@dreb/ai 2.31.1 → 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.
@@ -4079,6 +4079,24 @@ export const MODELS = {
4079
4079
  },
4080
4080
  },
4081
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
+ },
4082
4100
  "MiniMaxAI/MiniMax-M2.1": {
4083
4101
  id: "MiniMaxAI/MiniMax-M2.1",
4084
4102
  name: "MiniMax-M2.1",
@@ -4133,6 +4151,42 @@ export const MODELS = {
4133
4151
  contextWindow: 204800,
4134
4152
  maxTokens: 131072,
4135
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
+ },
4136
4190
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
4137
4191
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
4138
4192
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -4151,6 +4205,42 @@ export const MODELS = {
4151
4205
  contextWindow: 262144,
4152
4206
  maxTokens: 131072,
4153
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
+ },
4154
4244
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
4155
4245
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
4156
4246
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -4223,6 +4313,60 @@ export const MODELS = {
4223
4313
  contextWindow: 262144,
4224
4314
  maxTokens: 131072,
4225
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
+ },
4226
4370
  "Qwen/Qwen3.5-397B-A17B": {
4227
4371
  id: "Qwen/Qwen3.5-397B-A17B",
4228
4372
  name: "Qwen3.5-397B-A17B",
@@ -4241,6 +4385,42 @@ export const MODELS = {
4241
4385
  contextWindow: 262144,
4242
4386
  maxTokens: 32768,
4243
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
+ },
4244
4424
  "XiaomiMiMo/MiMo-V2-Flash": {
4245
4425
  id: "XiaomiMiMo/MiMo-V2-Flash",
4246
4426
  name: "MiMo-V2-Flash",
@@ -4259,6 +4439,24 @@ export const MODELS = {
4259
4439
  contextWindow: 262144,
4260
4440
  maxTokens: 4096,
4261
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
+ },
4262
4460
  "deepseek-ai/DeepSeek-R1-0528": {
4263
4461
  id: "deepseek-ai/DeepSeek-R1-0528",
4264
4462
  name: "DeepSeek-R1-0528",
@@ -4295,6 +4493,24 @@ export const MODELS = {
4295
4493
  contextWindow: 163840,
4296
4494
  maxTokens: 65536,
4297
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
+ },
4298
4514
  "deepseek-ai/DeepSeek-V4-Pro": {
4299
4515
  id: "deepseek-ai/DeepSeek-V4-Pro",
4300
4516
  name: "DeepSeek V4 Pro",
@@ -4313,6 +4529,60 @@ export const MODELS = {
4313
4529
  contextWindow: 1048576,
4314
4530
  maxTokens: 393216,
4315
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
+ },
4316
4586
  "moonshotai/Kimi-K2-Instruct": {
4317
4587
  id: "moonshotai/Kimi-K2-Instruct",
4318
4588
  name: "Kimi-K2-Instruct",
@@ -4346,12 +4616,120 @@ export const MODELS = {
4346
4616
  cacheRead: 0,
4347
4617
  cacheWrite: 0,
4348
4618
  },
4349
- contextWindow: 262144,
4350
- maxTokens: 16384,
4619
+ contextWindow: 262144,
4620
+ maxTokens: 16384,
4621
+ },
4622
+ "moonshotai/Kimi-K2-Thinking": {
4623
+ id: "moonshotai/Kimi-K2-Thinking",
4624
+ name: "Kimi-K2-Thinking",
4625
+ api: "openai-completions",
4626
+ provider: "huggingface",
4627
+ baseUrl: "https://router.huggingface.co/v1",
4628
+ compat: { "supportsDeveloperRole": false },
4629
+ reasoning: true,
4630
+ input: ["text"],
4631
+ cost: {
4632
+ input: 0.6,
4633
+ output: 2.5,
4634
+ cacheRead: 0.15,
4635
+ cacheWrite: 0,
4636
+ },
4637
+ contextWindow: 262144,
4638
+ maxTokens: 262144,
4639
+ },
