-Coder 30B-A3B Instruct",
5185
+ api: "openai-completions",
5186
+ provider: "huggingface",
5187
+ baseUrl: "https://router.huggingface.co/v1",
5188
+ compat: { "supportsDeveloperRole": false },
5189
+ reasoning: false,
5190
+ input: ["text"],
5191
+ cost: {
5192
+ input: 0.07,
5193
+ output: 0.26,
5194
+ cacheRead: 0,
5195
+ cacheWrite: 0,
5196
+ },
5197
+ contextWindow: 262144,
5198
+ maxTokens: 65536,
5199
+ },
5129
5200
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5130
5201
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5131
5202
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5198,6 +5269,60 @@ export const MODELS = {
5198
5269
  contextWindow: 262144,
5199
5270
  maxTokens: 131072,
5200
5271
  },
5272
+ "Qwen/Qwen3.5-122B-A10B": {
5273
+ id: "Qwen/Qwen3.5-122B-A10B",
5274
+ name: "Qwen3.5 122B-A10B",
5275
+ api: "openai-completions",
5276
+ provider: "huggingface",
5277
+ baseUrl: "https://router.huggingface.co/v1",
5278
+ compat: { "supportsDeveloperRole": false },
5279
+ reasoning: true,
5280
+ input: ["text", "image"],
5281
+ cost: {
5282
+ input: 0.4,
5283
+ output: 3.2,
5284
+ cacheRead: 0,
5285
+ cacheWrite: 0,
5286
+ },
5287
+ contextWindow: 262144,
5288
+ maxTokens: 65536,
5289
+ },
5290
+ "Qwen/Qwen3.5-27B": {
5291
+ id: "Qwen/Qwen3.5-27B",
5292
+ name: "Qwen3.5 27B",
5293
+ api: "openai-completions",
5294
+ provider: "huggingface",
5295
+ baseUrl: "https://router.huggingface.co/v1",
5296
+ compat: { "supportsDeveloperRole": false },
5297
+ reasoning: true,
5298
+ input: ["text", "image"],
5299
+ cost: {
5300
+ input: 0.3,
5301
+ output: 2.4,
5302
+ cacheRead: 0,
5303
+ cacheWrite: 0,
5304
+ },
5305
+ contextWindow: 262144,
5306
+ maxTokens: 65536,
5307
+ },
5308
+ "Qwen/Qwen3.5-35B-A3B": {
5309
+ id: "Qwen/Qwen3.5-35B-A3B",
5310
+ name: "Qwen3.5 35B-A3B",
5311
+ api: "openai-completions",
5312
+ provider: "huggingface",
5313
+ baseUrl: "https://router.huggingface.co/v1",
5314
+ compat: { "supportsDeveloperRole": false },
5315
+ reasoning: true,
5316
+ input: ["text", "image"],
5317
+ cost: {
5318
+ input: 0.25,
5319
+ output: 2,
5320
+ cacheRead: 0,
5321
+ cacheWrite: 0,
5322
+ },
5323
+ contextWindow: 262144,
5324
+ maxTokens: 65536,
5325
+ },
5201
5326
  "Qwen/Qwen3.5-397B-A17B": {
5202
5327
  id: "Qwen/Qwen3.5-397B-A17B",
5203
5328
  name: "Qwen3.5-397B-A17B",
@@ -5216,6 +5341,96 @@ export const MODELS = {
5216
5341
  contextWindow: 262144,
5217
5342
  maxTokens: 32768,
5218
5343
  },
5344
+ "Qwen/Qwen3.5-9B": {
5345
+ id: "Qwen/Qwen3.5-9B",
5346
+ name: "Qwen3.5 9B",
5347
+ api: "openai-completions",
5348
+ provider: "huggingface",
5349
+ baseUrl: "https://router.huggingface.co/v1",
5350
+ compat: { "supportsDeveloperRole": false },
5351
+ reasoning: true,
5352
+ input: ["text", "image"],
5353
+ cost: {
5354
+ input: 0.17,
5355
+ output: 0.25,
5356
+ cacheRead: 0,
5357
+ cacheWrite: 0,
5358
+ },
5359
+ contextWindow: 262144,
5360
+ maxTokens: 65536,
5361
+ },
5362
+ "Qwen/Qwen3.6-27B": {
5363
+ id: "Qwen/Qwen3.6-27B",
5364
+ name: "Qwen3.6 27B",
5365
+ api: "openai-completions",
5366
+ provider: "huggingface",
5367
+ baseUrl: "https://router.huggingface.co/v1",
5368
+ compat: { "supportsDeveloperRole": false },
5369
+ reasoning: true,
5370
+ input: ["text", "image"],
5371
+ cost: {
5372
+ input: 0.47,
5373
+ output: 3.19,
5374
+ cacheRead: 0,
5375
+ cacheWrite: 0,
5376
+ },
5377
+ contextWindow: 262144,
5378
+ maxTokens: 65536,
5379
+ },
5380
+ "Qwen/Qwen3.6-35B-A3B": {
5381
+ id: "Qwen/Qwen3.6-35B-A3B",
5382
+ name: "Qwen3.6 35B-A3B",
5383
+ api: "openai-completions",
5384
+ provider: "huggingface",
5385
+ baseUrl: "https://router.huggingface.co/v1",
5386
+ compat: { "supportsDeveloperRole": false },
5387
+ reasoning: true,
5388
+ input: ["text", "image"],
5389
+ cost: {
5390
+ input: 0.15,
5391
+ output: 0.95,
5392
+ cacheRead: 0,
5393
+ cacheWrite: 0,
5394
+ },
5395
+ contextWindow: 262144,
5396
+ maxTokens: 65536,
5397
+ },
5398
+ "XiaomiMiMo/MiMo-V2.5-Pro": {
5399
+ id: "XiaomiMiMo/MiMo-V2.5-Pro",
5400
+ name: "MiMo-V2.5-Pro",
5401
+ api: "openai-completions",
5402
+ provider: "huggingface",
5403
+ baseUrl: "https://router.huggingface.co/v1",
5404
+ compat: { "supportsDeveloperRole": false },
5405
+ reasoning: true,
5406
+ input: ["text"],
5407
+ cost: {
5408
+ input: 1,
5409
+ output: 3,
5410
+ cacheRead: 0,
5411
+ cacheWrite: 0,
5412
+ },
5413
+ contextWindow: 1048576,
5414
+ maxTokens: 131072,
5415
+ },
5416
+ "deepseek-ai/DeepSeek-R1": {
5417
+ id: "deepseek-ai/DeepSeek-R1",
5418
+ name: "DeepSeek-R1",
5419
+ api: "openai-completions",
5420
+ provider: "huggingface",
5421
+ baseUrl: "https://router.huggingface.co/v1",
5422
+ compat: { "supportsDeveloperRole": false },
5423
+ reasoning: true,
5424
+ input: ["text"],
5425
+ cost: {
5426
+ input: 0.7,
5427
+ output: 2.5,
5428
+ cacheRead: 0,
5429
+ cacheWrite: 0,
5430
+ },
5431
+ contextWindow: 64000,
5432
+ maxTokens: 32768,
5433
+ },
5219
5434
  "deepseek-ai/DeepSeek-R1-0528": {
5220
5435
  id: "deepseek-ai/DeepSeek-R1-0528",
5221
5436
  name: "DeepSeek-R1-0528",
@@ -5252,6 +5467,24 @@ export const MODELS = {
5252
5467
  contextWindow: 163840,
5253
5468
  maxTokens: 65536,
5254
5469
  },
5470
+ "deepseek-ai/DeepSeek-V4-Flash": {
5471
