@elyracode/ai 0.9.9 → 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.
@@ -2812,30 +2812,13 @@ export const MODELS = {
2812
2812
  reasoning: true,
2813
2813
  input: ["text"],
2814
2814
  cost: {
2815
- input: 0.25,
2816
- output: 0.69,
2815
+ input: 0.35,
2816
+ output: 0.75,
2817
2817
  cacheRead: 0,
2818
2818
  cacheWrite: 0,
2819
2819
  },
2820
2820
  contextWindow: 131072,
2821
- maxTokens: 32768,
2822
- },
2823
- "llama3.1-8b": {
2824
- id: "llama3.1-8b",
2825
- name: "Llama 3.1 8B",
2826
- api: "openai-completions",
2827
- provider: "cerebras",
2828
- baseUrl: "https://api.cerebras.ai/v1",
2829
- reasoning: false,
2830
- input: ["text"],
2831
- cost: {
2832
- input: 0.1,
2833
- output: 0.1,
2834
- cacheRead: 0,
2835
- cacheWrite: 0,
2836
- },
2837
- contextWindow: 32000,
2838
- maxTokens: 8000,
2821
+ maxTokens: 40960,
2839
2822
  },
2840
2823
  "zai-glm-4.7": {
2841
2824
  id: "zai-glm-4.7",
@@ -2843,7 +2826,7 @@ export const MODELS = {
2843
2826
  api: "openai-completions",
2844
2827
  provider: "cerebras",
2845
2828
  baseUrl: "https://api.cerebras.ai/v1",
2846
- reasoning: false,
2829
+ reasoning: true,
2847
2830
  input: ["text"],
2848
2831
  cost: {
2849
2832
  input: 2.25,
@@ -2852,7 +2835,7 @@ export const MODELS = {
2852
2835
  cacheWrite: 0,
2853
2836
  },
2854
2837
  contextWindow: 131072,
2855
- maxTokens: 40000,
2838
+ maxTokens: 40960,
2856
2839
  },
2857
2840
  },
2858
2841
  "cloudflare-ai-gateway": {
@@ -3559,6 +3542,24 @@ export const MODELS = {
3559
3542
  contextWindow: 262144,
3560
3543
  maxTokens: 256000,
3561
3544
  },
3545
+ "@cf/moonshotai/kimi-k2.7-code": {
3546
+ id: "@cf/moonshotai/kimi-k2.7-code",
3547
+ name: "Kimi K2.7 Code",
3548
+ api: "openai-completions",
3549
+ provider: "cloudflare-workers-ai",
3550
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3551
+ compat: { "sendSessionAffinityHeaders": true },
3552
+ reasoning: true,
3553
+ input: ["text", "image"],
3554
+ cost: {
3555
+ input: 0.95,
3556
+ output: 4,
3557
+ cacheRead: 0.19,
3558
+ cacheWrite: 0,
3559
+ },
3560
+ contextWindow: 262144,
3561
+ maxTokens: 262144,
3562
+ },
3562
3563
  "@cf/nvidia/nemotron-3-120b-a12b": {
3563
3564
  id: "@cf/nvidia/nemotron-3-120b-a12b",
3564
3565
  name: "Nemotron 3 Super 120B",
@@ -3649,6 +3650,24 @@ export const MODELS = {
3649
3650
  contextWindow: 131072,
3650
3651
  maxTokens: 131072,
3651
3652
  },
3653
+ "@cf/zai-org/glm-5.2": {
3654
+ id: "@cf/zai-org/glm-5.2",
3655
+ name: "Glm 5.2",
3656
+ api: "openai-completions",
3657
+ provider: "cloudflare-workers-ai",
3658
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3659
+ compat: { "sendSessionAffinityHeaders": true },
3660
+ reasoning: true,
3661
+ input: ["text"],
3662
+ cost: {
3663
+ input: 1.4,
3664
+ output: 4.4,
3665
+ cacheRead: 0.26,
3666
+ cacheWrite: 0,
3667
+ },
3668
+ contextWindow: 262144,
3669
+ maxTokens: 262144,
3670
+ },
3652
3671
  },
3653
3672
  "deepseek": {
3654
3673
  "deepseek-v4-flash": {
@@ -3702,7 +3721,7 @@ export const MODELS = {
3702
3721
  cost: {
3703
3722
  input: 0.14,
3704
3723
  output: 0.28,
3705
- cacheRead: 0.03,
3724
+ cacheRead: 0.028,
3706
3725
  cacheWrite: 0,
3707
3726
  },
3708
3727
  contextWindow: 1000000,
@@ -3742,6 +3761,23 @@ export const MODELS = {
3742
3761
  contextWindow: 202800,
3743
3762
  maxTokens: 131072,
3744
3763
  },
3764
+ "accounts/fireworks/models/glm-5p2": {
3765
+ id: "accounts/fireworks/models/glm-5p2",
3766
+ name: "GLM 5.2",
3767
+ api: "anthropic-messages",
3768
+ provider: "fireworks",
3769
+ baseUrl: "https://api.fireworks.ai/inference",
3770
+ reasoning: true,
3771
+ input: ["text"],
3772
+ cost: {
3773
+ input: 1.4,
3774
+ output: 4.4,
3775
+ cacheRead: 0.26,
3776
+ cacheWrite: 0,
3777
+ },
3778
+ contextWindow: 1048575,