4640
+ "moonshotai/Kimi-K2.5": {
4641
+ id: "moonshotai/Kimi-K2.5",
4642
+ name: "Kimi-K2.5",
4643
+ api: "openai-completions",
4644
+ provider: "huggingface",
4645
+ baseUrl: "https://router.huggingface.co/v1",
4646
+ compat: { "supportsDeveloperRole": false },
4647
+ reasoning: true,
4648
+ input: ["text", "image"],
4649
+ cost: {
4650
+ input: 0.6,
4651
+ output: 3,
4652
+ cacheRead: 0.1,
4653
+ cacheWrite: 0,
4654
+ },
4655
+ contextWindow: 262144,
4656
+ maxTokens: 262144,
4657
+ },
4658
+ "moonshotai/Kimi-K2.6": {
4659
+ id: "moonshotai/Kimi-K2.6",
4660
+ name: "Kimi-K2.6",
4661
+ api: "openai-completions",
4662
+ provider: "huggingface",
4663
+ baseUrl: "https://router.huggingface.co/v1",
4664
+ compat: { "supportsDeveloperRole": false },
4665
+ reasoning: true,
4666
+ input: ["text", "image"],
4667
+ cost: {
4668
+ input: 0.95,
4669
+ output: 4,
4670
+ cacheRead: 0.16,
4671
+ cacheWrite: 0,
4672
+ },
4673
+ contextWindow: 262144,
4674
+ maxTokens: 262144,
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,
4351
4729
  },
4352
- "moonshotai/Kimi-K2-Thinking": {
4353
- id: "moonshotai/Kimi-K2-Thinking",
4354
- name: "Kimi-K2-Thinking",
4730
+ "zai-org/GLM-4.5-Air": {
4731
+ id: "zai-org/GLM-4.5-Air",
4732
+ name: "GLM-4.5-Air",
4355
4733
  api: "openai-completions",
4356
4734
  provider: "huggingface",
4357
4735
  baseUrl: "https://router.huggingface.co/v1",
@@ -4359,17 +4737,17 @@ export const MODELS = {
4359
4737
  reasoning: true,
4360
4738
  input: ["text"],
4361
4739
  cost: {
4362
- input: 0.6,
4363
- output: 2.5,
4364
- cacheRead: 0.15,
4740
+ input: 0.13,
4741
+ output: 0.85,
4742
+ cacheRead: 0,
4365
4743
  cacheWrite: 0,
4366
4744
  },
4367
- contextWindow: 262144,
4368
- maxTokens: 262144,
4745
+ contextWindow: 131072,
4746
+ maxTokens: 98304,
4369
4747
  },
4370
- "moonshotai/Kimi-K2.5": {
4371
- id: "moonshotai/Kimi-K2.5",
4372
- name: "Kimi-K2.5",
4748
+ "zai-org/GLM-4.5V": {
4749
+ id: "zai-org/GLM-4.5V",
4750
+ name: "GLM-4.5V",
4373
4751
  api: "openai-completions",
4374
4752
  provider: "huggingface",
4375
4753
  baseUrl: "https://router.huggingface.co/v1",
@@ -4378,30 +4756,30 @@ export const MODELS = {
4378
4756
  input: ["text", "image"],
4379
4757
  cost: {
4380
4758
  input: 0.6,
4381
- output: 3,
4382
- cacheRead: 0.1,
4759
+ output: 1.8,
4760
+ cacheRead: 0,
4383
4761
  cacheWrite: 0,
4384
4762
  },
4385
- contextWindow: 262144,
4386
- maxTokens: 262144,
4763
+ contextWindow: 65536,
4764
+ maxTokens: 16384,
4387
4765
  },
4388
- "moonshotai/Kimi-K2.6": {
4389
- id: "moonshotai/Kimi-K2.6",
4390
- name: "Kimi-K2.6",
4766
+ "zai-org/GLM-4.6": {
4767
+ id: "zai-org/GLM-4.6",
4768
+ name: "GLM-4.6",
4391
4769
  api: "openai-completions",
4392
4770
  provider: "huggingface",
4393
4771
  baseUrl: "https://router.huggingface.co/v1",
4394
4772
  compat: { "supportsDeveloperRole": false },
4395
4773
  reasoning: true,
4396
- input: ["text", "image"],
4774
+ input: ["text"],
4397
4775
  cost: {
4398
- input: 0.95,
4399
- output: 4,
4400
- cacheRead: 0.16,
4776
+ input: 0.55,
4777
+ output: 2.2,
4778
+ cacheRead: 0,
4401
4779
  cacheWrite: 0,
4402
4780
  },
4403
- contextWindow: 262144,
4404
- maxTokens: 262144,
4781
+ contextWindow: 204800,
4782
+ maxTokens: 131072,
4405
4783
  },
4406
4784
  "zai-org/GLM-4.7": {
4407
4785
  id: "zai-org/GLM-4.7",