+ id: "deepseek-ai/DeepSeek-V4-Flash",
5472
+ name: "DeepSeek V4 Flash",
5473
+ api: "openai-completions",
5474
+ provider: "huggingface",
5475
+ baseUrl: "https://router.huggingface.co/v1",
5476
+ compat: { "supportsDeveloperRole": false },
5477
+ reasoning: true,
5478
+ input: ["text"],
5479
+ cost: {
5480
+ input: 0.14,
5481
+ output: 0.28,
5482
+ cacheRead: 0,
5483
+ cacheWrite: 0,
5484
+ },
5485
+ contextWindow: 1048576,
5486
+ maxTokens: 384000,
5487
+ },
5255
5488
  "deepseek-ai/DeepSeek-V4-Pro": {
5256
5489
  id: "deepseek-ai/DeepSeek-V4-Pro",
5257
5490
  name: "DeepSeek V4 Pro",
@@ -5270,6 +5503,60 @@ export const MODELS = {
5270
5503
  contextWindow: 1048576,
5271
5504
  maxTokens: 393216,
5272
5505
  },
5506
+ "google/gemma-4-26B-A4B-it": {
5507
+ id: "google/gemma-4-26B-A4B-it",
5508
+ name: "Gemma 4 26B A4B IT",
5509
+ api: "openai-completions",
5510
+ provider: "huggingface",
5511
+ baseUrl: "https://router.huggingface.co/v1",
5512
+ compat: { "supportsDeveloperRole": false },
5513
+ reasoning: true,
5514
+ input: ["text", "image"],
5515
+ cost: {
5516
+ input: 0.13,
5517
+ output: 0.4,
5518
+ cacheRead: 0,
5519
+ cacheWrite: 0,
5520
+ },
5521
+ contextWindow: 262144,
5522
+ maxTokens: 32768,
5523
+ },
5524
+ "google/gemma-4-31B-it": {
5525
+ id: "google/gemma-4-31B-it",
5526
+ name: "Gemma 4 31B IT",
5527
+ api: "openai-completions",
5528
+ provider: "huggingface",
5529
+ baseUrl: "https://router.huggingface.co/v1",
5530
+ compat: { "supportsDeveloperRole": false },
5531
+ reasoning: true,
5532
+ input: ["text", "image"],
5533
+ cost: {
5534
+ input: 0.14,
5535
+ output: 0.4,
5536
+ cacheRead: 0,
5537
+ cacheWrite: 0,
5538
+ },
5539
+ contextWindow: 262144,
5540
+ maxTokens: 32768,
5541
+ },
5542
+ "meta-llama/Llama-3.3-70B-Instruct": {
5543
+ id: "meta-llama/Llama-3.3-70B-Instruct",
5544
+ name: "Llama-3.3-70B-Instruct",
5545
+ api: "openai-completions",
5546
+ provider: "huggingface",
5547
+ baseUrl: "https://router.huggingface.co/v1",
5548
+ compat: { "supportsDeveloperRole": false },
5549
+ reasoning: false,
5550
+ input: ["text"],
5551
+ cost: {
5552
+ input: 0.59,
5553
+ output: 0.79,
5554
+ cacheRead: 0,
5555
+ cacheWrite: 0,
5556
+ },
5557
+ contextWindow: 131072,
5558
+ maxTokens: 4096,
5559
+ },
5273
5560
  "moonshotai/Kimi-K2-Instruct": {
5274
5561
  id: "moonshotai/Kimi-K2-Instruct",
5275
5562
  name: "Kimi-K2-Instruct",
@@ -5360,6 +5647,132 @@ export const MODELS = {
5360
5647
  contextWindow: 262144,
5361
5648
  maxTokens: 262144,
5362
5649
  },
5650
+ "moonshotai/Kimi-K2.7-Code": {
5651
+ id: "moonshotai/Kimi-K2.7-Code",
5652
+ name: "Kimi K2.7 Code",
5653
+ api: "openai-completions",
5654
+ provider: "huggingface",
5655
+ baseUrl: "https://router.huggingface.co/v1",
5656
+ compat: { "supportsDeveloperRole": false },
5657
+ reasoning: true,
5658
+ input: ["text", "image"],
5659
+ cost: {
5660
+ input: 0.95,
5661
+ output: 4,
5662
+ cacheRead: 0,
5663
+ cacheWrite: 0,
5664
+ },
5665
+ contextWindow: 262144,
5666
+ maxTokens: 262144,
5667
+ },
5668
+ "stepfun-ai/Step-3.5-Flash": {
5669
+ id: "stepfun-ai/Step-3.5-Flash",
5670
+ name: "Step 3.5 Flash",
5671
+ api: "openai-completions",
5672
+ provider: "huggingface",
5673
+ baseUrl: "https://router.huggingface.co/v1",
5674
+ compat: { "supportsDeveloperRole": false },
5675
+ reasoning: true,
5676
+ input: ["text"],
5677
+ cost: {
5678
+ input: 0.1,
5679
+ output: 0.3,
5680
+ cacheRead: 0,
5681
+ cacheWrite: 0,
5682
+ },
5683
+ contextWindow: 262144,
5684
+ maxTokens: 256000,
5685
+ },
5686
+ "stepfun-ai/Step-3.7-Flash": {
5687
+ id: "stepfun-ai/Step-3.7-Flash",
5688
+ name: "Step 3.7 Flash",
5689
+ api: "openai-completions",
5690
+ provider: "huggingface",
5691
+ baseUrl: "https://router.huggingface.co/v1",
5692
+ compat: { "supportsDeveloperRole": false },
5693
+ reasoning: true,
5694
+ input: ["text", "image"],
5695
+ cost: {
5696
+ input: 0.2,
5697
+ output: 1.15,
5698
+ cacheRead: 0,
5699
+ cacheWrite: 0,
5700
+ },
5701
+ contextWindow: 262144,
5702
+ maxTokens: 256000,
5703
+ },
5704
+ "zai-org/GLM-4.5": {
5705
+ id: "zai-org/GLM-4.5",
5706
+ name: "GLM-4.5",
5707
+ api: "openai-completions",
5708
+ provider: "huggingface",
5709
+ baseUrl: "https://router.huggingface.co/v1",
5710
+ compat: { "supportsDeveloperRole": false },
5711
+ reasoning: true,
5712
+ input: ["text"],
5713
+ cost: {
5714
+ input: 0.6,
5715
+ output: 2.2,
5716
+ cacheRead: 0,
5717
+ cacheWrite: 0,
5718
+ },
5719
+ contextWindow: 131072,
5720
+ maxTokens: 98304,
5721
+ },
5722
+ "zai-org/GLM-4.5-Air": {
5723
+ id: "zai-org/GLM-4.5-Air",
5724
+ name: "GLM-4.5-Air",
5725
+ api: "openai-completions",
5726
+ provider: "huggingface",
5727
+ baseUrl: "https://router.huggingface.co/v1",
5728
+ compat: { "supportsDeveloperRole": false },
5729
+ reasoning: true,
5730
+ input: ["text"],
5731
+ cost: {
5732
+ input: 0.13,
5733
+ output: 0.85,
5734
+ cacheRead: 0,
5735
+ cacheWrite: 0,
5736
+ },
5737
+ contextWindow: 131072,
5738
+ maxTokens: 98304,
5739
+ },
5740
+ "zai-org/GLM-4.5V": {
5741
+ id: "zai-org/GLM-4.5V",
5742
+ name: "GLM-4.5V",
5743
+ api: "openai-completions",
5744
+ provider: "huggingface",
5745
+ baseUrl: "https://router.huggingface.co/v1",
5746
+ compat: { "supportsDeveloperRole": false },
5747