3779
+ maxTokens: 131072,
3780
+ },
3745
3781
  "accounts/fireworks/models/gpt-oss-120b": {
3746
3782
  id: "accounts/fireworks/models/gpt-oss-120b",
3747
3783
  name: "GPT OSS 120B",
@@ -3776,26 +3812,26 @@ export const MODELS = {
3776
3812
  contextWindow: 131072,
3777
3813
  maxTokens: 32768,
3778
3814
  },
3779
- "accounts/fireworks/models/kimi-k2p5": {
3780
- id: "accounts/fireworks/models/kimi-k2p5",
3781
- name: "Kimi K2.5",
3815
+ "accounts/fireworks/models/kimi-k2p6": {
3816
+ id: "accounts/fireworks/models/kimi-k2p6",
3817
+ name: "Kimi K2.6",
3782
3818
  api: "anthropic-messages",
3783
3819
  provider: "fireworks",
3784
3820
  baseUrl: "https://api.fireworks.ai/inference",
3785
3821
  reasoning: true,
3786
3822
  input: ["text", "image"],
3787
3823
  cost: {
3788
- input: 0.6,
3789
- output: 3,
3790
- cacheRead: 0.1,
3824
+ input: 0.95,
3825
+ output: 4,
3826
+ cacheRead: 0.16,
3791
3827
  cacheWrite: 0,
3792
3828
  },
3793
- contextWindow: 256000,
3794
- maxTokens: 256000,
3829
+ contextWindow: 262000,
3830
+ maxTokens: 262000,
3795
3831
  },
3796
- "accounts/fireworks/models/kimi-k2p6": {
3797
- id: "accounts/fireworks/models/kimi-k2p6",
3798
- name: "Kimi K2.6",
3832
+ "accounts/fireworks/models/kimi-k2p7-code": {
3833
+ id: "accounts/fireworks/models/kimi-k2p7-code",
3834
+ name: "Kimi K2.7 Code",
3799
3835
  api: "anthropic-messages",
3800
3836
  provider: "fireworks",
3801
3837
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3804,15 +3840,15 @@ export const MODELS = {
3804
3840
  cost: {
3805
3841
  input: 0.95,
3806
3842
  output: 4,
3807
- cacheRead: 0.16,
3843
+ cacheRead: 0.19,
3808
3844
  cacheWrite: 0,
3809
3845
  },
3810
3846
  contextWindow: 262000,
3811
3847
  maxTokens: 262000,
3812
3848
  },
3813
- "accounts/fireworks/models/minimax-m2p5": {
3814
- id: "accounts/fireworks/models/minimax-m2p5",
3815
- name: "MiniMax-M2.5",
3849
+ "accounts/fireworks/models/minimax-m2p7": {
3850
+ id: "accounts/fireworks/models/minimax-m2p7",
3851
+ name: "MiniMax-M2.7",
3816
3852
  api: "anthropic-messages",
3817
3853
  provider: "fireworks",
3818
3854
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3821,15 +3857,15 @@ export const MODELS = {
3821
3857
  cost: {
3822
3858
  input: 0.3,
3823
3859
  output: 1.2,
3824
- cacheRead: 0.03,
3860
+ cacheRead: 0.06,
3825
3861
  cacheWrite: 0,
3826
3862
  },
3827
3863
  contextWindow: 196608,
3828
3864
  maxTokens: 196608,
3829
3865
  },
3830
- "accounts/fireworks/models/minimax-m2p7": {
3831
- id: "accounts/fireworks/models/minimax-m2p7",
3832
- name: "MiniMax-M2.7",
3866
+ "accounts/fireworks/models/minimax-m3": {
3867
+ id: "accounts/fireworks/models/minimax-m3",
3868
+ name: "MiniMax-M3",
3833
3869
  api: "anthropic-messages",
3834
3870
  provider: "fireworks",
3835
3871
  baseUrl: "https://api.fireworks.ai/inference",
@@ -3841,21 +3877,21 @@ export const MODELS = {
3841
3877
  cacheRead: 0.06,
3842
3878
  cacheWrite: 0,
3843
3879
  },
3844
- contextWindow: 196608,
3845
- maxTokens: 196608,
3880
+ contextWindow: 512000,
3881
+ maxTokens: 512000,
3846
3882
  },
3847
- "accounts/fireworks/models/qwen3p6-plus": {
3848
- id: "accounts/fireworks/models/qwen3p6-plus",
3849
- name: "Qwen 3.6 Plus",
3883
+ "accounts/fireworks/models/qwen3p7-plus": {
3884
+ id: "accounts/fireworks/models/qwen3p7-plus",
3885
+ name: "Qwen 3.7 Plus",
3850
3886
  api: "anthropic-messages",
3851
3887
  provider: "fireworks",
3852
3888
  baseUrl: "https://api.fireworks.ai/inference",
3853
3889
  reasoning: true,
3854
3890
  input: ["text", "image"],
3855
3891
  cost: {
3856
- input: 0.5,
3857
- output: 3,
3858
- cacheRead: 0.1,
3892
+ input: 0.4,
3893
+ output: 1.6,
3894
+ cacheRead: 0.08,
3859
3895
  cacheWrite: 0,
3860
3896
  },
3861
3897