@@ -4475,6 +4853,24 @@ export const MODELS = {
4475
4853
  contextWindow: 202752,
4476
4854
  maxTokens: 131072,
4477
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,
4872
+ maxTokens: 131072,
4873
+ },
4478
4874
  },
4479
4875
  "kimi-coding": {
4480
4876
  "k2p5": {
@@ -6294,6 +6690,23 @@ export const MODELS = {
6294
6690
  contextWindow: 204800,
6295
6691
  maxTokens: 131072,
6296
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
+ },
6297
6710
  "gpt-5": {
6298
6711
  id: "gpt-5",
6299
6712
  name: "GPT-5",
@@ -7675,23 +8088,6 @@ export const MODELS = {
7675
8088
  contextWindow: 1048576,
7676
8089
  maxTokens: 384000,
7677
8090
  },
7678
- "essentialai/rnj-1-instruct": {
7679
- id: "essentialai/rnj-1-instruct",
7680
- name: "EssentialAI: Rnj 1 Instruct",
7681
- api: "openai-completions",
7682
- provider: "openrouter",
7683
- baseUrl: "https://openrouter.ai/api/v1",
7684
- reasoning: false,
7685
- input: ["text"],
7686
- cost: {
7687
- input: 0.15,
7688
- output: 0.15,
7689
- cacheRead: 0,
7690
- cacheWrite: 0,
7691
- },
7692
- contextWindow: 32768,
7693
- maxTokens: 4096,
7694
- },
7695
8091
  "google/gemini-2.5-flash": {
7696
8092
  id: "google/gemini-2.5-flash",
7697
8093
  name: "Google: Gemini 2.5 Flash",
@@ -8313,13 +8709,13 @@ export const MODELS = {
8313
8709
  reasoning: true,
8314
8710
  input: ["text"],
8315
8711
  cost: {
8316
- input: 0.25,
8317
- output: 1,
8318
- cacheRead: 0.049999999999999996,
8712
+ input: 0.24,
8713
+ output: 0.96,
8714
+ cacheRead: 0,
8319
8715
  cacheWrite: 0,
8320
8716
  },
8321
8717
  contextWindow: 204800,
8322
- maxTokens: 131072,
8718
+ maxTokens: 196608,
8323
8719
  },
8324
8720
  "minimax/minimax-m3": {
8325
8721
  id: "minimax/minimax-m3",
@@ -8672,7 +9068,7 @@ export const MODELS = {
8672
9068
  cost: {
8673
9069
  input: 0.6,
8674
9070
  output: 2.5,
8675
- cacheRead: 0,
9071
+ cacheRead: 0.6,
8676
9072
  cacheWrite: 0,
8677
9073
  },
8678
9074
  contextWindow: 262144,
@@ -8721,13 +9117,13 @@ export const MODELS = {
8721
9117
  reasoning: true,
8722
9118
  input: ["text", "image"],
8723
9119
  cost: {
8724
- input: 0.612,
8725
- output: 3.0690000000000004,
8726
- cacheRead: 0.1296,
9120
+ input: 0.74,
9121
+ output: 3.5,
9122
+ cacheRead: 0.15,
8727
9123
  cacheWrite: 0,
8728
9124
  },
8729
9125
  contextWindow: 262144,
8730
- maxTokens: 262144,
9126
+ maxTokens: 16384,
8731
9127
  },
8732
9128
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
8733
9129
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -9936,23 +10332,6 @@ export const MODELS = {
9936
10332
  contextWindow: 262144,
9937
10333
  maxTokens: 32768,
9938
10334
  },
9939
- "prime-intellect/intellect-3": {
9940
- id: "prime-intellect/intellect-3",
9941
- name: "Prime Intellect: INTELLECT-3",
9942
- api: "openai-completions",
9943
- provider: "openrouter",
9944
- baseUrl: "https://openrouter.ai/api/v1",
9945
- reasoning: true,
9946
- input: ["text"],
9947
- cost: {
9948
- input: 0.19999999999999998,
9949
- output: 1.1,
9950
- cacheRead: 0,
9951
- cacheWrite: 0,
9952
- },
9953
- contextWindow: 131072,
9954
- maxTokens: 131072,
9955
- },
9956
10335
  "qwen/qwen-2.5-72b-instruct": {
9957
10336
  id: "qwen/qwen-2.5-72b-instruct",
9958
10337
  name: "Qwen2.5 72B Instruct",
@@ -10542,11 +10921,11 @@ export const MODELS = {
10542
10921