+ reasoning: true,
5748
+ input: ["text", "image"],
5749
+ cost: {
5750
+ input: 0.6,
5751
+ output: 1.8,
5752
+ cacheRead: 0,
5753
+ cacheWrite: 0,
5754
+ },
5755
+ contextWindow: 65536,
5756
+ maxTokens: 16384,
5757
+ },
5758
+ "zai-org/GLM-4.6": {
5759
+ id: "zai-org/GLM-4.6",
5760
+ name: "GLM-4.6",
5761
+ api: "openai-completions",
5762
+ provider: "huggingface",
5763
+ baseUrl: "https://router.huggingface.co/v1",
5764
+ compat: { "supportsDeveloperRole": false },
5765
+ reasoning: true,
5766
+ input: ["text"],
5767
+ cost: {
5768
+ input: 0.55,
5769
+ output: 2.2,
5770
+ cacheRead: 0,
5771
+ cacheWrite: 0,
5772
+ },
5773
+ contextWindow: 204800,
5774
+ maxTokens: 131072,
5775
+ },
5363
5776
  "zai-org/GLM-4.7": {
5364
5777
  id: "zai-org/GLM-4.7",
5365
5778
  name: "GLM-4.7",
@@ -5432,6 +5845,24 @@ export const MODELS = {
5432
5845
  contextWindow: 202752,
5433
5846
  maxTokens: 131072,
5434
5847
  },
5848
+ "zai-org/GLM-5.2": {
5849
+ id: "zai-org/GLM-5.2",
5850
+ name: "GLM-5.2",
5851
+ api: "openai-completions",
5852
+ provider: "huggingface",
5853
+ baseUrl: "https://router.huggingface.co/v1",
5854
+ compat: { "supportsDeveloperRole": false },
5855
+ reasoning: true,
5856
+ input: ["text"],
5857
+ cost: {
5858
+ input: 1.4,
5859
+ output: 4.4,
5860
+ cacheRead: 0,
5861
+ cacheWrite: 0,
5862
+ },
5863
+ contextWindow: 262144,
5864
+ maxTokens: 131072,
5865
+ },
5435
5866
  },
5436
5867
  "kimi-coding": {
5437
5868
  "k2p7": {
@@ -5533,12 +5964,12 @@ export const MODELS = {
5533
5964
  reasoning: true,
5534
5965
  input: ["text", "image"],
5535
5966
  cost: {
5536
- input: 0.6,
5537
- output: 2.4,
5538
- cacheRead: 0.12,
5967
+ input: 0.3,
5968
+ output: 1.2,
5969
+ cacheRead: 0.06,
5539
5970
  cacheWrite: 0,
5540
5971
  },
5541
- contextWindow: 512000,
5972
+ contextWindow: 1000000,
5542
5973
  maxTokens: 128000,
5543
5974
  },
5544
5975
  },
@@ -5586,12 +6017,12 @@ export const MODELS = {
5586
6017
  reasoning: true,
5587
6018
  input: ["text", "image"],
5588
6019
  cost: {
5589
- input: 0.6,
5590
- output: 2.4,
5591
- cacheRead: 0.12,
6020
+ input: 0.3,
6021
+ output: 1.2,
6022
+ cacheRead: 0.06,
5592
6023
  cacheWrite: 0,
5593
6024
  },
5594
- contextWindow: 512000,
6025
+ contextWindow: 1000000,
5595
6026
  maxTokens: 128000,
5596
6027
  },
5597
6028
  },
@@ -7606,6 +8037,23 @@ export const MODELS = {
7606
8037
  contextWindow: 204800,
7607
8038
  maxTokens: 131072,
7608
8039
  },
8040
+ "glm-5.2": {
8041
+ id: "glm-5.2",
8042
+ name: "GLM-5.2",
8043
+ api: "openai-completions",
8044
+ provider: "opencode",
8045
+ baseUrl: "https://opencode.ai/zen/v1",
8046
+ reasoning: true,
8047
+ input: ["text"],
8048
+ cost: {
8049
+ input: 1.4,
8050
+ output: 4.4,
8051
+ cacheRead: 0.26,
8052
+ cacheWrite: 0,
8053
+ },
8054
+ contextWindow: 1000000,
8055
+ maxTokens: 131072,
8056
+ },
7609
8057
  "gpt-5": {
7610
8058
  id: "gpt-5",
7611
8059
  name: "GPT-5",
@@ -8237,7 +8685,7 @@ export const MODELS = {
8237
8685
  cacheRead: 0.02,
8238
8686
  cacheWrite: 0,
8239
8687
  },
8240
- contextWindow: 512000,
8688
+ contextWindow: 1000000,
8241
8689
  maxTokens: 131072,
8242
8690
  },
8243
8691
  "qwen3.6-plus": {
@@ -8667,13 +9115,13 @@ export const MODELS = {
8667
9115
  reasoning: true,
8668
9116
  input: ["text"],
8669
9117
  cost: {
8670
- input: 0.22,
8671
- output: 0.85,
9118
+ input: 0.25,
9119
+ output: 0.7999999999999999,
8672
9120
  cacheRead: 0.06,
8673
9121
  cacheWrite: 0,
8674
9122
  },
8675
9123
  contextWindow: 262144,
8676
- maxTokens: 262144,
9124
+ maxTokens: 80000,
8677
9125
  },
8678
9126
  "arcee-ai/trinity-mini": {
8679
9127
  id: "arcee-ai/trinity-mini",
@@ -8958,7 +9406,7 @@ export const MODELS = {
8958
9406
  cost: {
8959
9407
  input: 0.2288,
8960
9408
  output: 0.3432,
8961
- cacheRead: 0,
9409
+ cacheRead: 0.02288,
8962
9410
  cacheWrite: 0,
8963
9411
  },
8964
9412
  contextWindow: 131072,
@@ -9019,23 +9467,6 @@ export const MODELS = {
9019
9467
  contextWindow: 1048576,
9020
9468
  maxTokens: 384000,
9021
9469
  },
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
9470
  "google/gemini-2.5-flash": {
9040
9471
  id: "google/gemini-2.5-flash",
9041
9472
  name: "Google: Gemini 2.5 Flash",
@@ -9640,9 +10071,9 @@ export const MODELS = {
9640
10071
  reasoning: true,
9641
10072
  input: ["text"],
9642
10073
  cost: {
9643
- input: 0.15,
9644
- output: 0.8999999999999999,
9645
- cacheRead: 0.049999999999999996,
10074
+ input: 0.12,
10075
+ output: 0.48,
10076
+ cacheRead: 0,
9646
10077
  cacheWrite: 0,
9647
10078
  },
9648
10079
  contextWindow: 204800,
@@ -9657,13 +10088,13 @@ export const MODELS = {
9657
10088
  reasoning: true,
9658
10089
  input: ["text"],
9659
10090
  cost: {
9660
- input: 0.25,
9661
- output: 1,
9662
- cacheRead: 0.049999999999999996,
10091
+ input: 0.18,
10092
+ output: 0.72,
10093
+ cacheRead: 0,
9663
10094
  cacheWrite: 0,
9664
10095
  },
9665
10096
  contextWindow: 204800,
9666
- maxTokens: 131072,
10097
+ maxTokens: 196608,
9667
10098
  },
9668
10099
  "minimax/minimax-m3": {
9669
10100
  id: "minimax/minimax-m3",
@@ -10016,7 +10447,7 @@ export const MODELS = {
10016
10447
  cost: {
10017
10448
  input: 0.6,
10018
10449
  output: 2.5,
10019
- cacheRead: 0,
10450
+ cacheRead: 0.6,
10020
10451
  cacheWrite: 0,