  contextWindow: 262144,
@@ -3878,6 +3914,23 @@ export const MODELS = {
3878
3914
  contextWindow: 202800,
3879
3915
  maxTokens: 131072,
3880
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
+ },
3881
3934
  "accounts/fireworks/routers/kimi-k2p6-fast": {
3882
3935
  id: "accounts/fireworks/routers/kimi-k2p6-fast",
3883
3936
  name: "Kimi K2.6 Fast",
@@ -3912,8 +3965,45 @@ export const MODELS = {
3912
3965
  contextWindow: 262000,
3913
3966
  maxTokens: 262000,
3914
3967
  },
3968
+ "accounts/fireworks/routers/kimi-k2p7-code-fast": {
3969
+ id: "accounts/fireworks/routers/kimi-k2p7-code-fast",
3970
+ name: "Kimi K2.7 Code Fast",
3971
+ api: "anthropic-messages",
3972
+ provider: "fireworks",
3973
+ baseUrl: "https://api.fireworks.ai/inference",
3974
+ reasoning: true,
3975
+ input: ["text", "image"],
3976
+ cost: {
3977
+ input: 1.9,
3978
+ output: 8,
3979
+ cacheRead: 0.38,
3980
+ cacheWrite: 0,
3981
+ },
3982
+ contextWindow: 262000,
3983
+ maxTokens: 262000,
3984
+ },
3915
3985
  },
3916
3986
  "github-copilot": {
3987
+ "claude-fable-5": {
3988
+ id: "claude-fable-5",
3989
+ name: "Claude Fable 5",
3990
+ api: "openai-completions",
3991
+ provider: "github-copilot",
3992
+ baseUrl: "https://api.individual.githubcopilot.com",
3993
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
3994
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3995
+ reasoning: true,
3996
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
3997
+ input: ["text", "image"],
3998
+ cost: {
3999
+ input: 10,
4000
+ output: 50,
4001
+ cacheRead: 1,
4002
+ cacheWrite: 12.5,
4003
+ },
4004
+ contextWindow: 1000000,
4005
+ maxTokens: 128000,
4006
+ },
3917
4007
  "claude-haiku-4.5": {
3918
4008
  id: "claude-haiku-4.5",
3919
4009
  name: "Claude Haiku 4.5 (latest)",
@@ -4311,25 +4401,6 @@ export const MODELS = {
4311
4401
  contextWindow: 400000,
4312
4402
  maxTokens: 128000,
4313
4403
  },
4314
- "raptor-mini": {
4315
- id: "raptor-mini",
4316
- name: "Raptor mini",
4317
- api: "openai-completions",
4318
- provider: "github-copilot",
4319
- baseUrl: "https://api.individual.githubcopilot.com",
4320
- headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4321
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4322
- reasoning: true,
4323
- input: ["text", "image"],
4324
- cost: {
4325
- input: 0.25,
4326
- output: 2,
4327
- cacheRead: 0.025,
4328
- cacheWrite: 0,
4329
- },
4330
- contextWindow: 400000,
4331
- maxTokens: 128000,
4332
- },
4333
4404
  },
4334
4405
  "google": {
4335
4406
  "gemini-2.0-flash": {
@@ -4964,6 +5035,24 @@ export const MODELS = {
4964
5035
  },
4965
5036
  },
4966
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
+ },
4967
5056
  "MiniMaxAI/MiniMax-M2.1": {
4968
5057
  id: "MiniMaxAI/MiniMax-M2.1",
4969
5058
  name: "MiniMax-M2.1",
@@ -5018,6 +5107,42 @@ export const MODELS = {
5018
5107
  contextWindow: 204800,
5019
5108
  maxTokens: 131072,
5020
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
+ },
5021
5146
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {
5022
5147
  id: "Qwen/Qwen3-235B-A22B-Thinking-2507",
5023
5148
  name: "Qwen3-235B-A22B-Thinking-2507",
@@ -5036,6 +5161,42 @@ export const MODELS = {
5036
5161
  contextWindow: 262144,
5037
5162
  maxTokens: 131072,
5038
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
+ },
5039
5200
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {
5040
5201
  id: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
5041
5202
  name: "Qwen3-Coder-480B-A35B-Instruct",
@@ -5108,9 +5269,9 @@ export const MODELS = {
5108
5269
  contextWindow: 262144,
5109
5270
  maxTokens: 131072,
5110
5271
  },
5111
- "Qwen/Qwen3.5-397B-A17B": {
5112
- id: "Qwen/Qwen3.5-397B-A17B",
5113
- name: "Qwen3.5-397B-A17B",
5272