  cost: {
10543
10922
  input: 0.14,
10544
10923
  output: 1,
10545
- cacheRead: 0,
10924
+ cacheRead: 0.049999999999999996,
10546
10925
  cacheWrite: 0,
10547
10926
  },
10548
10927
  contextWindow: 262144,
10549
- maxTokens: 262144,
10928
+ maxTokens: 81920,
10550
10929
  },
10551
10930
  "qwen/qwen3.5-397b-a17b": {
10552
10931
  id: "qwen/qwen3.5-397b-a17b",
@@ -10786,6 +11165,23 @@ export const MODELS = {
10786
11165
  contextWindow: 256000,
10787
11166
  maxTokens: 128000,
10788
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
+ },
10789
11185
  "sao10k/l3.1-euryale-70b": {
10790
11186
  id: "sao10k/l3.1-euryale-70b",
10791
11187
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -10854,23 +11250,6 @@ export const MODELS = {
10854
11250
  contextWindow: 262144,
10855
11251
  maxTokens: 4096,
10856
11252
  },
10857
- "thedrummer/rocinante-12b": {
10858
- id: "thedrummer/rocinante-12b",
10859
- name: "TheDrummer: Rocinante 12B",
10860
- api: "openai-completions",
10861
- provider: "openrouter",
10862
- baseUrl: "https://openrouter.ai/api/v1",
10863
- reasoning: false,
10864
- input: ["text"],
10865
- cost: {
10866
- input: 0.16999999999999998,
10867
- output: 0.43,
10868
- cacheRead: 0,
10869
- cacheWrite: 0,
10870
- },
10871
- contextWindow: 32768,
10872
- maxTokens: 32768,
10873
- },
10874
11253
  "thedrummer/unslopnemo-12b": {
10875
11254
  id: "thedrummer/unslopnemo-12b",
10876
11255
  name: "TheDrummer: UnslopNemo 12B",
@@ -10965,13 +11344,13 @@ export const MODELS = {
10965
11344
  reasoning: true,
10966
11345
  input: ["text", "image"],
10967
11346
  cost: {
10968
- input: 0.14,
11347
+ input: 0.105,
10969
11348
  output: 0.28,
10970
- cacheRead: 0.0028,
11349
+ cacheRead: 0,
10971
11350
  cacheWrite: 0,
10972
11351
  },
10973
11352
  contextWindow: 1048576,
10974
- maxTokens: 131072,
11353
+ maxTokens: 4096,
10975
11354
  },
10976
11355
  "xiaomi/mimo-v2.5-pro": {
10977
11356
  id: "xiaomi/mimo-v2.5-pro",
@@ -11154,11 +11533,11 @@ export const MODELS = {
11154
11533
  cost: {
11155
11534
  input: 0.98,
11156
11535
  output: 3.08,
11157
- cacheRead: 0.49,
11536
+ cacheRead: 0.182,
11158
11537
  cacheWrite: 0,
11159
11538
  },
11160
11539
  contextWindow: 202752,
11161
- maxTokens: 65535,
11540
+ maxTokens: 4096,
11162
11541
  },
11163
11542
  "z-ai/glm-5.2": {
11164
11543
  id: "z-ai/glm-5.2",
@@ -11169,13 +11548,13 @@ export const MODELS = {
11169
11548
  reasoning: true,
11170
11549
  input: ["text"],
11171
11550
  cost: {
11172
- input: 0.98,
11173
- output: 3.08,
11174
- cacheRead: 0.182,
11551
+ input: 0.95,
11552
+ output: 3,
11553
+ cacheRead: 0.18,
11175
11554
  cacheWrite: 0,
11176
11555
  },
11177
11556
  contextWindow: 1048576,
11178
- maxTokens: 4096,
11557
+ maxTokens: 32768,
11179
11558
  },
11180
11559
  "z-ai/glm-5v-turbo": {
11181
11560
  id: "z-ai/glm-5v-turbo",
@@ -11457,7 +11836,7 @@ export const MODELS = {
11457
11836
  api: "anthropic-messages",
11458
11837
  provider: "vercel-ai-gateway",
11459
11838
  baseUrl: "https://ai-gateway.vercel.sh",
11460
- reasoning: true,
11839
+ reasoning: false,
11461
11840
  input: ["text"],
11462
11841
  cost: {
11463
11842
  input: 1.5,
@@ -11474,7 +11853,7 @@ export const MODELS = {
11474
11853
  api: "anthropic-messages",
11475
11854
  provider: "vercel-ai-gateway",
11476
11855