10021
10452
  },
10022
10453
  contextWindow: 262144,
@@ -10048,9 +10479,9 @@ export const MODELS = {
10048
10479
  reasoning: true,
10049
10480
  input: ["text", "image"],
10050
10481
  cost: {
10051
- input: 0.67,
10052
- output: 3.5,
10053
- cacheRead: 0.19999999999999998,
10482
+ input: 0.66,
10483
+ output: 3.41,
10484
+ cacheRead: 0.144,
10054
10485
  cacheWrite: 0,
10055
10486
  },
10056
10487
  contextWindow: 262144,
@@ -10073,23 +10504,6 @@ export const MODELS = {
10073
10504
  contextWindow: 262144,
10074
10505
  maxTokens: 16384,
10075
10506
  },
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
10507
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
10094
10508
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
10095
10509
  name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
@@ -10167,13 +10581,13 @@ export const MODELS = {
10167
10581
  reasoning: true,
10168
10582
  input: ["text"],
10169
10583
  cost: {
10170
- input: 0.09,
10171
- output: 0.44999999999999996,
10584
+ input: 0.08499999999999999,
10585
+ output: 0.39999999999999997,
10172
10586
  cacheRead: 0,
10173
10587
  cacheWrite: 0,
10174
10588
  },
10175
10589
  contextWindow: 1000000,
10176
- maxTokens: 4096,
10590
+ maxTokens: 16384,
10177
10591
  },
10178
10592
  "nvidia/nemotron-3-super-120b-a12b:free": {
10179
10593
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -10859,13 +11273,13 @@ export const MODELS = {
10859
11273
  reasoning: true,
10860
11274
  input: ["text"],
10861
11275
  cost: {
10862
- input: 0.039,
10863
- output: 0.18,
11276
+ input: 0.03,
11277
+ output: 0.15,
10864
11278
  cacheRead: 0,
10865
11279
  cacheWrite: 0,
10866
11280
  },
10867
11281
  contextWindow: 131072,
10868
- maxTokens: 4096,
11282
+ maxTokens: 131072,
10869
11283
  },
10870
11284
  "openai/gpt-oss-120b:free": {
10871
11285
  id: "openai/gpt-oss-120b:free",
@@ -11207,23 +11621,6 @@ export const MODELS = {
11207
11621
  contextWindow: 262144,
11208
11622
  maxTokens: 32768,
11209
11623
  },
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
11624
  "qwen/qwen-2.5-72b-instruct": {
11228
11625
  id: "qwen/qwen-2.5-72b-instruct",
11229
11626
  name: "Qwen2.5 72B Instruct",
@@ -11813,11 +12210,11 @@ export const MODELS = {
11813
12210
  cost: {
11814
12211
  input: 0.14,
11815
12212
  output: 1,
11816
- cacheRead: 0,
12213
+ cacheRead: 0.049999999999999996,
11817
12214
  cacheWrite: 0,
11818
12215
  },
11819
12216
  contextWindow: 262144,
11820
- maxTokens: 262144,
12217
+ maxTokens: 81920,
11821
12218
  },
11822
12219
  "qwen/qwen3.5-397b-a17b": {
11823
12220
  id: "qwen/qwen3.5-397b-a17b",
@@ -11914,7 +12311,7 @@ export const MODELS = {
11914
12311
  input: ["text", "image"],
11915
12312
  cost: {
11916
12313
  input: 0.28850000000000003,
11917
- output: 3.17,
12314
+ output: 2.65,
11918
12315
  cacheRead: 0,
11919
12316
  cacheWrite: 0,
11920
12317
  },
@@ -12057,6 +12454,23 @@ export const MODELS = {
12057
12454
  contextWindow: 256000,
12058
12455
  maxTokens: 128000,
12059
12456
  },
12457
+ "sakana/fugu-ultra": {
12458
+ id: "sakana/fugu-ultra",
12459
+ name: "Sakana: Fugu Ultra",
12460
+ api: "openai-completions",
12461
+ provider: "openrouter",
12462
+ baseUrl: "https://openrouter.ai/api/v1",
12463
+ reasoning: true,
12464
+ input: ["text", "image"],
12465
+ cost: {
12466
+ input: 5,
12467
+ output: 30,
12468
+ cacheRead: 0.5,
12469
+ cacheWrite: 0,
12470
+ },
12471
+ contextWindow: 1000000,
12472
+ maxTokens: 128000,
12473
+ },
12060
12474
  "sao10k/l3.1-euryale-70b": {
12061
12475
  id: "sao10k/l3.1-euryale-70b",
12062
12476
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -12108,23 +12522,6 @@ export const MODELS = {
12108
12522
  contextWindow: 256000,
12109
12523
  maxTokens: 256000,
12110
12524
  },
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
12525
  "thedrummer/unslopnemo-12b": {
12129
12526
  id: "thedrummer/unslopnemo-12b",
12130
12527
  name: "TheDrummer: UnslopNemo 12B",
@@ -12219,13 +12616,13 @@ export const MODELS = {
12219
12616
  reasoning: true,
12220
12617
  input: ["text", "image"],
12221
12618
  cost: {
12222
- input: 0.14,
12619
+ input: 0.105,
12223
12620
  output: 0.28,
12224
- cacheRead: 0.0028,
12621
+ cacheRead: 0,
12225
12622
  cacheWrite: 0,
12226
12623
  },
12227
12624
  contextWindow: 1048576,
12228
- maxTokens: 131072,
12625
+ maxTokens: 4096,
12229
12626
  },
12230
12627
  "xiaomi/mimo-v2.5-pro": {
12231
12628
  id: "xiaomi/mimo-v2.5-pro",
@@ -12423,12 +12820,29 @@ export const MODELS = {
12423
12820
  reasoning: true,
12424
12821
  input: ["text"],
12425
12822
  cost: {
12426
- input: 1.2,
12427
- output: 4.1,
12428
- cacheRead: 0.19999999999999998,
12823
+ input: 0.95,
12824
+ output: 3,
12825
+ cacheRead: 0.18,
12429
12826
  cacheWrite: 0,
12430
12827
  },
12431
12828
  contextWindow: 1048576,
12829
+ maxTokens: 32768,
12830
+ },
12831
+ "z-ai/glm-5v-turbo": {
12832
+ id: "z-ai/glm-5v-turbo",
12833
+ name: "Z.ai: GLM 5V Turbo",
12834
+ api: "openai-completions",
12835
+ provider: "openrouter",
12836
+ baseUrl: "https://openrouter.ai/api/v1",
12837
+ reasoning: true,
12838
+ input: ["text", "image"],
12839
+ cost: {
12840
+ input: 1.2,
12841
+ output: 4,
12842
+ cacheRead: 0.24,
12843
+ cacheWrite: 0,
