@dreb/ai 2.31.0 → 2.32.0

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.
@@ -3491,40 +3491,6 @@ export const MODELS = {
3491
3491
  contextWindow: 262144,
3492
3492
  maxTokens: 32768,
3493
3493
  },
3494
- "gemma-4-E2B-it": {
3495
- id: "gemma-4-E2B-it",
3496
- name: "Gemma 4 E2B IT",
3497
- api: "google-generative-ai",
3498
- provider: "google",
3499
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3500
- reasoning: true,
3501
- input: ["text", "image"],
3502
- cost: {
3503
- input: 0,
3504
- output: 0,
3505
- cacheRead: 0,
3506
- cacheWrite: 0,
3507
- },
3508
- contextWindow: 131072,
3509
- maxTokens: 8192,
3510
- },
3511
- "gemma-4-E4B-it": {
3512
- id: "gemma-4-E4B-it",
3513
- name: "Gemma 4 E4B IT",
3514
- api: "google-generative-ai",
3515
- provider: "google",
3516
- baseUrl: "https://generativelanguage.googleapis.com/v1beta",
3517
- reasoning: true,
3518
- input: ["text", "image"],
3519
- cost: {
3520
- input: 0,
3521
- output: 0,
3522
- cacheRead: 0,
3523
- cacheWrite: 0,
3524
- },
3525
- contextWindow: 131072,
3526
- maxTokens: 8192,
3527
- },
3528
3494
  },
3529
3495
  "google-antigravity": {
3530
3496
  "claude-opus-4-5-thinking": {
@@ -4113,6 +4079,24 @@ export const MODELS = {
4113
4079
  },
4114
4080
  },
4115
4081
  "huggingface": {
4082
+ "MiniMaxAI/MiniMax-M2": {
4083
+ id: "MiniMaxAI/MiniMax-M2",
4084
+ name: "MiniMax-M2",
4085
+ api: "openai-completions",
4086
+ provider: "huggingface",
4087
+ baseUrl: "https://router.huggingface.co/v1",
4088
+ compat: { "supportsDeveloperRole": false },
4089
+ reasoning: true,
4090
+ input: ["text"],
4091
+ cost: {
4092
+ input: 0.3,
4093
+ output: 1.2,
4094
+ cacheRead: 0,
4095
+ cacheWrite: 0,
4096
+ },
4097
+ contextWindow: 204800,
4098
+ maxTokens: 128000,
4099
+ },
4116
4100
  "MiniMaxAI/MiniMax-M2.1": {
4117
4101
  id: "MiniMaxAI/MiniMax-M2.1",
4118
4102
  name: "MiniMax-M2.1",
@@ -4167,6 +4151,42 @@ export const MODELS = {
4167
4151
  contextWindow: 204800,
4168
4152
  maxTokens: 131072,
4169
4153
  },
4154
+ "MiniMaxAI/MiniMax-M3": {
4155
+ id: "MiniMaxAI/MiniMax-M3",
4156
+ name: "MiniMax-M3",
4157
+ api: "openai-completions",
4158
+ provider: "huggingface",
4159
+ baseUrl: "https://router.huggingface.co/v1",
4160
+ compat: { "supportsDeveloperRole": false },
4161
+ reasoning: true,
4162
+ input: ["text", "image"],
4163
+ cost: {
4164
+ input: 0.3,
4165
+ output: 1.2,
4166
+ cacheRead: 0,
4167
+ cacheWrite: 0,
4168
+ },
4169
+ contextWindow: 524288,
4170
+ maxTokens: 128000,
4171
+ },
4172
+ "Qwen/Qwen3-235B-A22B": {
4173
+ id: "Qwen/Qwen3-235B-A22B",
4174
+ name: "Qwen3 235B-A22B",
4175
+ api: "openai-completions",
4176
+ provider: "huggingface",
4177
+ baseUrl: "https://router.huggingface.co/v1",
4178
+ compat: { "supportsDeveloperRole": false },
4179
+ reasoning: true,
4180
+ input: ["text"],
4181
+ cost: {
4182
+ input: 0.2,
4183
+ output: 0.8,
4184
+ cacheRead: 0,
4185
+ cacheWrite: 0,
4186
+ },
4187
+ contextWindow: 40960,
4188
+ maxTokens: 16384,
4189
+ },
4170
4190
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4171
4191
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4172
4192
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4185,6 +4205,42 @@ export const MODELS = {
4185
4205
  contextWindow: 262144,
4186
4206
  maxTokens: 131072,
4187
4207
  },
4208
+ "Qwen/Qwen3-32B": {
4209
+ id: "Qwen/Qwen3-32B",
4210
+ name: "Qwen3 32B",
4211
+ api: "openai-completions",
4212
+ provider: "huggingface",
4213
+ baseUrl: "https://router.huggingface.co/v1",
4214
+ compat: { "supportsDeveloperRole": false },
4215
+ reasoning: true,
4216
+ input: ["text"],
4217
+ cost: {
4218
+ input: 0.29,
4219
+ output: 0.59,
4220
+ cacheRead: 0,
4221
+ cacheWrite: 0,
4222
+ },
4223
+ contextWindow: 131072,
4224
+ maxTokens: 16384,
4225
+ },
4226
+ "Qwen/Qwen3-Coder-30B-A3B-Instruct": {
4227
+ id: "Qwen/Qwen3-Coder-30B-A3B-Instruct",
4228
+ name: "Qwen3-Coder 30B-A3B Instruct",
4229
+ api: "openai-completions",
4230
+ provider: "huggingface",
4231
+ baseUrl: "https://router.huggingface.co/v1",
4232
+ compat: { "supportsDeveloperRole": false },
4233
+ reasoning: false,
4234
+ input: ["text"],
4235
+ cost: {
4236
+ input: 0.07,
4237
+ output: 0.26,
4238
+ cacheRead: 0,
4239
+ cacheWrite: 0,
4240
+ },
4241
+ contextWindow: 262144,
4242
+ maxTokens: 65536,
4243
+ },
4188
4244
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
4189
4245
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
4190
4246
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -4257,6 +4313,60 @@ export const MODELS = {
4257
4313
  contextWindow: 262144,
4258
4314
  maxTokens: 131072,
4259
4315
  },
4316
+ "Qwen/Qwen3.5-122B-A10B": {
4317
+ id: "Qwen/Qwen3.5-122B-A10B",
4318
+ name: "Qwen3.5 122B-A10B",
4319
+ api: "openai-completions",
4320
+ provider: "huggingface",
4321
+ baseUrl: "https://router.huggingface.co/v1",
4322
+ compat: { "supportsDeveloperRole": false },
4323
+ reasoning: true,
4324
+ input: ["text", "image"],
4325
+ cost: {
4326
+ input: 0.4,
4327
+ output: 3.2,
4328
+ cacheRead: 0,
4329
+ cacheWrite: 0,
4330
+ },
4331
+ contextWindow: 262144,
4332
+ maxTokens: 65536,
4333
+ },
4334
+ "Qwen/Qwen3.5-27B": {
4335
+ id: "Qwen/Qwen3.5-27B",
4336
+ name: "Qwen3.5 27B",
4337
+ api: "openai-completions",
4338
+ provider: "huggingface",
4339
+ baseUrl: "https://router.huggingface.co/v1",
4340
+ compat: { "supportsDeveloperRole": false },