  baseUrl: "https://ai-gateway.vercel.sh",
11477
- reasoning: true,
11856
+ reasoning: false,
11478
11857
  input: ["text"],
11479
11858
  cost: {
11480
11859
  input: 0.15,
@@ -11491,7 +11870,7 @@ export const MODELS = {
11491
11870
  api: "anthropic-messages",
11492
11871
  provider: "vercel-ai-gateway",
11493
11872
  baseUrl: "https://ai-gateway.vercel.sh",
11494
- reasoning: true,
11873
+ reasoning: false,
11495
11874
  input: ["text"],
11496
11875
  cost: {
11497
11876
  input: 0.5,
@@ -11604,6 +11983,40 @@ export const MODELS = {
11604
11983
  contextWindow: 131072,
11605
11984
  maxTokens: 32768,
11606
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
+ },
11607
12020
  "alibaba/qwen3-vl-thinking": {
11608
12021
  id: "alibaba/qwen3-vl-thinking",
11609
12022
  name: "Qwen3 VL 235B A22B Thinking",
@@ -11698,30 +12111,98 @@ export const MODELS = {
11698
12111
  reasoning: true,
11699
12112
  input: ["text"],
11700
12113
  cost: {
11701
- input: 1.25,
11702
- output: 3.75,
11703
- cacheRead: 0.25,
11704
- 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,
11705
12186
  },
11706
- contextWindow: 991000,
11707
- maxTokens: 64000,
12187
+ contextWindow: 128000,
12188
+ maxTokens: 8192,
11708
12189
  },
11709
- "alibaba/qwen3.7-plus": {
11710
- id: "alibaba/qwen3.7-plus",
11711
- name: "Qwen 3.7 Plus",
12190
+ "amazon/nova-pro": {
12191
+ id: "amazon/nova-pro",
12192
+ name: "Nova Pro",
11712
12193
  api: "anthropic-messages",
11713
12194
  provider: "vercel-ai-gateway",
11714
12195
  baseUrl: "https://ai-gateway.vercel.sh",
11715
- reasoning: true,
12196
+ reasoning: false,
11716
12197
  input: ["text", "image"],
11717
12198
  cost: {
11718
- input: 0.39999999999999997,
11719
- output: 1.5999999999999999,
11720
- cacheRead: 0.08,
11721
- cacheWrite: 0.5,
12199
+ input: 0.7999999999999999,
12200
+ output: 3.1999999999999997,
12201
+ cacheRead: 0,
12202
+ cacheWrite: 0,
11722
12203
  },
11723
- contextWindow: 1000000,
11724
- maxTokens: 64000,
12204
+ contextWindow: 300000,
12205
+ maxTokens: 8192,
11725
12206
  },
11726
12207
  "anthropic/claude-3-haiku": {
11727
12208
  id: "anthropic/claude-3-haiku",
@@ -11961,6 +12442,23 @@ export const MODELS = {
11961
12442
  contextWindow: 262100,
11962
12443
  maxTokens: 80000,
11963
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
+ },
11964
12462
  "bytedance/seed-1.6": {
11965
12463
  id: "bytedance/seed-1.6",
11966
12464
  name: "Seed 1.6",
@@ -11968,7 +12466,7 @@ export const MODELS = {
11968
12466
  provider: "vercel-ai-gateway",
11969
12467
  baseUrl: "https://ai-gateway.vercel.sh",
11970
12468
  reasoning: true,
11971
- input: ["text"],
12469
+ input: ["text", "image"],
11972
12470
  cost: {
11973
12471
  input: 0.25,
11974
12472
  output: 2,
@@ -11978,6 +12476,23 @@ export const MODELS = {
11978
12476
  contextWindow: 256000,
11979
12477
  maxTokens: 32000,
11980
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
+ },
11981
12496
  "cohere/command-a": {
11982
12497
  id: "cohere/command-a",
11983
12498
  name: "Command A",
@@ -12069,8 +12584,8 @@ export const MODELS = {
12069
12584
  api: "anthropic-messages",
12070
12585
  provider: "vercel-ai-gateway",
12071
12586
  baseUrl: "https://ai-gateway.vercel.sh",
12072
- reasoning: true,
12073
- input: ["text", "image"],
12587
+ reasoning: false,
12588