+ "Qwen/Qwen3.5-122B-A10B": {
5273
+ id: "Qwen/Qwen3.5-122B-A10B",
5274
+ name: "Qwen3.5 122B-A10B",
5114
5275
  api: "openai-completions",
5115
5276
  provider: "huggingface",
5116
5277
  baseUrl: "https://router.huggingface.co/v1",
@@ -5118,25 +5279,169 @@ export const MODELS = {
5118
5279
  reasoning: true,
5119
5280
  input: ["text", "image"],
5120
5281
  cost: {
5121
- input: 0.6,
5122
- output: 3.6,
5282
+ input: 0.4,
5283
+ output: 3.2,
5123
5284
  cacheRead: 0,
5124
5285
  cacheWrite: 0,
5125
5286
  },
5126
5287
  contextWindow: 262144,
5127
- maxTokens: 32768,
5288
+ maxTokens: 65536,
5128
5289
  },
5129
- "deepseek-ai/DeepSeek-R1-0528": {
5130
- id: "deepseek-ai/DeepSeek-R1-0528",
5131
- name: "DeepSeek-R1-0528",
5290
+ "Qwen/Qwen3.5-27B": {
5291
+ id: "Qwen/Qwen3.5-27B",
5292
+ name: "Qwen3.5 27B",
5132
5293
  api: "openai-completions",
5133
5294
  provider: "huggingface",
5134
5295
  baseUrl: "https://router.huggingface.co/v1",
5135
5296
  compat: { "supportsDeveloperRole": false },
5136
5297
  reasoning: true,
5137
- input: ["text"],
5298
+ input: ["text", "image"],
5138
5299
  cost: {
5139
- input: 3,
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
+ },
5326
+ "Qwen/Qwen3.5-397B-A17B": {
5327
+ id: "Qwen/Qwen3.5-397B-A17B",
5328
+ name: "Qwen3.5-397B-A17B",
5329
+ api: "openai-completions",
5330
+ provider: "huggingface",
5331
+ baseUrl: "https://router.huggingface.co/v1",
5332
+ compat: { "supportsDeveloperRole": false },
5333
+ reasoning: true,
5334
+ input: ["text", "image"],
5335
+ cost: {
5336
+ input: 0.6,
5337
+ output: 3.6,
5338
+ cacheRead: 0,
5339
+ cacheWrite: 0,
5340
+ },
5341
+ contextWindow: 262144,
5342
+ maxTokens: 32768,
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
+ },
5434
+ "deepseek-ai/DeepSeek-R1-0528": {
5435
+ id: "deepseek-ai/DeepSeek-R1-0528",
5436
+ name: "DeepSeek-R1-0528",
5437
+ api: "openai-completions",
5438
+ provider: "huggingface",
5439
+ baseUrl: "https://router.huggingface.co/v1",
5440
+ compat: { "supportsDeveloperRole": false },
5441
+ reasoning: true,
5442
+ input: ["text"],
5443
+ cost: {
5444
+ input: 3,
5140
5445
  output: 5,
5141
5446
  cacheRead: 0,
5142
5447
  cacheWrite: 0,
@@ -5162,6 +5467,24 @@ export const MODELS = {
5162
5467
  contextWindow: 163840,
5163
5468
  maxTokens: 65536,
5164
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
+ },
5165
5488
  "deepseek-ai/DeepSeek-V4-Pro": {
5166
5489
  id: "deepseek-ai/DeepSeek-V4-Pro",
5167
5490
  name: "DeepSeek V4 Pro",
@@ -5180,6 +5503,60 @@ export const MODELS = {
5180
5503
  contextWindow: 1048576,
5181
5504
  maxTokens: 393216,
5182
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
+ },
5183
5560
  "moonshotai/Kimi-K2-Instruct": {
5184
5561
  id: "moonshotai/Kimi-K2-Instruct",
5185
5562
  name: "Kimi-K2-Instruct",
@@ -5270,6 +5647,132 @@ export const MODELS = {
5270
5647
  contextWindow: 262144,
5271
5648
  maxTokens: 262144,
5272
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
+ },
5273
5776
  "zai-org/GLM-4.7": {
5274
5777
  id: "zai-org/GLM-4.7",
5275
5778
  name: "GLM-4.7",
@@ -5342,8 +5845,44 @@ export const MODELS = {
5342
5845
  contextWindow: 202752,
5343
5846
  maxTokens: 131072,
5344
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
+ },
5345
5866
  },
5346
5867
  "kimi-coding": {
5868
+ "k2p7": {
5869
+ id: "k2p7",
5870
+ name: "Kimi K2.7 Code",
5871
+ api: "anthropic-messages",
5872
+ provider: "kimi-coding",
5873
+ baseUrl: "https://api.kimi.com/coding",
5874
+ headers: { "User-Agent": "KimiCLI/1.5" },
5875
+ reasoning: true,
5876
+ input: ["text", "image"],
5877
+ cost: {
5878
+ input: 0,
5879
+ output: 0,
5880
+ cacheRead: 0,
5881
+ cacheWrite: 0,
5882
+ },
5883
+ contextWindow: 262144,
5884
+ maxTokens: 32768,
5885
+ },
5347
5886