4341
+ reasoning: true,
4342
+ input: ["text", "image"],
4343
+ cost: {
4344
+ input: 0.3,
4345
+ output: 2.4,
4346
+ cacheRead: 0,
4347
+ cacheWrite: 0,
4348
+ },
4349
+ contextWindow: 262144,
4350
+ maxTokens: 65536,
4351
+ },
4352
+ "Qwen/Qwen3.5-35B-A3B": {
4353
+ id: "Qwen/Qwen3.5-35B-A3B",
4354
+ name: "Qwen3.5 35B-A3B",
4355
+ api: "openai-completions",
4356
+ provider: "huggingface",
4357
+ baseUrl: "https://router.huggingface.co/v1",
4358
+ compat: { "supportsDeveloperRole": false },
4359
+ reasoning: true,
4360
+ input: ["text", "image"],
4361
+ cost: {
4362
+ input: 0.25,
4363
+ output: 2,
4364
+ cacheRead: 0,
4365
+ cacheWrite: 0,
4366
+ },
4367
+ contextWindow: 262144,
4368
+ maxTokens: 65536,
4369
+ },
4260
4370
  "Qwen/Qwen3.5-397B-A17B": {
4261
4371
  id: "Qwen/Qwen3.5-397B-A17B",
4262
4372
  name: "Qwen3.5-397B-A17B",
@@ -4275,6 +4385,42 @@ export const MODELS = {
4275
4385
  contextWindow: 262144,
4276
4386
  maxTokens: 32768,
4277
4387
  },
4388
+ "Qwen/Qwen3.5-9B": {
4389
+ id: "Qwen/Qwen3.5-9B",
4390
+ name: "Qwen3.5 9B",
4391
+ api: "openai-completions",
4392
+ provider: "huggingface",
4393
+ baseUrl: "https://router.huggingface.co/v1",
4394
+ compat: { "supportsDeveloperRole": false },
4395
+ reasoning: true,
4396
+ input: ["text", "image"],
4397
+ cost: {
4398
+ input: 0.17,
4399
+ output: 0.25,
4400
+ cacheRead: 0,
4401
+ cacheWrite: 0,
4402
+ },
4403
+ contextWindow: 262144,
4404
+ maxTokens: 65536,
4405
+ },
4406
+ "Qwen/Qwen3.6-35B-A3B": {
4407
+ id: "Qwen/Qwen3.6-35B-A3B",
4408
+ name: "Qwen3.6 35B-A3B",
4409
+ api: "openai-completions",
4410
+ provider: "huggingface",
4411
+ baseUrl: "https://router.huggingface.co/v1",
4412
+ compat: { "supportsDeveloperRole": false },
4413
+ reasoning: true,
4414
+ input: ["text", "image"],
4415
+ cost: {
4416
+ input: 0.15,
4417
+ output: 0.95,
4418
+ cacheRead: 0,
4419
+ cacheWrite: 0,
4420
+ },
4421
+ contextWindow: 262144,
4422
+ maxTokens: 65536,
4423
+ },
4278
4424
  "XiaomiMiMo/MiMo-V2-Flash": {
4279
4425
  id: "XiaomiMiMo/MiMo-V2-Flash",
4280
4426
  name: "MiMo-V2-Flash",
@@ -4293,6 +4439,24 @@ export const MODELS = {
4293
4439
  contextWindow: 262144,
4294
4440
  maxTokens: 4096,
4295
4441
  },
4442
+ "deepseek-ai/DeepSeek-R1": {
4443
+ id: "deepseek-ai/DeepSeek-R1",
4444
+ name: "DeepSeek-R1",
4445
+ api: "openai-completions",
4446
+ provider: "huggingface",
4447
+ baseUrl: "https://router.huggingface.co/v1",
4448
+ compat: { "supportsDeveloperRole": false },
4449
+ reasoning: true,
4450
+ input: ["text"],
4451
+ cost: {
4452
+ input: 0.7,
4453
+ output: 2.5,
4454
+ cacheRead: 0,
4455
+ cacheWrite: 0,
4456
+ },
4457
+ contextWindow: 64000,
4458
+ maxTokens: 32768,
4459
+ },
4296
4460
  "deepseek-ai/DeepSeek-R1-0528": {
4297
4461
  id: "deepseek-ai/DeepSeek-R1-0528",
4298
4462
  name: "DeepSeek-R1-0528",
@@ -4329,6 +4493,24 @@ export const MODELS = {
4329
4493
  contextWindow: 163840,
4330
4494
  maxTokens: 65536,
4331
4495
  },
4496
+ "deepseek-ai/DeepSeek-V4-Flash": {
4497
+ id: "deepseek-ai/DeepSeek-V4-Flash",
4498
+ name: "DeepSeek V4 Flash",
4499
+ api: "openai-completions",
4500
+ provider: "huggingface",
4501
+ baseUrl: "https://router.huggingface.co/v1",
4502
+ compat: { "supportsDeveloperRole": false },
4503
+ reasoning: true,
4504
+ input: ["text"],
4505
+ cost: {
4506
+ input: 0.14,
4507
+ output: 0.28,
4508
+ cacheRead: 0,
4509
+ cacheWrite: 0,
4510
+ },
4511
+ contextWindow: 1048576,
4512
+ maxTokens: 384000,
4513
+ },
4332
4514
  "deepseek-ai/DeepSeek-V4-Pro": {
4333
4515
  id: "deepseek-ai/DeepSeek-V4-Pro",
4334
4516
  name: "DeepSeek V4 Pro",
@@ -4347,6 +4529,60 @@ export const MODELS = {
4347
4529
  contextWindow: 1048576,
4348
4530
  maxTokens: 393216,
4349
4531
  },
4532
+ "google/gemma-4-26B-A4B-it": {
4533
+ id: "google/gemma-4-26B-A4B-it",
4534
+ name: "Gemma 4 26B A4B IT",
4535
+ api: "openai-completions",
4536
+ provider: "huggingface",
4537
+ baseUrl: "https://router.huggingface.co/v1",
4538
+ compat: { "supportsDeveloperRole": false },
4539
+ reasoning: true,
4540
+ input: ["text", "image"],
4541
+ cost: {
4542
+ input: 0.13,
4543
+ output: 0.4,
4544
+ cacheRead: 0,
4545
+ cacheWrite: 0,
4546
+ },
4547
+ contextWindow: 262144,
4548
+ maxTokens: 32768,
4549
+ },
4550
+ "google/gemma-4-31B-it": {
4551
+ id: "google/gemma-4-31B-it",
4552
+ name: "Gemma 4 31B IT",
4553
+ api: "openai-completions",
4554
+ provider: "huggingface",
4555
+ baseUrl: "https://router.huggingface.co/v1",
4556
+ compat: { "supportsDeveloperRole": false },
4557
+ reasoning: true,
4558
+ input: ["text", "image"],
4559
+ cost: {
4560
+ input: 0.14,
4561
+ output: 0.4,
4562
+ cacheRead: 0,
4563
+ cacheWrite: 0,
4564
+ },
4565
+ contextWindow: 262144,
4566
+ maxTokens: 32768,
4567
+ },
4568
+ "meta-llama/Llama-3.3-70B-Instruct": {
4569
+ id: "meta-llama/Llama-3.3-70B-Instruct",
4570
+ name: "Llama-3.3-70B-Instruct",
4571
+ api: "openai-completions",
4572
+ provider: "huggingface",
4573
+ baseUrl: "https://router.huggingface.co/v1",
4574
+ compat: { "supportsDeveloperRole": false },
4575
+ reasoning: false,
4576
+ input: ["text"],
4577
+ cost: {
4578
+ input: 0.59,
4579
+ output: 0.79,
4580
+ cacheRead: 0,
4581
+ cacheWrite: 0,
4582
+ },
4583
+ contextWindow: 131072,
4584
+ maxTokens: 4096,
4585
+ },
4350
4586
  "moonshotai/Kimi-K2-Instruct": {
4351
4587
  id: "moonshotai/Kimi-K2-Instruct",