+ input: ["text"],
12074
12589
  cost: {
12075
12590
  input: 0.28,
12076
12591
  output: 0.42,
@@ -12087,7 +12602,7 @@ export const MODELS = {
12087
12602
  provider: "vercel-ai-gateway",
12088
12603
  baseUrl: "https://ai-gateway.vercel.sh",
12089
12604
  reasoning: true,
12090
- input: ["text", "image"],
12605
+ input: ["text"],
12091
12606
  cost: {
12092
12607
  input: 0.62,
12093
12608
  output: 1.85,
@@ -12104,7 +12619,7 @@ export const MODELS = {
12104
12619
  provider: "vercel-ai-gateway",
12105
12620
  baseUrl: "https://ai-gateway.vercel.sh",
12106
12621
  reasoning: true,
12107
- input: ["text", "image"],
12622
+ input: ["text"],
12108
12623
  cost: {
12109
12624
  input: 0.14,
12110
12625
  output: 0.28,
@@ -12307,7 +12822,7 @@ export const MODELS = {
12307
12822
  api: "anthropic-messages",
12308
12823
  provider: "vercel-ai-gateway",
12309
12824
  baseUrl: "https://ai-gateway.vercel.sh",
12310
- reasoning: false,
12825
+ reasoning: true,
12311
12826
  input: ["text", "image"],
12312
12827
  cost: {
12313
12828
  input: 0.14,
@@ -12352,6 +12867,40 @@ export const MODELS = {
12352
12867
  contextWindow: 32000,
12353
12868
  maxTokens: 16384,
12354
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
+ },
12355
12904
  "kwaipilot/kat-coder-pro-v2": {
12356
12905
  id: "kwaipilot/kat-coder-pro-v2",
12357
12906
  name: "Kat Coder Pro V2",
@@ -12386,6 +12935,23 @@ export const MODELS = {
12386
12935
  contextWindow: 128000,
12387
12936
  maxTokens: 100000,
12388
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
+ },
12389
12955
  "meta/llama-3.1-70b": {
12390
12956
  id: "meta/llama-3.1-70b",
12391
12957
  name: "Llama 3.1 70B Instruct",
@@ -12699,7 +13265,7 @@ export const MODELS = {
12699
13265
  provider: "vercel-ai-gateway",
12700
13266
  baseUrl: "https://ai-gateway.vercel.sh",
12701
13267
  reasoning: false,
12702
- input: ["text"],
13268
+ input: ["text", "image"],
12703
13269
  cost: {
12704
13270
  input: 0.09999999999999999,
12705
13271
  output: 0.3,
@@ -12709,6 +13275,57 @@ export const MODELS = {
12709
13275
  contextWindow: 256000,
12710
13276
  maxTokens: 256000,
12711
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
+ },
12712
13329
  "mistral/ministral-3b": {
12713
13330
  id: "mistral/ministral-3b",
12714
13331
  name: "Ministral 3B",
@@ -12743,6 +13360,23 @@ export const MODELS = {
12743
13360
  contextWindow: 128000,
12744
13361
  maxTokens: 4000,
12745
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
+ },
12746
13380
  "mistral/mistral-medium": {
12747
13381
  id: "mistral/mistral-medium",
12748
13382
  name: "Mistral Medium 3.1",
@@ -12767,7 +13401,7 @@ export const MODELS = {
12767
13401
  provider: "vercel-ai-gateway",
12768
13402
  baseUrl: "https://ai-gateway.vercel.sh",
12769
13403
  reasoning: true,
12770
- input: ["text"],
13404
+ input: ["text", "image"],
12771
13405
  cost: {
12772
13406
  input: 1.5,
12773
13407
  output: 7.5,
@@ -12947,6 +13581,23 @@ export const MODELS = {
12947
13581
  contextWindow: 262144,
12948
13582
  maxTokens: 32768,
12949
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
+ },
12950
13601
  "nvidia/nemotron-3-super-120b-a12b": {
12951
13602
  id: "nvidia/nemotron-3-super-120b-a12b",
12952
13603
  name: "NVIDIA Nemotron 3 Super 120B A12B",
@@ -13015,6 +13666,23 @@ export const MODELS = {