12844
+ },
12845
+ contextWindow: 202752,
12432
12846
  maxTokens: 131072,
12433
12847
  },
12434
12848
  "~anthropic/claude-fable-latest": {
@@ -12543,9 +12957,9 @@ export const MODELS = {
12543
12957
  reasoning: true,
12544
12958
  input: ["text", "image"],
12545
12959
  cost: {
12546
- input: 0.67,
12547
- output: 3.5,
12548
- cacheRead: 0.19999999999999998,
12960
+ input: 0.66,
12961
+ output: 3.41,
12962
+ cacheRead: 0.144,
12549
12963
  cacheWrite: 0,
12550
12964
  },
12551
12965
  contextWindow: 262144,
@@ -12943,6 +13357,25 @@ export const MODELS = {
12943
13357
  contextWindow: 202752,
12944
13358
  maxTokens: 131072,
12945
13359
  },
13360
+ "zai-org/GLM-5.2": {
13361
+ id: "zai-org/GLM-5.2",
13362
+ name: "GLM-5.2",
13363
+ api: "openai-completions",
13364
+ provider: "together",
13365
+ baseUrl: "https://api.together.ai/v1",
13366
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13367
+ reasoning: true,
13368
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13369
+ input: ["text"],
13370
+ cost: {
13371
+ input: 1.4,
13372
+ output: 4.4,
13373
+ cacheRead: 0.26,
13374
+ cacheWrite: 0,
13375
+ },
13376
+ contextWindow: 262144,
13377
+ maxTokens: 164000,
13378
+ },
12946
13379
  },
12947
13380
  "vercel-ai-gateway": {
12948
13381
  "alibaba/qwen-3-14b": {
@@ -13053,7 +13486,7 @@ export const MODELS = {
13053
13486
  api: "anthropic-messages",
13054
13487
  provider: "vercel-ai-gateway",
13055
13488
  baseUrl: "https://ai-gateway.vercel.sh",
13056
- reasoning: true,
13489
+ reasoning: false,
13057
13490
  input: ["text"],
13058
13491
  cost: {
13059
13492
  input: 1.5,
@@ -13070,7 +13503,7 @@ export const MODELS = {
13070
13503
  api: "anthropic-messages",
13071
13504
  provider: "vercel-ai-gateway",
13072
13505
  baseUrl: "https://ai-gateway.vercel.sh",
13073
- reasoning: true,
13506
+ reasoning: false,
13074
13507
  input: ["text"],
13075
13508
  cost: {
13076
13509
  input: 0.15,
@@ -13087,7 +13520,7 @@ export const MODELS = {
13087
13520
  api: "anthropic-messages",
13088
13521
  provider: "vercel-ai-gateway",
13089
13522
  baseUrl: "https://ai-gateway.vercel.sh",
13090
- reasoning: true,
13523
+ reasoning: false,
13091
13524
  input: ["text"],
13092
13525
  cost: {
13093
13526
  input: 0.5,
@@ -13200,6 +13633,40 @@ export const MODELS = {
13200
13633
  contextWindow: 131072,
13201
13634
  maxTokens: 32768,
13202
13635
  },
13636
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
13637
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
13638
+ name: "Qwen3 VL 235B A22B Instruct",
13639
+ api: "anthropic-messages",
13640
+ provider: "vercel-ai-gateway",
13641
+ baseUrl: "https://ai-gateway.vercel.sh",
13642
+ reasoning: false,
13643
+ input: ["text", "image"],
13644
+ cost: {
13645
+ input: 0.39999999999999997,
13646
+ output: 1.5999999999999999,
13647
+ cacheRead: 0,
13648
+ cacheWrite: 0,
13649
+ },
13650
+ contextWindow: 131072,
13651
+ maxTokens: 129024,
13652
+ },
13653
+ "alibaba/qwen3-vl-instruct": {
13654
+ id: "alibaba/qwen3-vl-instruct",
13655
+ name: "Qwen3 VL 235B A22B Instruct",
13656
+ api: "anthropic-messages",
13657
+ provider: "vercel-ai-gateway",
13658
+ baseUrl: "https://ai-gateway.vercel.sh",
13659
+ reasoning: false,
13660
+ input: ["text", "image"],
13661
+ cost: {
13662
+ input: 0.39999999999999997,
13663
+ output: 1.5999999999999999,
13664
+ cacheRead: 0,
13665
+ cacheWrite: 0,
13666
+ },
13667
+ contextWindow: 131072,
13668
+ maxTokens: 129024,
13669
+ },
13203
13670
  "alibaba/qwen3-vl-thinking": {
13204
13671
  id: "alibaba/qwen3-vl-thinking",
13205
13672
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13319,6 +13786,74 @@ export const MODELS = {
13319
13786
  contextWindow: 1000000,
13320
13787
  maxTokens: 64000,
13321
13788
  },
13789
+ "amazon/nova-2-lite": {
13790
+ id: "amazon/nova-2-lite",
13791
+ name: "Nova 2 Lite",
13792
+ api: "anthropic-messages",
13793
+ provider: "vercel-ai-gateway",
13794
+ baseUrl: "https://ai-gateway.vercel.sh",
13795
+ reasoning: true,
13796
+ input: ["text", "image"],
13797
+ cost: {
13798
+ input: 0.3,
13799
+ output: 2.5,
13800
+ cacheRead: 0.075,
13801
+ cacheWrite: 0,
13802
+ },
13803
+ contextWindow: 1000000,
13804
+ maxTokens: 1000000,
13805
+ },
13806
+ "amazon/nova-lite": {
13807
+ id: "amazon/nova-lite",
13808
+ name: "Nova Lite",
13809
+ api: "anthropic-messages",
13810
+ provider: "vercel-ai-gateway",
13811
+ baseUrl: "https://ai-gateway.vercel.sh",
13812
+ reasoning: false,
13813
+ input: ["text", "image"],
13814
+ cost: {
13815
+ input: 0.06,
13816
+ output: 0.24,
13817
+ cacheRead: 0,
13818
+ cacheWrite: 0,
13819
+ },
13820
+ contextWindow: 300000,
13821
+ maxTokens: 8192,
13822
+ },
13823
+ "amazon/nova-micro": {
13824
+ id: "amazon/nova-micro",
13825
+ name: "Nova Micro",
13826
+ api: "anthropic-messages",
13827
+ provider: "vercel-ai-gateway",
13828
+ baseUrl: "https://ai-gateway.vercel.sh",
13829
+ reasoning: false,
13830
+ input: ["text"],
13831
+ cost: {
13832
+ input: 0.035,
13833
+ output: 0.14,
13834
+ cacheRead: 0,
13835
+ cacheWrite: 0,
13836
+ },
13837
+ contextWindow: 128000,
13838
+ maxTokens: 8192,
13839
+ },
13840
+ "amazon/nova-pro": {
13841
+ id: "amazon/nova-pro",
13842
+ name: "Nova Pro",
13843
+ api: "anthropic-messages",