4352
4588
  name: "Kimi-K2-Instruct",
@@ -4437,6 +4673,114 @@ export const MODELS = {
4437
4673
  contextWindow: 262144,
4438
4674
  maxTokens: 262144,
4439
4675
  },
4676
+ "moonshotai/Kimi-K2.7-Code": {
4677
+ id: "moonshotai/Kimi-K2.7-Code",
4678
+ name: "Kimi K2.7 Code",
4679
+ api: "openai-completions",
4680
+ provider: "huggingface",
4681
+ baseUrl: "https://router.huggingface.co/v1",
4682
+ compat: { "supportsDeveloperRole": false },
4683
+ reasoning: true,
4684
+ input: ["text", "image"],
4685
+ cost: {
4686
+ input: 0.95,
4687
+ output: 4,
4688
+ cacheRead: 0,
4689
+ cacheWrite: 0,
4690
+ },
4691
+ contextWindow: 262144,
4692
+ maxTokens: 262144,
4693
+ },
4694
+ "stepfun-ai/Step-3.5-Flash": {
4695
+ id: "stepfun-ai/Step-3.5-Flash",
4696
+ name: "Step 3.5 Flash",
4697
+ api: "openai-completions",
4698
+ provider: "huggingface",
4699
+ baseUrl: "https://router.huggingface.co/v1",
4700
+ compat: { "supportsDeveloperRole": false },
4701
+ reasoning: true,
4702
+ input: ["text"],
4703
+ cost: {
4704
+ input: 0.1,
4705
+ output: 0.3,
4706
+ cacheRead: 0,
4707
+ cacheWrite: 0,
4708
+ },
4709
+ contextWindow: 262144,
4710
+ maxTokens: 256000,
4711
+ },
4712
+ "zai-org/GLM-4.5": {
4713
+ id: "zai-org/GLM-4.5",
4714
+ name: "GLM-4.5",
4715
+ api: "openai-completions",
4716
+ provider: "huggingface",
4717
+ baseUrl: "https://router.huggingface.co/v1",
4718
+ compat: { "supportsDeveloperRole": false },
4719
+ reasoning: true,
4720
+ input: ["text"],
4721
+ cost: {
4722
+ input: 0.6,
4723
+ output: 2.2,
4724
+ cacheRead: 0,
4725
+ cacheWrite: 0,
4726
+ },
4727
+ contextWindow: 131072,
4728
+ maxTokens: 98304,
4729
+ },
4730
+ "zai-org/GLM-4.5-Air": {
4731
+ id: "zai-org/GLM-4.5-Air",
4732
+ name: "GLM-4.5-Air",
4733
+ api: "openai-completions",
4734
+ provider: "huggingface",
4735
+ baseUrl: "https://router.huggingface.co/v1",
4736
+ compat: { "supportsDeveloperRole": false },
4737
+ reasoning: true,
4738
+ input: ["text"],
4739
+ cost: {
4740
+ input: 0.13,
4741
+ output: 0.85,
4742
+ cacheRead: 0,
4743
+ cacheWrite: 0,
4744
+ },
4745
+ contextWindow: 131072,
4746
+ maxTokens: 98304,
4747
+ },
4748
+ "zai-org/GLM-4.5V": {
4749
+ id: "zai-org/GLM-4.5V",
4750
+ name: "GLM-4.5V",
4751
+ api: "openai-completions",
4752
+ provider: "huggingface",
4753
+ baseUrl: "https://router.huggingface.co/v1",
4754
+ compat: { "supportsDeveloperRole": false },
4755
+ reasoning: true,
4756
+ input: ["text", "image"],
4757
+ cost: {
4758
+ input: 0.6,
4759
+ output: 1.8,
4760
+ cacheRead: 0,
4761
+ cacheWrite: 0,
4762
+ },
4763
+ contextWindow: 65536,
4764
+ maxTokens: 16384,
4765
+ },
4766
+ "zai-org/GLM-4.6": {
4767
+ id: "zai-org/GLM-4.6",
4768
+ name: "GLM-4.6",
4769
+ api: "openai-completions",
4770
+ provider: "huggingface",
4771
+ baseUrl: "https://router.huggingface.co/v1",
4772
+ compat: { "supportsDeveloperRole": false },
4773
+ reasoning: true,
4774
+ input: ["text"],
4775
+ cost: {
4776
+ input: 0.55,
4777
+ output: 2.2,
4778
+ cacheRead: 0,
4779
+ cacheWrite: 0,
4780
+ },
4781
+ contextWindow: 204800,
4782
+ maxTokens: 131072,
4783
+ },
4440
4784
  "zai-org/GLM-4.7": {
4441
4785
  id: "zai-org/GLM-4.7",
4442
4786
  name: "GLM-4.7",
@@ -4506,7 +4850,25 @@ export const MODELS = {
4506
4850
  cacheRead: 0.2,
4507
4851
  cacheWrite: 0,
4508
4852
  },
4509
- contextWindow: 202752,
4853
+ contextWindow: 202752,
4854
+ maxTokens: 131072,
4855
+ },
4856
+ "zai-org/GLM-5.2": {
4857
+ id: "zai-org/GLM-5.2",
4858
+ name: "GLM-5.2",
4859
+ api: "openai-completions",
4860
+ provider: "huggingface",
4861
+ baseUrl: "https://router.huggingface.co/v1",
4862
+ compat: { "supportsDeveloperRole": false },
4863
+ reasoning: true,
4864
+ input: ["text"],
4865
+ cost: {
4866
+ input: 1.4,
4867
+ output: 4.4,
4868
+ cacheRead: 0,
4869
+ cacheWrite: 0,
4870
+ },
4871
+ contextWindow: 262144,
4510
4872
  maxTokens: 131072,
4511
4873
  },
4512
4874
  },
@@ -6328,6 +6690,23 @@ export const MODELS = {
6328
6690
  contextWindow: 204800,
6329
6691
  maxTokens: 131072,
6330
6692
  },
6693
+ "glm-5.2": {
6694
+ id: "glm-5.2",
6695
+ name: "GLM-5.2",
6696
+ api: "openai-completions",
6697
+ provider: "opencode",
6698
+ baseUrl: "https://opencode.ai/zen/v1",
6699
+ reasoning: true,
6700
+ input: ["text"],
6701
+ cost: {
6702
+ input: 1.4,
6703
+ output: 4.4,
6704
+ cacheRead: 0.26,
6705
+ cacheWrite: 0,
6706
+ },
6707
+ contextWindow: 1000000,
6708
+ maxTokens: 131072,
6709
+ },
6331
6710
  "gpt-5": {
6332
6711
  id: "gpt-5",
6333
6712
  name: "GPT-5",
@@ -7114,23 +7493,6 @@ export const MODELS = {
7114
7493
  contextWindow: 200000,
7115
7494
  maxTokens: 4096,
7116
7495
  },
7117
- "anthropic/claude-3.5-haiku": {
7118
- id: "anthropic/claude-3.5-haiku",
7119
- name: "Anthropic: Claude 3.5 Haiku",
7120
- api: "openai-completions",
7121
- provider: "openrouter",
7122
- baseUrl: "https://openrouter.ai/api/v1",
7123
- reasoning: false,
7124
- input: ["text", "image"],
7125
- cost: {
7126
- input: 0.7999999999999999,
7127
- output: 4,
7128
- cacheRead: 0.08,
7129
- cacheWrite: 1,
7130
- },
7131
- contextWindow: 200000,
7132
- maxTokens: 8192,
7133
- },
7134
7496
  "anthropic/claude-fable-5": {
7135
7497
  id: "anthropic/claude-fable-5",
7136
7498
  name: "Anthropic: Claude Fable 5",
@@ -7378,13 +7740,13 @@ export const MODELS = {
7378
7740
  reasoning: true,
7379
7741
  input: ["text"],
7380
7742
  cost: {
7381