  "kimi-for-coding": {
5348
5887
  id: "kimi-for-coding",
5349
5888
  name: "Kimi For Coding",
@@ -5425,12 +5964,12 @@ export const MODELS = {
5425
5964
  reasoning: true,
5426
5965
  input: ["text", "image"],
5427
5966
  cost: {
5428
- input: 0.6,
5429
- output: 2.4,
5430
- cacheRead: 0.12,
5967
+ input: 0.3,
5968
+ output: 1.2,
5969
+ cacheRead: 0.06,
5431
5970
  cacheWrite: 0,
5432
5971
  },
5433
- contextWindow: 512000,
5972
+ contextWindow: 1000000,
5434
5973
  maxTokens: 128000,
5435
5974
  },
5436
5975
  },
@@ -5478,12 +6017,12 @@ export const MODELS = {
5478
6017
  reasoning: true,
5479
6018
  input: ["text", "image"],
5480
6019
  cost: {
5481
- input: 0.6,
5482
- output: 2.4,
5483
- cacheRead: 0.12,
6020
+ input: 0.3,
6021
+ output: 1.2,
6022
+ cacheRead: 0.06,
5484
6023
  cacheWrite: 0,
5485
6024
  },
5486
- contextWindow: 512000,
6025
+ contextWindow: 1000000,
5487
6026
  maxTokens: 128000,
5488
6027
  },
5489
6028
  },
@@ -6126,6 +6665,42 @@ export const MODELS = {
6126
6665
  contextWindow: 262144,
6127
6666
  maxTokens: 262144,
6128
6667
  },
6668
+ "kimi-k2.7-code": {
6669
+ id: "kimi-k2.7-code",
6670
+ name: "Kimi K2.7 Code",
6671
+ api: "openai-completions",
6672
+ provider: "moonshotai",
6673
+ baseUrl: "https://api.moonshot.ai/v1",
6674
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6675
+ reasoning: true,
6676
+ input: ["text", "image"],
6677
+ cost: {
6678
+ input: 0.95,
6679
+ output: 4,
6680
+ cacheRead: 0.19,
6681
+ cacheWrite: 0,
6682
+ },
6683
+ contextWindow: 262144,
6684
+ maxTokens: 262144,
6685
+ },
6686
+ "kimi-k2.7-code-highspeed": {
6687
+ id: "kimi-k2.7-code-highspeed",
6688
+ name: "Kimi K2.7 Code HighSpeed",
6689
+ api: "openai-completions",
6690
+ provider: "moonshotai",
6691
+ baseUrl: "https://api.moonshot.ai/v1",
6692
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6693
+ reasoning: true,
6694
+ input: ["text", "image"],
6695
+ cost: {
6696
+ input: 1.9,
6697
+ output: 8,
6698
+ cacheRead: 0.38,
6699
+ cacheWrite: 0,
6700
+ },
6701
+ contextWindow: 262144,
6702
+ maxTokens: 262144,
6703
+ },
6129
6704
  },
6130
6705
  "moonshotai-cn": {
6131
6706
  "kimi-k2-0711-preview": {
@@ -6254,6 +6829,42 @@ export const MODELS = {
6254
6829
  contextWindow: 262144,
6255
6830
  maxTokens: 262144,
6256
6831
  },
6832
+ "kimi-k2.7-code": {
6833
+ id: "kimi-k2.7-code",
6834
+ name: "Kimi K2.7 Code",
6835
+ api: "openai-completions",
6836
+ provider: "moonshotai-cn",
6837
+ baseUrl: "https://api.moonshot.cn/v1",
6838
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6839
+ reasoning: true,
6840
+ input: ["text", "image"],
6841
+ cost: {
6842
+ input: 0.95,
6843
+ output: 4,
6844
+ cacheRead: 0.19,
6845
+ cacheWrite: 0,
6846
+ },
6847
+ contextWindow: 262144,
6848
+ maxTokens: 262144,
6849
+ },
6850
+ "kimi-k2.7-code-highspeed": {
6851
+ id: "kimi-k2.7-code-highspeed",
6852
+ name: "Kimi K2.7 Code HighSpeed",
6853
+ api: "openai-completions",
6854
+ provider: "moonshotai-cn",
6855
+ baseUrl: "https://api.moonshot.cn/v1",
6856
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6857
+ reasoning: true,
6858
+ input: ["text", "image"],
6859
+ cost: {
6860
+ input: 1.9,
6861
+ output: 8,
6862
+ cacheRead: 0.38,
6863
+ cacheWrite: 0,
6864
+ },
6865
+ contextWindow: 262144,
6866
+ maxTokens: 262144,
6867
+ },
6257
6868
  },
6258
6869
  "openai": {
6259
6870
  "gpt-4": {
@@ -7125,24 +7736,6 @@ export const MODELS = {
7125
7736
  contextWindow: 200000,
7126
7737
  maxTokens: 32000,
7127
7738
  },
7128
- "claude-fable-5": {
7129
- id: "claude-fable-5",
7130
- name: "Claude Fable 5",
7131
- api: "anthropic-messages",
7132
- provider: "opencode",
7133
- baseUrl: "https://opencode.ai/zen",
7134
- reasoning: true,
7135
- thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