13015
13666
  contextWindow: 131072,
13016
13667
  maxTokens: 131072,
13017
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
+ },
13018
13686
  "openai/gpt-4-turbo": {
13019
13687
  id: "openai/gpt-4-turbo",
13020
13688
  name: "GPT-4 Turbo",
@@ -13140,7 +13808,7 @@ export const MODELS = {
13140
13808
  api: "anthropic-messages",
13141
13809
  provider: "vercel-ai-gateway",
13142
13810
  baseUrl: "https://ai-gateway.vercel.sh",
13143
- reasoning: true,
13811
+ reasoning: false,
13144
13812
  input: ["text", "image"],
13145
13813
  cost: {
13146
13814
  input: 1.25,
@@ -13276,7 +13944,7 @@ export const MODELS = {
13276
13944
  api: "anthropic-messages",
13277
13945
  provider: "vercel-ai-gateway",
13278
13946
  baseUrl: "https://ai-gateway.vercel.sh",
13279
- reasoning: true,
13947
+ reasoning: false,
13280
13948
  input: ["text", "image"],
13281
13949
  cost: {
13282
13950
  input: 1.25,
@@ -13327,7 +13995,7 @@ export const MODELS = {
13327
13995
  api: "anthropic-messages",
13328
13996
  provider: "vercel-ai-gateway",
13329
13997
  baseUrl: "https://ai-gateway.vercel.sh",
13330
- reasoning: true,
13998
+ reasoning: false,
13331
13999
  input: ["text", "image"],
13332
14000
  cost: {
13333
14001
  input: 1.75,
@@ -13378,7 +14046,7 @@ export const MODELS = {
13378
14046
  api: "anthropic-messages",
13379
14047
  provider: "vercel-ai-gateway",
13380
14048
  baseUrl: "https://ai-gateway.vercel.sh",
13381
- reasoning: true,
14049
+ reasoning: false,
13382
14050
  input: ["text", "image"],
13383
14051
  cost: {
13384
14052
  input: 1.75,
@@ -13661,40 +14329,6 @@ export const MODELS = {
13661
14329
  contextWindow: 200000,
13662
14330
  maxTokens: 100000,
13663
14331
  },
13664
- "perplexity/sonar": {
13665
- id: "perplexity/sonar",
13666
- name: "Sonar",
13667
- api: "anthropic-messages",
13668
- provider: "vercel-ai-gateway",
13669
- baseUrl: "https://ai-gateway.vercel.sh",
13670
- reasoning: false,
13671
- input: ["text", "image"],
13672
- cost: {
13673
- input: 0,
13674
- output: 0,
13675
- cacheRead: 0,
13676
- cacheWrite: 0,
13677
- },
13678
- contextWindow: 127000,
13679
- maxTokens: 8000,
13680
- },
13681
- "perplexity/sonar-pro": {
13682
- id: "perplexity/sonar-pro",
13683
- name: "Sonar Pro",
13684
- api: "anthropic-messages",
13685
- provider: "vercel-ai-gateway",
13686
- baseUrl: "https://ai-gateway.vercel.sh",
13687
- reasoning: false,
13688
- input: ["text", "image"],
13689
- cost: {
13690
- input: 0,
13691
- output: 0,
13692
- cacheRead: 0,
13693
- cacheWrite: 0,
13694
- },
13695
- contextWindow: 200000,
13696
- maxTokens: 8000,
13697
- },
13698
14332
  "sakana/fugu-ultra": {
13699
14333
  id: "sakana/fugu-ultra",
13700
14334
  name: "Fugu Ultra",
@@ -14178,7 +14812,7 @@ export const MODELS = {
14178
14812
  provider: "vercel-ai-gateway",
14179
14813
  baseUrl: "https://ai-gateway.vercel.sh",
14180
14814
  reasoning: true,
14181
- input: ["text", "image"],
14815
+ input: ["text"],
14182
14816
  cost: {
14183
14817
  input: 1.4,
14184
14818
  output: 4.4,
@@ -14205,6 +14839,23 @@ export const MODELS = {
14205
14839
  contextWindow: 1000000,
14206
14840
  maxTokens: 128000,
14207
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
+ },
14208
14859
  "zai/glm-5v-turbo": {
14209
14860
  id: "zai/glm-5v-turbo",
14210
14861
  name: "GLM 5V Turbo",