- input: 0.22,
7382
- output: 0.85,
7743
+ input: 0.25,
7744
+ output: 0.7999999999999999,
7383
7745
  cacheRead: 0.06,
7384
7746
  cacheWrite: 0,
7385
7747
  },
7386
7748
  contextWindow: 262144,
7387
- maxTokens: 262144,
7749
+ maxTokens: 80000,
7388
7750
  },
7389
7751
  "arcee-ai/trinity-mini": {
7390
7752
  id: "arcee-ai/trinity-mini",
@@ -7726,23 +8088,6 @@ export const MODELS = {
7726
8088
  contextWindow: 1048576,
7727
8089
  maxTokens: 384000,
7728
8090
  },
7729
- "essentialai/rnj-1-instruct": {
7730
- id: "essentialai/rnj-1-instruct",
7731
- name: "EssentialAI: Rnj 1 Instruct",
7732
- api: "openai-completions",
7733
- provider: "openrouter",
7734
- baseUrl: "https://openrouter.ai/api/v1",
7735
- reasoning: false,
7736
- input: ["text"],
7737
- cost: {
7738
- input: 0.15,
7739
- output: 0.15,
7740
- cacheRead: 0,
7741
- cacheWrite: 0,
7742
- },
7743
- contextWindow: 32768,
7744
- maxTokens: 4096,
7745
- },
7746
8091
  "google/gemini-2.5-flash": {
7747
8092
  id: "google/gemini-2.5-flash",
7748
8093
  name: "Google: Gemini 2.5 Flash",
@@ -8364,13 +8709,13 @@ export const MODELS = {
8364
8709
  reasoning: true,
8365
8710
  input: ["text"],
8366
8711
  cost: {
8367
- input: 0.25,
8368
- output: 1,
8369
- cacheRead: 0.049999999999999996,
8712
+ input: 0.24,
8713
+ output: 0.96,
8714
+ cacheRead: 0,
8370
8715
  cacheWrite: 0,
8371
8716
  },
8372
8717
  contextWindow: 204800,
8373
- maxTokens: 131072,
8718
+ maxTokens: 196608,
8374
8719
  },
8375
8720
  "minimax/minimax-m3": {
8376
8721
  id: "minimax/minimax-m3",
@@ -8723,7 +9068,7 @@ export const MODELS = {
8723
9068
  cost: {
8724
9069
  input: 0.6,
8725
9070
  output: 2.5,
8726
- cacheRead: 0,
9071
+ cacheRead: 0.6,
8727
9072
  cacheWrite: 0,
8728
9073
  },
8729
9074
  contextWindow: 262144,
@@ -8755,9 +9100,9 @@ export const MODELS = {
8755
9100
  reasoning: true,
8756
9101
  input: ["text", "image"],
8757
9102
  cost: {
8758
- input: 0.67,
8759
- output: 3.5,
8760
- cacheRead: 0.19999999999999998,
9103
+ input: 0.66,
9104
+ output: 3.41,
9105
+ cacheRead: 0.144,
8761
9106
  cacheWrite: 0,
8762
9107
  },
8763
9108
  contextWindow: 262144,
@@ -8772,30 +9117,13 @@ export const MODELS = {
8772
9117
  reasoning: true,
8773
9118
  input: ["text", "image"],
8774
9119
  cost: {
8775
- input: 0.6799999999999999,
8776
- output: 3.41,
8777
- cacheRead: 0.144,
8778
- cacheWrite: 0,
8779
- },
8780
- contextWindow: 262144,
8781
- maxTokens: 262144,
8782
- },
8783
- "nex-agi/nex-n2-pro:free": {
8784
- id: "nex-agi/nex-n2-pro:free",
8785
- name: "Nex AGI: Nex-N2-Pro (free)",
8786
- api: "openai-completions",
8787
- provider: "openrouter",
8788
- baseUrl: "https://openrouter.ai/api/v1",
8789
- reasoning: true,
8790
- input: ["text", "image"],
8791
- cost: {
8792
- input: 0,
8793
- output: 0,
8794
- cacheRead: 0,
9120
+ input: 0.74,
9121
+ output: 3.5,
9122
+ cacheRead: 0.15,
8795
9123
  cacheWrite: 0,
8796
9124
  },
8797
9125
  contextWindow: 262144,
8798
- maxTokens: 262144,
9126
+ maxTokens: 16384,
8799
9127
  },
8800
9128
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
8801
9129
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -10004,23 +10332,6 @@ export const MODELS = {
10004
10332
  contextWindow: 262144,
10005
10333
  maxTokens: 32768,
10006
10334
  },
10007
- "prime-intellect/intellect-3": {
10008
- id: "prime-intellect/intellect-3",
10009
- name: "Prime Intellect: INTELLECT-3",
10010
- api: "openai-completions",
10011
- provider: "openrouter",
10012
- baseUrl: "https://openrouter.ai/api/v1",
10013
- reasoning: true,
10014
- input: ["text"],
10015
- cost: {
10016
- input: 0.19999999999999998,
10017
- output: 1.1,
10018
- cacheRead: 0,
10019
- cacheWrite: 0,
10020
- },
10021
- contextWindow: 131072,
10022
- maxTokens: 131072,
10023
- },
10024
10335
  "qwen/qwen-2.5-72b-instruct": {
10025
10336
  id: "qwen/qwen-2.5-72b-instruct",
10026
10337
  name: "Qwen2.5 72B Instruct",
@@ -10610,11 +10921,11 @@ export const MODELS = {
10610
10921
  cost: {
10611
10922
  input: 0.14,
10612
10923
  output: 1,
10613
- cacheRead: 0,
10924
+ cacheRead: 0.049999999999999996,
10614
10925
  cacheWrite: 0,
10615
10926
  },
10616
10927
  contextWindow: 262144,
10617
- maxTokens: 262144,
10928
+ maxTokens: 81920,
10618
10929
  },
10619
10930
  "qwen/qwen3.5-397b-a17b": {
10620
10931
  id: "qwen/qwen3.5-397b-a17b",
@@ -10854,6 +11165,23 @@ export const MODELS = {
10854
11165
  contextWindow: 256000,
10855
11166
  maxTokens: 128000,
10856
11167
  },
11168
+ "sakana/fugu-ultra": {
11169
+ id: "sakana/fugu-ultra",
11170
+ name: "Sakana: Fugu Ultra",
11171
+ api: "openai-completions",
11172
+ provider: "openrouter",
11173
+ baseUrl: "https://openrouter.ai/api/v1",
11174
+ reasoning: true,
11175
+ input: ["text", "image"],
11176
+ cost: {
11177
+ input: 5,
11178
+ output: 30,
11179
+ cacheRead: 0.5,
11180
+ cacheWrite: 0,
11181
+ },
11182
+ contextWindow: 1000000,
11183
+ maxTokens: 128000,
11184
+ },
10857
11185
  "sao10k/l3.1-euryale-70b": {
10858
11186
  id: "sao10k/l3.1-euryale-70b",
10859
11187
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -10914,30 +11242,13 @@ export const MODELS = {
10914
11242
  reasoning: true,
10915
11243
  input: ["text"],
10916
11244
  cost: {
10917
- input: 0.06599999999999999,
10918
- output: 0.26,
10919
- cacheRead: 0.029,
11245
+ input: 0.063,
11246
+ output: 0.21,
11247
+ cacheRead: 0.020999999999999998,