7136
- input: ["text", "image"],
7137
- cost: {
7138
- input: 10,
7139
- output: 50,
7140
- cacheRead: 1,
7141
- cacheWrite: 12.5,
7142
- },
7143
- contextWindow: 1000000,
7144
- maxTokens: 128000,
7145
- },
7146
7739
  "claude-haiku-4-5": {
7147
7740
  id: "claude-haiku-4-5",
7148
7741
  name: "Claude Haiku 4.5",
@@ -7444,6 +8037,23 @@ export const MODELS = {
7444
8037
  contextWindow: 204800,
7445
8038
  maxTokens: 131072,
7446
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
+ },
7447
8057
  "gpt-5": {
7448
8058
  id: "gpt-5",
7449
8059
  name: "GPT-5",
@@ -7942,26 +8552,26 @@ export const MODELS = {
7942
8552
  contextWindow: 1000000,
7943
8553
  maxTokens: 384000,
7944
8554
  },
7945
- "glm-5": {
7946
- id: "glm-5",
7947
- name: "GLM-5",
8555
+ "glm-5.1": {
8556
+ id: "glm-5.1",
8557
+ name: "GLM-5.1",
7948
8558
  api: "openai-completions",
7949
8559
  provider: "opencode-go",
7950
8560
  baseUrl: "https://opencode.ai/zen/go/v1",
7951
8561
  reasoning: true,
7952
8562
  input: ["text"],
7953
8563
  cost: {
7954
- input: 1,
7955
- output: 3.2,
7956
- cacheRead: 0.2,
8564
+ input: 1.4,
8565
+ output: 4.4,
8566
+ cacheRead: 0.26,
7957
8567
  cacheWrite: 0,
7958
8568
  },
7959
8569
  contextWindow: 202752,
7960
8570
  maxTokens: 32768,
7961
8571
  },
7962
- "glm-5.1": {
7963
- id: "glm-5.1",
7964
- name: "GLM-5.1",
8572
+ "glm-5.2": {
8573
+ id: "glm-5.2",
8574
+ name: "GLM-5.2",
7965
8575
  api: "openai-completions",
7966
8576
  provider: "opencode-go",
7967
8577
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -7973,29 +8583,29 @@ export const MODELS = {
7973
8583
  cacheRead: 0.26,
7974
8584
  cacheWrite: 0,
7975
8585
  },
7976
- contextWindow: 202752,
7977
- maxTokens: 32768,
8586
+ contextWindow: 1000000,
8587
+ maxTokens: 131072,
7978
8588
  },
7979
- "kimi-k2.5": {
7980
- id: "kimi-k2.5",
7981
- name: "Kimi K2.5",
8589
+ "kimi-k2.6": {
8590
+ id: "kimi-k2.6",
8591
+ name: "Kimi K2.6",
7982
8592
  api: "openai-completions",
7983
8593
  provider: "opencode-go",
7984
8594
  baseUrl: "https://opencode.ai/zen/go/v1",
7985
8595
  reasoning: true,
7986
8596
  input: ["text", "image"],
7987
8597
  cost: {
7988
- input: 0.6,
7989
- output: 3,
7990
- cacheRead: 0.1,
8598
+ input: 0.95,
8599
+ output: 4,
8600
+ cacheRead: 0.16,
7991
8601
  cacheWrite: 0,
7992
8602
  },
7993
8603
  contextWindow: 262144,
7994
8604
  maxTokens: 65536,
7995
8605
  },
7996
- "kimi-k2.6": {
7997
- id: "kimi-k2.6",
7998
- name: "Kimi K2.6",
8606
+ "kimi-k2.7-code": {
8607
+ id: "kimi-k2.7-code",
8608
+ name: "Kimi K2.7 Code",
7999
8609
  api: "openai-completions",
8000
8610
  provider: "opencode-go",
8001
8611
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8004,11 +8614,11 @@ export const MODELS = {
8004
8614
  cost: {
8005
8615
  input: 0.95,
8006
8616
  output: 4,
8007
- cacheRead: 0.16,
8617
+ cacheRead: 0.19,
8008
8618
  cacheWrite: 0,
8009
8619
  },
8010
8620
  contextWindow: 262144,
8011
- maxTokens: 65536,
8621
+ maxTokens: 262144,
8012
8622
  },
8013
8623
  "mimo-v2.5": {
8014
8624
  id: "mimo-v2.5",
@@ -8044,23 +8654,6 @@ export const MODELS = {
8044
8654
  contextWindow: 1048576,
8045
8655
  maxTokens: 128000,
8046
8656
  },
8047
- "minimax-m2.5": {
8048
- id: "minimax-m2.5",
8049
- name: "MiniMax M2.5",
8050
- api: "anthropic-messages",
8051
- provider: "opencode-go",
8052
- baseUrl: "https://opencode.ai/zen/go",
8053
- reasoning: true,
8054
- input: ["text"],
8055
- cost: {
8056
- input: 0.3,
8057
- output: 1.2,
8058
- cacheRead: 0.03,
8059
- cacheWrite: 0,
8060
- },
8061
- contextWindow: 204800,
8062
- maxTokens: 65536,
8063
- },
8064
8657
  "minimax-m2.7": {
8065
8658
  id: "minimax-m2.7",
8066
8659
  name: "MiniMax M2.7",
@@ -8080,19 +8673,19 @@ export const MODELS = {