13844
+ provider: "vercel-ai-gateway",
13845
+ baseUrl: "https://ai-gateway.vercel.sh",
13846
+ reasoning: false,
13847
+ input: ["text", "image"],
13848
+ cost: {
13849
+ input: 0.7999999999999999,
13850
+ output: 3.1999999999999997,
13851
+ cacheRead: 0,
13852
+ cacheWrite: 0,
13853
+ },
13854
+ contextWindow: 300000,
13855
+ maxTokens: 8192,
13856
+ },
13322
13857
  "anthropic/claude-3-haiku": {
13323
13858
  id: "anthropic/claude-3-haiku",
13324
13859
  name: "Claude 3 Haiku",
@@ -13543,6 +14078,23 @@ export const MODELS = {
13543
14078
  contextWindow: 262100,
13544
14079
  maxTokens: 80000,
13545
14080
  },
14081
+ "arcee-ai/trinity-mini": {
14082
+ id: "arcee-ai/trinity-mini",
14083
+ name: "Trinity Mini",
14084
+ api: "anthropic-messages",
14085
+ provider: "vercel-ai-gateway",
14086
+ baseUrl: "https://ai-gateway.vercel.sh",
14087
+ reasoning: false,
14088
+ input: ["text"],
14089
+ cost: {
14090
+ input: 0.045,
14091
+ output: 0.15,
14092
+ cacheRead: 0,
14093
+ cacheWrite: 0,
14094
+ },
14095
+ contextWindow: 131072,
14096
+ maxTokens: 131072,
14097
+ },
13546
14098
  "bytedance/seed-1.6": {
13547
14099
  id: "bytedance/seed-1.6",
13548
14100
  name: "Seed 1.6",
@@ -13550,7 +14102,7 @@ export const MODELS = {
13550
14102
  provider: "vercel-ai-gateway",
13551
14103
  baseUrl: "https://ai-gateway.vercel.sh",
13552
14104
  reasoning: true,
13553
- input: ["text"],
14105
+ input: ["text", "image"],
13554
14106
  cost: {
13555
14107
  input: 0.25,
13556
14108
  output: 2,
@@ -13560,6 +14112,23 @@ export const MODELS = {
13560
14112
  contextWindow: 256000,
13561
14113
  maxTokens: 32000,
13562
14114
  },
14115
+ "bytedance/seed-1.8": {
14116
+ id: "bytedance/seed-1.8",
14117
+ name: "Bytedance Seed 1.8",
14118
+ api: "anthropic-messages",
14119
+ provider: "vercel-ai-gateway",
14120
+ baseUrl: "https://ai-gateway.vercel.sh",
14121
+ reasoning: true,
14122
+ input: ["text", "image"],
14123
+ cost: {
14124
+ input: 0.25,
14125
+ output: 2,
14126
+ cacheRead: 0.049999999999999996,
14127
+ cacheWrite: 0,
14128
+ },
14129
+ contextWindow: 256000,
14130
+ maxTokens: 64000,
14131
+ },
13563
14132
  "cohere/command-a": {
13564
14133
  id: "cohere/command-a",
13565
14134
  name: "Command A",
@@ -13651,8 +14220,8 @@ export const MODELS = {
13651
14220
  api: "anthropic-messages",
13652
14221
  provider: "vercel-ai-gateway",
13653
14222
  baseUrl: "https://ai-gateway.vercel.sh",
13654
- reasoning: true,
13655
- input: ["text", "image"],
14223
+ reasoning: false,
14224
+ input: ["text"],
13656
14225
  cost: {
13657
14226
  input: 0.28,
13658
14227
  output: 0.42,
@@ -13669,7 +14238,7 @@ export const MODELS = {
13669
14238
  provider: "vercel-ai-gateway",
13670
14239
  baseUrl: "https://ai-gateway.vercel.sh",
13671
14240
  reasoning: true,
13672
- input: ["text", "image"],
14241
+ input: ["text"],
13673
14242
  cost: {
13674
14243
  input: 0.62,
13675
14244
  output: 1.85,
@@ -13686,7 +14255,7 @@ export const MODELS = {
13686
14255
  provider: "vercel-ai-gateway",
13687
14256
  baseUrl: "https://ai-gateway.vercel.sh",
13688
14257
  reasoning: true,
13689
- input: ["text", "image"],
14258
+ input: ["text"],
13690
14259
  cost: {
13691
14260
  input: 0.14,
13692
14261
  output: 0.28,
@@ -13889,7 +14458,7 @@ export const MODELS = {
13889
14458
  api: "anthropic-messages",
13890
14459
  provider: "vercel-ai-gateway",
13891
14460
  baseUrl: "https://ai-gateway.vercel.sh",
13892
- reasoning: false,
14461
+ reasoning: true,
13893
14462
  input: ["text", "image"],
13894
14463
  cost: {
13895
14464
  input: 0.14,
@@ -13934,6 +14503,40 @@ export const MODELS = {
13934
14503
  contextWindow: 32000,
13935
14504
  maxTokens: 16384,
13936
14505
  },
14506
+ "interfaze/interfaze-beta": {
14507
+ id: "interfaze/interfaze-beta",
14508
+ name: "Interfaze Beta",
14509
+ api: "anthropic-messages",
14510
+ provider: "vercel-ai-gateway",
14511
+ baseUrl: "https://ai-gateway.vercel.sh",
14512
+ reasoning: true,
14513
+ input: ["text", "image"],
14514
+ cost: {
14515
+ input: 1.5,
14516
+ output: 3.5,
14517
+ cacheRead: 0,
14518
+ cacheWrite: 0,
14519
+ },
14520
+ contextWindow: 1000000,
14521
+ maxTokens: 32000,
14522
+ },
14523
+ "kwaipilot/kat-coder-pro-v1": {
14524
+ id: "kwaipilot/kat-coder-pro-v1",
14525
+ name: "KAT-Coder-Pro V1",
14526
+ api: "anthropic-messages",
14527
+ provider: "vercel-ai-gateway",
14528
+ baseUrl: "https://ai-gateway.vercel.sh",
14529
+ reasoning: false,
14530
+ input: ["text"],
14531
+ cost: {
14532
+ input: 0.3,
14533
+ output: 1.2,
14534
+ cacheRead: 0.06,
14535
+ cacheWrite: 0,
14536
+ },
14537
+ contextWindow: 256000,
14538
+ maxTokens: 32000,
14539
+ },
13937
14540
  "kwaipilot/kat-coder-pro-v2": {
13938
14541
  id: "kwaipilot/kat-coder-pro-v2",
13939
14542
  name: "Kat Coder Pro V2",
@@ -13968,6 +14571,23 @@ export const MODELS = {
13968
14571
  contextWindow: 128000,
13969
14572
  maxTokens: 100000,
13970
14573
  },
14574
+ "meituan/longcat-flash-thinking-2601": {
14575
+ id: "meituan/longcat-flash-thinking-2601",
14576
+ name: "LongCat Flash Thinking 2601",
14577
+ api: "anthropic-messages",
14578
+ provider: "vercel-ai-gateway",
14579
+ baseUrl: "https://ai-gateway.vercel.sh",
14580
+ reasoning: true,
14581
+ input: ["text"],