10920
11248
  cacheWrite: 0,
10921
11249
  },
10922
11250
  contextWindow: 262144,
10923
- maxTokens: 262144,
10924
- },
10925
- "thedrummer/rocinante-12b": {
10926
- id: "thedrummer/rocinante-12b",
10927
- name: "TheDrummer: Rocinante 12B",
10928
- api: "openai-completions",
10929
- provider: "openrouter",
10930
- baseUrl: "https://openrouter.ai/api/v1",
10931
- reasoning: false,
10932
- input: ["text"],
10933
- cost: {
10934
- input: 0.16999999999999998,
10935
- output: 0.43,
10936
- cacheRead: 0,
10937
- cacheWrite: 0,
10938
- },
10939
- contextWindow: 32768,
10940
- maxTokens: 32768,
11251
+ maxTokens: 4096,
10941
11252
  },
10942
11253
  "thedrummer/unslopnemo-12b": {
10943
11254
  id: "thedrummer/unslopnemo-12b",
@@ -11033,13 +11344,13 @@ export const MODELS = {
11033
11344
  reasoning: true,
11034
11345
  input: ["text", "image"],
11035
11346
  cost: {
11036
- input: 0.14,
11347
+ input: 0.105,
11037
11348
  output: 0.28,
11038
- cacheRead: 0.0028,
11349
+ cacheRead: 0,
11039
11350
  cacheWrite: 0,
11040
11351
  },
11041
11352
  contextWindow: 1048576,
11042
- maxTokens: 131072,
11353
+ maxTokens: 4096,
11043
11354
  },
11044
11355
  "xiaomi/mimo-v2.5-pro": {
11045
11356
  id: "xiaomi/mimo-v2.5-pro",
@@ -11237,12 +11548,29 @@ export const MODELS = {
11237
11548
  reasoning: true,
11238
11549
  input: ["text"],
11239
11550
  cost: {
11240
- input: 1.2,
11241
- output: 4.1,
11242
- cacheRead: 0.19999999999999998,
11551
+ input: 0.95,
11552
+ output: 3,
11553
+ cacheRead: 0.18,
11243
11554
  cacheWrite: 0,
11244
11555
  },
11245
11556
  contextWindow: 1048576,
11557
+ maxTokens: 32768,
11558
+ },
11559
+ "z-ai/glm-5v-turbo": {
11560
+ id: "z-ai/glm-5v-turbo",
11561
+ name: "Z.ai: GLM 5V Turbo",
11562
+ api: "openai-completions",
11563
+ provider: "openrouter",
11564
+ baseUrl: "https://openrouter.ai/api/v1",
11565
+ reasoning: true,
11566
+ input: ["text", "image"],
11567
+ cost: {
11568
+ input: 1.2,
11569
+ output: 4,
11570
+ cacheRead: 0.24,
11571
+ cacheWrite: 0,
11572
+ },
11573
+ contextWindow: 202752,
11246
11574
  maxTokens: 131072,
11247
11575
  },
11248
11576
  "~anthropic/claude-fable-latest": {
@@ -11356,9 +11684,9 @@ export const MODELS = {
11356
11684
  reasoning: true,
11357
11685
  input: ["text", "image"],
11358
11686
  cost: {
11359
- input: 0.67,
11360
- output: 3.5,
11361
- cacheRead: 0.19999999999999998,
11687
+ input: 0.66,
11688
+ output: 3.41,
11689
+ cacheRead: 0.144,
11362
11690
  cacheWrite: 0,
11363
11691
  },
11364
11692
  contextWindow: 262144,
@@ -11508,7 +11836,7 @@ export const MODELS = {
11508
11836
  api: "anthropic-messages",
11509
11837
  provider: "vercel-ai-gateway",
11510
11838
  baseUrl: "https://ai-gateway.vercel.sh",
11511
- reasoning: true,
11839
+ reasoning: false,
11512
11840
  input: ["text"],
11513
11841
  cost: {
11514
11842
  input: 1.5,
@@ -11525,7 +11853,7 @@ export const MODELS = {
11525
11853
  api: "anthropic-messages",
11526
11854
  provider: "vercel-ai-gateway",
11527
11855
  baseUrl: "https://ai-gateway.vercel.sh",
11528
- reasoning: true,
11856
+ reasoning: false,
11529
11857
  input: ["text"],
11530
11858
  cost: {
11531
11859
  input: 0.15,
@@ -11542,7 +11870,7 @@ export const MODELS = {
11542
11870
  api: "anthropic-messages",
11543
11871
  provider: "vercel-ai-gateway",
11544
11872
  baseUrl: "https://ai-gateway.vercel.sh",
11545
- reasoning: true,
11873
+ reasoning: false,
11546
11874
  input: ["text"],
11547
11875
  cost: {
11548
11876
  input: 0.5,
@@ -11655,6 +11983,40 @@ export const MODELS = {
11655
11983
  contextWindow: 131072,
11656
11984
  maxTokens: 32768,
11657
11985
  },
11986
+ "alibaba/qwen3-vl-235b-a22b-instruct": {
11987
+ id: "alibaba/qwen3-vl-235b-a22b-instruct",
11988
+ name: "Qwen3 VL 235B A22B Instruct",
11989
+ api: "anthropic-messages",
11990
+ provider: "vercel-ai-gateway",
11991
+ baseUrl: "https://ai-gateway.vercel.sh",
11992
+ reasoning: false,
11993
+ input: ["text", "image"],
11994
+ cost: {
11995
+ input: 0.39999999999999997,
11996
+ output: 1.5999999999999999,
11997
+ cacheRead: 0,
11998
+ cacheWrite: 0,
11999
+ },
12000
+ contextWindow: 131072,
12001
+ maxTokens: 129024,
12002
+ },
12003
+ "alibaba/qwen3-vl-instruct": {
12004
+ id: "alibaba/qwen3-vl-instruct",
12005
+ name: "Qwen3 VL 235B A22B Instruct",
12006
+ api: "anthropic-messages",
12007
+ provider: "vercel-ai-gateway",
12008
+ baseUrl: "https://ai-gateway.vercel.sh",
12009
+ reasoning: false,
12010
+ input: ["text", "image"],
12011
+ cost: {
12012
+ input: 0.39999999999999997,
12013
+ output: 1.5999999999999999,
12014
+ cacheRead: 0,
12015
+ cacheWrite: 0,
12016
+ },
12017
+ contextWindow: 131072,
12018
+ maxTokens: 129024,
12019
+ },
11658
12020
  "alibaba/qwen3-vl-thinking": {
11659
12021
  id: "alibaba/qwen3-vl-thinking",
11660
12022
  name: "Qwen3 VL 235B A22B Thinking",
@@ -11749,30 +12111,98 @@ export const MODELS = {
11749
12111
  reasoning: true,
11750
12112
  input: ["text"],
11751
12113
  cost: {
11752
- input: 1.25,
11753
- output: 3.75,
11754
- cacheRead: 0.25,
11755
- cacheWrite: 1.5625,
12114
+ input: 1.25,
12115
+ output: 3.75,
12116
+ cacheRead: 0.25,
12117
+ cacheWrite: 1.5625,
12118
+ },
12119
+ contextWindow: 991000,
12120
+ maxTokens: 64000,
12121
+ },
12122
+ "alibaba/qwen3.7-plus": {
12123
+ id: "alibaba/qwen3.7-plus",
12124
+ name: "Qwen 3.7 Plus",
12125
+ api: "anthropic-messages",