8080
8673
  },
8081
8674
  "minimax-m3": {
8082
8675
  id: "minimax-m3",
8083
- name: "MiniMax M3",
8676
+ name: "MiniMax M3 (3x usage)",
8084
8677
  api: "anthropic-messages",
8085
8678
  provider: "opencode-go",
8086
8679
  baseUrl: "https://opencode.ai/zen/go",
8087
8680
  reasoning: true,
8088
8681
  input: ["text", "image"],
8089
8682
  cost: {
8090
- input: 0.3,
8091
- output: 1.2,
8092
- cacheRead: 0.06,
8683
+ input: 0.1,
8684
+ output: 0.4,
8685
+ cacheRead: 0.02,
8093
8686
  cacheWrite: 0,
8094
8687
  },
8095
- contextWindow: 512000,
8688
+ contextWindow: 1000000,
8096
8689
  maxTokens: 131072,
8097
8690
  },
8098
8691
  "qwen3.6-plus": {
@@ -8522,13 +9115,13 @@ export const MODELS = {
8522
9115
  reasoning: true,
8523
9116
  input: ["text"],
8524
9117
  cost: {
8525
- input: 0.22,
8526
- output: 0.85,
9118
+ input: 0.25,
9119
+ output: 0.7999999999999999,
8527
9120
  cacheRead: 0.06,
8528
9121
  cacheWrite: 0,
8529
9122
  },
8530
9123
  contextWindow: 262144,
8531
- maxTokens: 262144,
9124
+ maxTokens: 80000,
8532
9125
  },
8533
9126
  "arcee-ai/trinity-mini": {
8534
9127
  id: "arcee-ai/trinity-mini",
@@ -8683,6 +9276,23 @@ export const MODELS = {
8683
9276
  contextWindow: 128000,
8684
9277
  maxTokens: 4000,
8685
9278
  },
9279
+ "cohere/north-mini-code:free": {
9280
+ id: "cohere/north-mini-code:free",
9281
+ name: "Cohere: North Mini Code (free)",
9282
+ api: "openai-completions",
9283
+ provider: "openrouter",
9284
+ baseUrl: "https://openrouter.ai/api/v1",
9285
+ reasoning: true,
9286
+ input: ["text"],
9287
+ cost: {
9288
+ input: 0,
9289
+ output: 0,
9290
+ cacheRead: 0,
9291
+ cacheWrite: 0,
9292
+ },
9293
+ contextWindow: 256000,
9294
+ maxTokens: 64000,
9295
+ },
8686
9296
  "deepseek/deepseek-chat": {
8687
9297
  id: "deepseek/deepseek-chat",
8688
9298
  name: "DeepSeek: DeepSeek V3",
@@ -8714,7 +9324,7 @@ export const MODELS = {
8714
9324
  cacheRead: 0.135,
8715
9325
  cacheWrite: 0,
8716
9326
  },
8717
- contextWindow: 131072,
9327
+ contextWindow: 163840,
8718
9328
  maxTokens: 16384,
8719
9329
  },
8720
9330
  "deepseek/deepseek-chat-v3.1": {
@@ -8796,7 +9406,7 @@ export const MODELS = {
8796
9406
  cost: {
8797
9407
  input: 0.2288,
8798
9408
  output: 0.3432,
8799
- cacheRead: 0,
9409
+ cacheRead: 0.02288,
8800
9410
  cacheWrite: 0,
8801
9411
  },
8802
9412
  contextWindow: 131072,
@@ -8830,13 +9440,13 @@ export const MODELS = {
8830
9440
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8831
9441
  input: ["text"],
8832
9442
  cost: {
8833
- input: 0.0983,
8834
- output: 0.1966,
8835
- cacheRead: 0.019700000000000002,
9443
+ input: 0.09,
9444
+ output: 0.18,
9445
+ cacheRead: 0.02,
8836
9446
  cacheWrite: 0,
8837
9447
  },
8838
9448
  contextWindow: 1048576,
8839
- maxTokens: 131072,
9449
+ maxTokens: 65536,
8840
9450
  },
8841
9451
  "deepseek/deepseek-v4-pro": {
8842
9452
  id: "deepseek/deepseek-v4-pro",
@@ -8857,23 +9467,6 @@ export const MODELS = {
8857
9467
  contextWindow: 1048576,
8858
9468
  maxTokens: 384000,
8859
9469
  },
8860
- "essentialai/rnj-1-instruct": {
8861
- id: "essentialai/rnj-1-instruct",
8862
- name: "EssentialAI: Rnj 1 Instruct",
8863
- api: "openai-completions",
8864
- provider: "openrouter",
8865
- baseUrl: "https://openrouter.ai/api/v1",
8866
- reasoning: false,
8867
- input: ["text"],
8868
- cost: {
8869
- input: 0.15,
8870
- output: 0.15,
8871
- cacheRead: 0,
8872
- cacheWrite: 0,
8873
- },
8874
- contextWindow: 32768,
8875
- maxTokens: 4096,
8876
- },
8877
9470
  "google/gemini-2.5-flash": {
8878
9471
  id: "google/gemini-2.5-flash",
8879
9472
  name: "Google: Gemini 2.5 Flash",
@@ -8991,7 +9584,24 @@ export const MODELS = {
8991
9584
  cacheWrite: 0.08333333333333334,
8992
9585
  },
8993
9586
  contextWindow: 1048576,
8994