14582
+ cost: {
14583
+ input: 0,
14584
+ output: 0,
14585
+ cacheRead: 0,
14586
+ cacheWrite: 0,
14587
+ },
14588
+ contextWindow: 32768,
14589
+ maxTokens: 32768,
14590
+ },
13971
14591
  "meta/llama-3.1-70b": {
13972
14592
  id: "meta/llama-3.1-70b",
13973
14593
  name: "Llama 3.1 70B Instruct",
@@ -14281,7 +14901,7 @@ export const MODELS = {
14281
14901
  provider: "vercel-ai-gateway",
14282
14902
  baseUrl: "https://ai-gateway.vercel.sh",
14283
14903
  reasoning: false,
14284
- input: ["text"],
14904
+ input: ["text", "image"],
14285
14905
  cost: {
14286
14906
  input: 0.09999999999999999,
14287
14907
  output: 0.3,
@@ -14291,6 +14911,57 @@ export const MODELS = {
14291
14911
  contextWindow: 256000,
14292
14912
  maxTokens: 256000,
14293
14913
  },
14914
+ "mistral/magistral-medium": {
14915
+ id: "mistral/magistral-medium",
14916
+ name: "Magistral Medium 2509",
14917
+ api: "anthropic-messages",
14918
+ provider: "vercel-ai-gateway",
14919
+ baseUrl: "https://ai-gateway.vercel.sh",
14920
+ reasoning: true,
14921
+ input: ["text", "image"],
14922
+ cost: {
14923
+ input: 2,
14924
+ output: 5,
14925
+ cacheRead: 0,
14926
+ cacheWrite: 0,
14927
+ },
14928
+ contextWindow: 128000,
14929
+ maxTokens: 64000,
14930
+ },
14931
+ "mistral/magistral-small": {
14932
+ id: "mistral/magistral-small",
14933
+ name: "Magistral Small 2509",
14934
+ api: "anthropic-messages",
14935
+ provider: "vercel-ai-gateway",
14936
+ baseUrl: "https://ai-gateway.vercel.sh",
14937
+ reasoning: true,
14938
+ input: ["text", "image"],
14939
+ cost: {
14940
+ input: 0.5,
14941
+ output: 1.5,
14942
+ cacheRead: 0,
14943
+ cacheWrite: 0,
14944
+ },
14945
+ contextWindow: 128000,
14946
+ maxTokens: 64000,
14947
+ },
14948
+ "mistral/ministral-14b": {
14949
+ id: "mistral/ministral-14b",
14950
+ name: "Ministral 14B",
14951
+ api: "anthropic-messages",
14952
+ provider: "vercel-ai-gateway",
14953
+ baseUrl: "https://ai-gateway.vercel.sh",
14954
+ reasoning: false,
14955
+ input: ["text", "image"],
14956
+ cost: {
14957
+ input: 0.19999999999999998,
14958
+ output: 0.19999999999999998,
14959
+ cacheRead: 0,
14960
+ cacheWrite: 0,
14961
+ },
14962
+ contextWindow: 256000,
14963
+ maxTokens: 256000,
14964
+ },
14294
14965
  "mistral/ministral-3b": {
14295
14966
  id: "mistral/ministral-3b",
14296
14967
  name: "Ministral 3B",
@@ -14325,6 +14996,23 @@ export const MODELS = {
14325
14996
  contextWindow: 128000,
14326
14997
  maxTokens: 4000,
14327
14998
  },
14999
+ "mistral/mistral-large-3": {
15000
+ id: "mistral/mistral-large-3",
15001
+ name: "Mistral Large 3",
15002
+ api: "anthropic-messages",
15003
+ provider: "vercel-ai-gateway",
15004
+ baseUrl: "https://ai-gateway.vercel.sh",
15005
+ reasoning: false,
15006
+ input: ["text", "image"],
15007
+ cost: {
15008
+ input: 0.5,
15009
+ output: 1.5,
15010
+ cacheRead: 0,
15011
+ cacheWrite: 0,
15012
+ },
15013
+ contextWindow: 256000,
15014
+ maxTokens: 256000,
15015
+ },
14328
15016
  "mistral/mistral-medium": {
14329
15017
  id: "mistral/mistral-medium",
14330
15018
  name: "Mistral Medium 3.1",
@@ -14349,7 +15037,7 @@ export const MODELS = {
14349
15037
  provider: "vercel-ai-gateway",
14350
15038
  baseUrl: "https://ai-gateway.vercel.sh",
14351
15039
  reasoning: true,
14352
- input: ["text"],
15040
+ input: ["text", "image"],
14353
15041
  cost: {
14354
15042
  input: 1.5,
14355
15043
  output: 7.5,
@@ -14529,6 +15217,23 @@ export const MODELS = {
14529
15217
  contextWindow: 262144,
14530
15218
  maxTokens: 32768,
14531
15219
  },
15220
+ "nvidia/nemotron-3-nano-30b-a3b": {
15221
+ id: "nvidia/nemotron-3-nano-30b-a3b",
15222
+ name: "Nemotron 3 Nano 30B A3B",
15223
+ api: "anthropic-messages",
15224
+ provider: "vercel-ai-gateway",
15225
+ baseUrl: "https://ai-gateway.vercel.sh",
15226
+ reasoning: true,
15227
+ input: ["text"],
15228
+ cost: {
15229
+ input: 0.049999999999999996,
15230
+ output: 0.24,
15231
+ cacheRead: 0,
15232
+ cacheWrite: 0,
15233
+ },
15234
+ contextWindow: 262144,
15235
+ maxTokens: 262144,
15236
+ },
14532
15237
  "nvidia/nemotron-3-super-120b-a12b": {
14533
15238
  id: "nvidia/nemotron-3-super-120b-a12b",
14534
15239
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -14597,6 +15302,23 @@ export const MODELS = {
14597
15302
  contextWindow: 131072,
14598
15303
  maxTokens: 131072,
14599
15304
  },
15305
+ "openai/gpt-3.5-turbo": {
15306
+ id: "openai/gpt-3.5-turbo",
15307
+ name: "GPT-3.5 Turbo",
15308
+ api: "anthropic-messages",
15309
+ provider: "vercel-ai-gateway",
15310
+ baseUrl: "https://ai-gateway.vercel.sh",
15311
+ reasoning: false,
15312
+ input: ["text"],
15313
+ cost: {
15314
+ input: 0.5,
15315
+ output: 1.5,
15316
+ cacheRead: 0,
15317
+ cacheWrite: 0,
15318
+ },
15319
+ contextWindow: 16385,
15320
+ maxTokens: 4096,
15321
+ },
14600
15322
  "openai/gpt-4.1": {
14601
15323
  id: "openai/gpt-4.1",
14602
15324
  name: "GPT-4.1",
@@ -14705,7 +15427,7 @@ export const MODELS = {
14705
15427
  api: "anthropic-messages",
14706
15428
  provider: "vercel-ai-gateway",
14707
15429
  baseUrl: "https://ai-gateway.vercel.sh",
14708
- reasoning: true,
15430
+ reasoning: false,
14709
15431
  input: ["text", "image"],
14710
15432
  cost: {
14711
15433
  input: 1.25,
@@ -14841,7 +15563,7 @@ export const MODELS = {