12126
+ provider: "vercel-ai-gateway",
12127
+ baseUrl: "https://ai-gateway.vercel.sh",
12128
+ reasoning: true,
12129
+ input: ["text", "image"],
12130
+ cost: {
12131
+ input: 0.39999999999999997,
12132
+ output: 1.5999999999999999,
12133
+ cacheRead: 0.08,
12134
+ cacheWrite: 0.5,
12135
+ },
12136
+ contextWindow: 1000000,
12137
+ maxTokens: 64000,
12138
+ },
12139
+ "amazon/nova-2-lite": {
12140
+ id: "amazon/nova-2-lite",
12141
+ name: "Nova 2 Lite",
12142
+ api: "anthropic-messages",
12143
+ provider: "vercel-ai-gateway",
12144
+ baseUrl: "https://ai-gateway.vercel.sh",
12145
+ reasoning: true,
12146
+ input: ["text", "image"],
12147
+ cost: {
12148
+ input: 0.3,
12149
+ output: 2.5,
12150
+ cacheRead: 0.075,
12151
+ cacheWrite: 0,
12152
+ },
12153
+ contextWindow: 1000000,
12154
+ maxTokens: 1000000,
12155
+ },
12156
+ "amazon/nova-lite": {
12157
+ id: "amazon/nova-lite",
12158
+ name: "Nova Lite",
12159
+ api: "anthropic-messages",
12160
+ provider: "vercel-ai-gateway",
12161
+ baseUrl: "https://ai-gateway.vercel.sh",
12162
+ reasoning: false,
12163
+ input: ["text", "image"],
12164
+ cost: {
12165
+ input: 0.06,
12166
+ output: 0.24,
12167
+ cacheRead: 0,
12168
+ cacheWrite: 0,
12169
+ },
12170
+ contextWindow: 300000,
12171
+ maxTokens: 8192,
12172
+ },
12173
+ "amazon/nova-micro": {
12174
+ id: "amazon/nova-micro",
12175
+ name: "Nova Micro",
12176
+ api: "anthropic-messages",
12177
+ provider: "vercel-ai-gateway",
12178
+ baseUrl: "https://ai-gateway.vercel.sh",
12179
+ reasoning: false,
12180
+ input: ["text"],
12181
+ cost: {
12182
+ input: 0.035,
12183
+ output: 0.14,
12184
+ cacheRead: 0,
12185
+ cacheWrite: 0,
11756
12186
  },
11757
- contextWindow: 991000,
11758
- maxTokens: 64000,
12187
+ contextWindow: 128000,
12188
+ maxTokens: 8192,
11759
12189
  },
11760
- "alibaba/qwen3.7-plus": {
11761
- id: "alibaba/qwen3.7-plus",
11762
- name: "Qwen 3.7 Plus",
12190
+ "amazon/nova-pro": {
12191
+ id: "amazon/nova-pro",
12192
+ name: "Nova Pro",
11763
12193
  api: "anthropic-messages",
11764
12194
  provider: "vercel-ai-gateway",
11765
12195
  baseUrl: "https://ai-gateway.vercel.sh",
11766
- reasoning: true,
12196
+ reasoning: false,
11767
12197
  input: ["text", "image"],
11768
12198
  cost: {
11769
- input: 0.39999999999999997,
11770
- output: 1.5999999999999999,
11771
- cacheRead: 0.08,
11772
- cacheWrite: 0.5,
12199
+ input: 0.7999999999999999,
12200
+ output: 3.1999999999999997,
12201
+ cacheRead: 0,
12202
+ cacheWrite: 0,
11773
12203
  },
11774
- contextWindow: 1000000,
11775
- maxTokens: 64000,
12204
+ contextWindow: 300000,
12205
+ maxTokens: 8192,
11776
12206
  },
11777
12207
  "anthropic/claude-3-haiku": {
11778
12208
  id: "anthropic/claude-3-haiku",
@@ -12012,6 +12442,23 @@ export const MODELS = {
12012
12442
  contextWindow: 262100,
12013
12443
  maxTokens: 80000,
12014
12444
  },
12445
+ "arcee-ai/trinity-mini": {
12446
+ id: "arcee-ai/trinity-mini",
12447
+ name: "Trinity Mini",
12448
+ api: "anthropic-messages",
12449
+ provider: "vercel-ai-gateway",
12450
+ baseUrl: "https://ai-gateway.vercel.sh",
12451
+ reasoning: false,
12452
+ input: ["text"],
12453
+ cost: {
12454
+ input: 0.045,
12455
+ output: 0.15,
12456
+ cacheRead: 0,
12457
+ cacheWrite: 0,
12458
+ },
12459
+ contextWindow: 131072,
12460
+ maxTokens: 131072,
12461
+ },
12015
12462
  "bytedance/seed-1.6": {
12016
12463
  id: "bytedance/seed-1.6",
12017
12464
  name: "Seed 1.6",
@@ -12019,7 +12466,7 @@ export const MODELS = {
12019
12466
  provider: "vercel-ai-gateway",
12020
12467
  baseUrl: "https://ai-gateway.vercel.sh",
12021
12468
  reasoning: true,
12022
- input: ["text"],
12469
+ input: ["text", "image"],
12023
12470
  cost: {
12024
12471
  input: 0.25,
12025
12472
  output: 2,
@@ -12029,6 +12476,23 @@ export const MODELS = {
12029
12476
  contextWindow: 256000,
12030
12477
  maxTokens: 32000,
12031
12478
  },
12479
+ "bytedance/seed-1.8": {
12480
+ id: "bytedance/seed-1.8",
12481
+ name: "Bytedance Seed 1.8",
12482
+ api: "anthropic-messages",
12483
+ provider: "vercel-ai-gateway",
12484
+ baseUrl: "https://ai-gateway.vercel.sh",
12485
+ reasoning: true,
12486
+ input: ["text", "image"],
12487
+ cost: {
12488
+ input: 0.25,
12489
+ output: 2,
12490
+ cacheRead: 0.049999999999999996,
12491
+ cacheWrite: 0,
12492
+ },
12493
+ contextWindow: 256000,
12494
+ maxTokens: 64000,
12495
+ },
12032
12496
  "cohere/command-a": {
12033
12497
  id: "cohere/command-a",
12034
12498
  name: "Command A",
@@ -12120,8 +12584,8 @@ export const MODELS = {
12120
12584
  api: "anthropic-messages",
12121
12585
  provider: "vercel-ai-gateway",
12122
12586
  baseUrl: "https://ai-gateway.vercel.sh",
12123
- reasoning: true,
12124
- input: ["text", "image"],
12587
+ reasoning: false,
12588
+ input: ["text"],
12125
12589
  cost: {
12126
12590
  input: 0.28,
12127
12591
  output: 0.42,
@@ -12138,7 +12602,7 @@ export const MODELS = {
12138
12602
  provider: "vercel-ai-gateway",
12139
12603
  baseUrl: "https://ai-gateway.vercel.sh",
12140
12604
  reasoning: true,
12141
- input: ["text", "image"],
12605
+ input: ["text"],
12142
12606
  cost: {
12143
12607
  input: 0.62,
12144
12608
  output: 1.85,
@@ -12155,7 +12619,7 @@ export const MODELS = {
12155
12619
  provider: "vercel-ai-gateway",
12156
12620
  baseUrl: "https://ai-gateway.vercel.sh",
12157
12621
  reasoning: true,
12158
- input: ["text", "image"],
12622
+ input: ["text"],
12159
12623
  cost: {
12160
12624
  input: 0.14,
12161
12625
  output: 0.28,
@@ -12358,7 +12822,7 @@ export const MODELS = {
12358
12822
  api: "anthropic-messages",
12359
12823
  provider: "vercel-ai-gateway",
12360
12824
  baseUrl: "https://ai-gateway.vercel.sh",
12361
- reasoning: false,
12825
+ reasoning: true,
12362
12826
  input: ["text", "image"],
12363
12827
  cost: {
12364
12828
  input: 0.14,
@@ -12403,6 +12867,40 @@ export const MODELS = {
12403
12867
  contextWindow: 32000,
12404
12868
  maxTokens: 16384,
12405
12869
  },
12870
+ "interfaze/interfaze-beta": {
12871
+ id: "interfaze/interfaze-beta",
12872
+ name: "Interfaze Beta",
12873
+ api: "anthropic-messages",
12874
+ provider: "vercel-ai-gateway",
12875
+ baseUrl: "https://ai-gateway.vercel.sh",
12876
+ reasoning: true,
12877
+ input: ["text", "image"],
12878
+ cost: {
12879
+ input: 1.5,
12880
+ output: 3.5,
12881
+ cacheRead: 0,
12882
+ cacheWrite: 0,
12883
+ },
12884
+ contextWindow: 1000000,
12885
+ maxTokens: 32000,
12886
+ },
12887
+ "kwaipilot/kat-coder-pro-v1": {
12888
+ id: "kwaipilot/kat-coder-pro-v1",
12889
+ name: "KAT-Coder-Pro V1",
12890
+ api: "anthropic-messages",
12891
+ provider: "vercel-ai-gateway",
12892
+ baseUrl: "https://ai-gateway.vercel.sh",
12893
+ reasoning: false,
12894
+ input: ["text"],
12895
+ cost: {
12896
+ input: 0.3,
12897
+ output: 1.2,
12898
+ cacheRead: 0.06,
12899
+ cacheWrite: 0,
12900
+ },
12901
+ contextWindow: 256000,
12902
+ maxTokens: 32000,
12903
+ },
12406
12904
  "kwaipilot/kat-coder-pro-v2": {
12407
12905
  id: "kwaipilot/kat-coder-pro-v2",
12408
12906
  name: "Kat Coder Pro V2",
@@ -12437,6 +12935,23 @@ export const MODELS = {
12437
12935
  contextWindow: 128000,
12438
12936
  maxTokens: 100000,
12439
12937
  },
12938
+ "meituan/longcat-flash-thinking-2601": {
12939
+ id: "meituan/longcat-flash-thinking-2601",
12940
+ name: "LongCat Flash Thinking 2601",
12941
+ api: "anthropic-messages",
12942
+ provider: "vercel-ai-gateway",
12943
+ baseUrl: "https://ai-gateway.vercel.sh",
12944
+ reasoning: true,
12945
+ input: ["text"],
12946
+ cost: {
12947
+ input: 0,
12948
+ output: 0,
12949
+ cacheRead: 0,
12950
+ cacheWrite: 0,
12951
+ },
12952
+ contextWindow: 32768,
12953
+ maxTokens: 32768,
12954
+ },
12440
12955
  "meta/llama-3.1-70b": {
12441
12956
  id: "meta/llama-3.1-70b",
12442
12957
  name: "Llama 3.1 70B Instruct",
@@ -12750,7 +13265,7 @@ export const MODELS = {
12750
13265
  provider: "vercel-ai-gateway",
12751
13266
  baseUrl: "https://ai-gateway.vercel.sh",
12752
13267
  reasoning: false,
12753
- input: ["text"],
13268
+ input: ["text", "image"],
12754
13269
  cost: {
12755
13270
  input: 0.09999999999999999,
12756
13271
  output: 0.3,
@@ -12760,6 +13275,57 @@ export const MODELS = {
12760
13275
  contextWindow: 256000,
12761
13276
  maxTokens: 256000,
12762
13277
  },
13278
+ "mistral/magistral-medium": {
13279
+ id: "mistral/magistral-medium",
13280
+ name: "Magistral Medium 2509",
13281
+ api: "anthropic-messages",
13282
+ provider: "vercel-ai-gateway",
13283
+ baseUrl: "https://ai-gateway.vercel.sh",
13284
+ reasoning: true,
13285
+ input: ["text", "image"],
13286
+ cost: {
13287
+ input: 2,
13288
+ output: 5,
13289
+ cacheRead: 0,
13290
+ cacheWrite: 0,
13291
+ },
13292
+ contextWindow: 128000,
13293
+ maxTokens: 64000,
13294
+ },
13295
+ "mistral/magistral-small": {
13296
+ id: "mistral/magistral-small",
13297
+ name: "Magistral Small 2509",
13298
+ api: "anthropic-messages",
13299
+ provider: "vercel-ai-gateway",
13300
+ baseUrl: "https://ai-gateway.vercel.sh",
13301
+ reasoning: true,
13302
+ input: ["text", "image"],
13303
+ cost: {
13304
+ input: 0.5,
13305
+ output: 1.5,
13306
+ cacheRead: 0,
13307
+ cacheWrite: 0,
13308
+ },
13309
+ contextWindow: 128000,
13310
+ maxTokens: 64000,
13311
+ },
13312
+ "mistral/ministral-14b": {
13313
+ id: "mistral/ministral-14b",
13314
+ name: "Ministral 14B",
13315
+ api: "anthropic-messages",
13316
+ provider: "vercel-ai-gateway",
13317
+ baseUrl: "https://ai-gateway.vercel.sh",
13318
+ reasoning: false,
13319
+ input: ["text", "image"],
13320
+ cost: {
13321
+ input: 0.19999999999999998,
13322
+ output: 0.19999999999999998,
13323
+ cacheRead: 0,
13324
+ cacheWrite: 0,
13325
+ },
13326
+ contextWindow: 256000,
13327
+ maxTokens: 256000,
13328
+ },
12763
13329
  "mistral/ministral-3b": {
12764
13330
  id: "mistral/ministral-3b",
12765
13331
  name: "Ministral 3B",
@@ -12794,6 +13360,23 @@ export const MODELS = {
12794
13360
  contextWindow: 128000,
12795
13361
  maxTokens: 4000,
12796
13362
  },
13363
+ "mistral/mistral-large-3": {
13364
+ id: "mistral/mistral-large-3",
13365
+ name: "Mistral Large 3",
13366
+ api: "anthropic-messages",
13367
+ provider: "vercel-ai-gateway",
13368
+ baseUrl: "https://ai-gateway.vercel.sh",
13369
+ reasoning: false,
13370
+ input: ["text", "image"],
13371
+ cost: {
13372
+ input: 0.5,
13373
+ output: 1.5,
13374
+ cacheRead: 0,
13375
+ cacheWrite: 0,
13376
+ },
13377
+ contextWindow: 256000,
13378
+ maxTokens: 256000,
13379
+ },
12797
13380
  "mistral/mistral-medium": {
12798
13381
  id: "mistral/mistral-medium",
12799
13382
  name: "Mistral Medium 3.1",
@@ -12818,7 +13401,7 @@ export const MODELS = {
12818
13401
  provider: "vercel-ai-gateway",
12819
13402
  baseUrl: "https://ai-gateway.vercel.sh",
12820
13403
  reasoning: true,
12821