14841
15563
  api: "anthropic-messages",
14842
15564
  provider: "vercel-ai-gateway",
14843
15565
  baseUrl: "https://ai-gateway.vercel.sh",
14844
- reasoning: true,
15566
+ reasoning: false,
14845
15567
  input: ["text", "image"],
14846
15568
  cost: {
14847
15569
  input: 1.25,
@@ -14893,7 +15615,7 @@ export const MODELS = {
14893
15615
  api: "anthropic-messages",
14894
15616
  provider: "vercel-ai-gateway",
14895
15617
  baseUrl: "https://ai-gateway.vercel.sh",
14896
- reasoning: true,
15618
+ reasoning: false,
14897
15619
  thinkingLevelMap: { "xhigh": "xhigh" },
14898
15620
  input: ["text", "image"],
14899
15621
  cost: {
@@ -14947,7 +15669,7 @@ export const MODELS = {
14947
15669
  api: "anthropic-messages",
14948
15670
  provider: "vercel-ai-gateway",
14949
15671
  baseUrl: "https://ai-gateway.vercel.sh",
14950
- reasoning: true,
15672
+ reasoning: false,
14951
15673
  thinkingLevelMap: { "xhigh": "xhigh" },
14952
15674
  input: ["text", "image"],
14953
15675
  cost: {
@@ -15094,13 +15816,13 @@ export const MODELS = {
15094
15816
  reasoning: true,
15095
15817
  input: ["text"],
15096
15818
  cost: {
15097
- input: 0.35,
15098
- output: 0.75,
15099
- cacheRead: 0.25,
15819
+ input: 0.09999999999999999,
15820
+ output: 0.5,
15821
+ cacheRead: 0,
15100
15822
  cacheWrite: 0,
15101
15823
  },
15102
15824
  contextWindow: 131072,
15103
- maxTokens: 131000,
15825
+ maxTokens: 131072,
15104
15826
  },
15105
15827
  "openai/gpt-oss-20b": {
15106
15828
  id: "openai/gpt-oss-20b",
@@ -15238,39 +15960,22 @@ export const MODELS = {
15238
15960
  contextWindow: 200000,
15239
15961
  maxTokens: 100000,
15240
15962
  },
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",
15963
+ "sakana/fugu-ultra": {
15964
+ id: "sakana/fugu-ultra",
15965
+ name: "Fugu Ultra",
15261
15966
  api: "anthropic-messages",
15262
15967
  provider: "vercel-ai-gateway",
15263
15968
  baseUrl: "https://ai-gateway.vercel.sh",
15264
- reasoning: false,
15969
+ reasoning: true,
15265
15970
  input: ["text", "image"],
15266
15971
  cost: {
15267
- input: 0,
15268
- output: 0,
15269
- cacheRead: 0,
15972
+ input: 5,
15973
+ output: 30,
15974
+ cacheRead: 0.5,
15270
15975
  cacheWrite: 0,
15271
15976
  },
15272
- contextWindow: 200000,
15273
- maxTokens: 8000,
15977
+ contextWindow: 1000000,
15978
+ maxTokens: 1000000,
15274
15979
  },
15275
15980
  "stepfun/step-3.5-flash": {
15276
15981
  id: "stepfun/step-3.5-flash",
@@ -15283,8 +15988,8 @@ export const MODELS = {
15283
15988
  cost: {
15284
15989
  input: 0.09,
15285
15990
  output: 0.3,
15286
- cacheRead: 0,
15287
- cacheWrite: 0.02,
15991
+ cacheRead: 0.02,
15992
+ cacheWrite: 0,
15288
15993
  },
15289
15994
  contextWindow: 262114,
15290
15995
  maxTokens: 262114,
@@ -15500,7 +16205,7 @@ export const MODELS = {
15500
16205
  provider: "vercel-ai-gateway",
15501
16206
  baseUrl: "https://ai-gateway.vercel.sh",
15502
16207
  reasoning: true,
15503
- input: ["text", "image"],
16208
+ input: ["text"],
15504
16209
  cost: {
15505
16210
  input: 0.435,
15506
16211
  output: 0.87,
@@ -15621,13 +16326,13 @@ export const MODELS = {
15621
16326
  reasoning: true,
15622
16327
  input: ["text"],
15623
16328
  cost: {
15624
- input: 2.25,
15625
- output: 2.75,
15626
- cacheRead: 2.25,
16329
+ input: 0.6,
16330
+ output: 2.2,
16331
+ cacheRead: 0.12,
15627
16332
  cacheWrite: 0,
15628
16333
  },
15629
- contextWindow: 131000,
15630
- maxTokens: 40000,
16334
+ contextWindow: 200000,
16335
+ maxTokens: 120000,
15631
16336
  },
15632
16337
  "zai/glm-4.7-flash": {
15633
16338
  id: "zai/glm-4.7-flash",
@@ -15672,8 +16377,8 @@ export const MODELS = {
15672
16377
  reasoning: true,
15673
16378
  input: ["text"],
15674
16379
  cost: {
15675
- input: 1,
15676
- output: 3.1999999999999997,
16380
+ input: 0.95,
16381
+ output: 3.15,
15677
16382
  cacheRead: 0.19999999999999998,
15678
16383
  cacheWrite: 0,
15679
16384
  },
@@ -15704,15 +16409,15 @@ export const MODELS = {
15704
16409
  provider: "vercel-ai-gateway",
15705
16410
  baseUrl: "https://ai-gateway.vercel.sh",
15706
16411
  reasoning: true,
15707
- input: ["text", "image"],
16412
+ input: ["text"],
15708
16413
  cost: {
15709
- input: 1.4,
15710
- output: 4.4,
16414
+ input: 1.3,
16415
+ output: 4.300000000000001,
15711
16416
  cacheRead: 0.26,
15712
16417
  cacheWrite: 0,
15713
16418
  },
15714
- contextWindow: 202800,
15715
- maxTokens: 64000,
16419
+ contextWindow: 202000,
16420
+ maxTokens: 202000,
15716
16421
  },
15717
16422
  "zai/glm-5.2": {
15718
16423
  id: "zai/glm-5.2",
@@ -15731,6 +16436,23 @@ export const MODELS = {
15731
16436
  contextWindow: 1000000,
15732
16437
  maxTokens: 128000,
15733
16438
  },
16439
+ "zai/glm-5.2-fast": {
16440
+ id: "zai/glm-5.2-fast",
16441
+ name: "GLM 5.2 Fast",
16442
+ api: "anthropic-messages",
16443
+ provider: "vercel-ai-gateway",
16444
+ baseUrl: "https://ai-gateway.vercel.sh",
16445
+ reasoning: true,
16446
+ input: ["text"],
16447
+ cost: {
16448
+ input: 3,
16449
+ output: 10.25,
16450
+ cacheRead: 0.5,
16451
+ cacheWrite: 0,
16452
+ },
16453
+ contextWindow: 1000000,
16454
+ maxTokens: 128000,
16455
+ },
15734
16456
  "zai/glm-5v-turbo": {
15735
16457
  id: "zai/glm-5v-turbo",
15736
16458
  name: "GLM 5V Turbo",