- input: ["text"],
13404
+ input: ["text", "image"],
12822
13405
  cost: {
12823
13406
  input: 1.5,
12824
13407
  output: 7.5,
@@ -12998,6 +13581,23 @@ export const MODELS = {
12998
13581
  contextWindow: 262144,
12999
13582
  maxTokens: 32768,
13000
13583
  },
13584
+ "nvidia/nemotron-3-nano-30b-a3b": {
13585
+ id: "nvidia/nemotron-3-nano-30b-a3b",
13586
+ name: "Nemotron 3 Nano 30B A3B",
13587
+ api: "anthropic-messages",
13588
+ provider: "vercel-ai-gateway",
13589
+ baseUrl: "https://ai-gateway.vercel.sh",
13590
+ reasoning: true,
13591
+ input: ["text"],
13592
+ cost: {
13593
+ input: 0.049999999999999996,
13594
+ output: 0.24,
13595
+ cacheRead: 0,
13596
+ cacheWrite: 0,
13597
+ },
13598
+ contextWindow: 262144,
13599
+ maxTokens: 262144,
13600
+ },
13001
13601
  "nvidia/nemotron-3-super-120b-a12b": {
13002
13602
  id: "nvidia/nemotron-3-super-120b-a12b",
13003
13603
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -13066,6 +13666,23 @@ export const MODELS = {
13066
13666
  contextWindow: 131072,
13067
13667
  maxTokens: 131072,
13068
13668
  },
13669
+ "openai/gpt-3.5-turbo": {
13670
+ id: "openai/gpt-3.5-turbo",
13671
+ name: "GPT-3.5 Turbo",
13672
+ api: "anthropic-messages",
13673
+ provider: "vercel-ai-gateway",
13674
+ baseUrl: "https://ai-gateway.vercel.sh",
13675
+ reasoning: false,
13676
+ input: ["text"],
13677
+ cost: {
13678
+ input: 0.5,
13679
+ output: 1.5,
13680
+ cacheRead: 0,
13681
+ cacheWrite: 0,
13682
+ },
13683
+ contextWindow: 16385,
13684
+ maxTokens: 4096,
13685
+ },
13069
13686
  "openai/gpt-4-turbo": {
13070
13687
  id: "openai/gpt-4-turbo",
13071
13688
  name: "GPT-4 Turbo",
@@ -13191,7 +13808,7 @@ export const MODELS = {
13191
13808
  api: "anthropic-messages",
13192
13809
  provider: "vercel-ai-gateway",
13193
13810
  baseUrl: "https://ai-gateway.vercel.sh",
13194
- reasoning: true,
13811
+ reasoning: false,
13195
13812
  input: ["text", "image"],
13196
13813
  cost: {
13197
13814
  input: 1.25,
@@ -13327,7 +13944,7 @@ export const MODELS = {
13327
13944
  api: "anthropic-messages",
13328
13945
  provider: "vercel-ai-gateway",
13329
13946
  baseUrl: "https://ai-gateway.vercel.sh",
13330
- reasoning: true,
13947
+ reasoning: false,
13331
13948
  input: ["text", "image"],
13332
13949
  cost: {
13333
13950
  input: 1.25,
@@ -13378,7 +13995,7 @@ export const MODELS = {
13378
13995
  api: "anthropic-messages",
13379
13996
  provider: "vercel-ai-gateway",
13380
13997
  baseUrl: "https://ai-gateway.vercel.sh",
13381
- reasoning: true,
13998
+ reasoning: false,
13382
13999
  input: ["text", "image"],
13383
14000
  cost: {
13384
14001
  input: 1.75,
@@ -13429,7 +14046,7 @@ export const MODELS = {
13429
14046
  api: "anthropic-messages",
13430
14047
  provider: "vercel-ai-gateway",
13431
14048
  baseUrl: "https://ai-gateway.vercel.sh",
13432
- reasoning: true,
14049
+ reasoning: false,
13433
14050
  input: ["text", "image"],
13434
14051
  cost: {
13435
14052
  input: 1.75,
@@ -13712,39 +14329,22 @@ export const MODELS = {
13712
14329
  contextWindow: 200000,
13713
14330
  maxTokens: 100000,
13714
14331
  },
13715
- "perplexity/sonar": {
13716
- id: "perplexity/sonar",
13717
- name: "Sonar",
13718
- api: "anthropic-messages",
13719
- provider: "vercel-ai-gateway",
13720
- baseUrl: "https://ai-gateway.vercel.sh",
13721
- reasoning: false,
13722
- input: ["text", "image"],
13723
- cost: {
13724
- input: 0,
13725
- output: 0,
13726
- cacheRead: 0,
13727
- cacheWrite: 0,
13728
- },
13729
- contextWindow: 127000,
13730
- maxTokens: 8000,
13731
- },
13732
- "perplexity/sonar-pro": {
13733
- id: "perplexity/sonar-pro",
13734
- name: "Sonar Pro",
14332
+ "sakana/fugu-ultra": {
14333
+ id: "sakana/fugu-ultra",
14334
+ name: "Fugu Ultra",
13735
14335
  api: "anthropic-messages",
13736
14336
  provider: "vercel-ai-gateway",
13737
14337
  baseUrl: "https://ai-gateway.vercel.sh",
13738
- reasoning: false,
14338
+ reasoning: true,
13739
14339
  input: ["text", "image"],
13740
14340
  cost: {
13741
- input: 0,
13742
- output: 0,
13743
- cacheRead: 0,
14341
+ input: 5,
14342
+ output: 30,
14343
+ cacheRead: 0.5,
13744
14344
  cacheWrite: 0,
13745
14345
  },
13746
- contextWindow: 200000,
13747
- maxTokens: 8000,
14346
+ contextWindow: 1000000,
14347
+ maxTokens: 1000000,
13748
14348
  },
13749
14349
  "stepfun/step-3.5-flash": {
13750
14350
  id: "stepfun/step-3.5-flash",
@@ -14212,7 +14812,7 @@ export const MODELS = {
14212
14812
  provider: "vercel-ai-gateway",
14213
14813
  baseUrl: "https://ai-gateway.vercel.sh",
14214
14814
  reasoning: true,
14215
- input: ["text", "image"],
14815
+ input: ["text"],
14216
14816
  cost: {
14217
14817
  input: 1.4,
14218
14818
  output: 4.4,
@@ -14239,6 +14839,23 @@ export const MODELS = {
14239
14839
  contextWindow: 1000000,
14240
14840
  maxTokens: 128000,
14241
14841
  },
14842
+ "zai/glm-5.2-fast": {
14843
+ id: "zai/glm-5.2-fast",
14844
+ name: "GLM 5.2 Fast",
14845
+ api: "anthropic-messages",
14846
+ provider: "vercel-ai-gateway",
14847
+ baseUrl: "https://ai-gateway.vercel.sh",
14848
+ reasoning: true,
14849
+ input: ["text"],
14850
+ cost: {
14851
+ input: 3,
14852
+ output: 10.25,
14853
+ cacheRead: 0.5,
14854
+ cacheWrite: 0,
14855
+ },
14856
+ contextWindow: 1000000,
14857
+ maxTokens: 128000,
14858
+ },
14242
14859
  "zai/glm-5v-turbo": {
14243
14860
  id: "zai/glm-5v-turbo",
14244
14861
  name: "GLM 5V Turbo",