- maxTokens: 65536,
9587
+ maxTokens: 65535,
9588
+ },
9589
+ "google/gemini-3-pro-image": {
9590
+ id: "google/gemini-3-pro-image",
9591
+ name: "Google: Nano Banana Pro (Gemini 3 Pro Image)",
9592
+ api: "openai-completions",
9593
+ provider: "openrouter",
9594
+ baseUrl: "https://openrouter.ai/api/v1",
9595
+ reasoning: true,
9596
+ input: ["text", "image"],
9597
+ cost: {
9598
+ input: 2,
9599
+ output: 12,
9600
+ cacheRead: 0.19999999999999998,
9601
+ cacheWrite: 0.375,
9602
+ },
9603
+ contextWindow: 65536,
9604
+ maxTokens: 32768,
8995
9605
  },
8996
9606
  "google/gemini-3.1-flash-lite": {
8997
9607
  id: "google/gemini-3.1-flash-lite",
@@ -9156,12 +9766,12 @@ export const MODELS = {
9156
9766
  input: ["text", "image"],
9157
9767
  cost: {
9158
9768
  input: 0.12,
9159
- output: 0.36,
9769
+ output: 0.35,
9160
9770
  cacheRead: 0.09,
9161
9771
  cacheWrite: 0,
9162
9772
  },
9163
9773
  contextWindow: 262144,
9164
- maxTokens: 8192,
9774
+ maxTokens: 262144,
9165
9775
  },
9166
9776
  "google/gemma-4-31b-it:free": {
9167
9777
  id: "google/gemma-4-31b-it:free",
@@ -9178,7 +9788,7 @@ export const MODELS = {
9178
9788
  cacheWrite: 0,
9179
9789
  },
9180
9790
  contextWindow: 262144,
9181
- maxTokens: 32768,
9791
+ maxTokens: 8192,
9182
9792
  },
9183
9793
  "ibm-granite/granite-4.1-8b": {
9184
9794
  id: "ibm-granite/granite-4.1-8b",
@@ -9282,6 +9892,23 @@ export const MODELS = {
9282
9892
  contextWindow: 256000,
9283
9893
  maxTokens: 80000,
9284
9894
  },
9895
+ "liquid/lfm-2.5-1.2b-thinking:free": {
9896
+ id: "liquid/lfm-2.5-1.2b-thinking:free",
9897
+ name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
9898
+ api: "openai-completions",
9899
+ provider: "openrouter",
9900
+ baseUrl: "https://openrouter.ai/api/v1",
9901
+ reasoning: true,
9902
+ input: ["text"],
9903
+ cost: {
9904
+ input: 0,
9905
+ output: 0,
9906
+ cacheRead: 0,
9907
+ cacheWrite: 0,
9908
+ },
9909
+ contextWindow: 32768,
9910
+ maxTokens: 4096,
9911
+ },
9285
9912
  "meta-llama/llama-3.1-70b-instruct": {
9286
9913
  id: "meta-llama/llama-3.1-70b-instruct",
9287
9914
  name: "Meta: Llama 3.1 70B Instruct",
@@ -9444,9 +10071,9 @@ export const MODELS = {
9444
10071
  reasoning: true,
9445
10072
  input: ["text"],
9446
10073
  cost: {
9447
- input: 0.15,
9448
- output: 0.8999999999999999,
9449
- cacheRead: 0.049999999999999996,
10074
+ input: 0.12,
10075
+ output: 0.48,
10076
+ cacheRead: 0,
9450
10077
  cacheWrite: 0,
9451
10078
  },
9452
10079
  contextWindow: 204800,
@@ -9461,13 +10088,13 @@ export const MODELS = {
9461
10088
  reasoning: true,
9462
10089
  input: ["text"],
9463
10090
  cost: {
9464
- input: 0.27,
9465
- output: 1.08,
9466
- cacheRead: 0.054,
10091
+ input: 0.18,
10092
+ output: 0.72,
10093
+ cacheRead: 0,
9467
10094
  cacheWrite: 0,
9468
10095
  },
9469
10096
  contextWindow: 204800,
9470
- maxTokens: 131072,
10097
+ maxTokens: 196608,
9471
10098
  },
9472
10099
  "minimax/minimax-m3": {
9473
10100
  id: "minimax/minimax-m3",
@@ -9820,7 +10447,7 @@ export const MODELS = {
9820
10447
  cost: {
9821
10448
  input: 0.6,
9822
10449
  output: 2.5,
9823
- cacheRead: 0,
10450
+ cacheRead: 0.6,
9824
10451
  cacheWrite: 0,
9825
10452
  },
9826
10453
  contextWindow: 262144,
@@ -9852,30 +10479,30 @@ export const MODELS = {
9852
10479
  reasoning: true,
9853
10480
  input: ["text", "image"],
9854
10481
  cost: {
9855
- input: 0.6799999999999999,
10482
+ input: 0.66,
9856
10483
  output: 3.41,
9857
- cacheRead: 0.33999999999999997,
10484
+ cacheRead: 0.144,
9858
10485
  cacheWrite: 0,
9859
10486
  },
9860
10487
  contextWindow: 262144,
9861
- maxTokens: 262142,
10488
+ maxTokens: 262144,
9862
10489
  },
9863
- "nex-agi/nex-n2-pro:free": {
9864
- id: "nex-agi/nex-n2-pro:free",
9865
- name: "Nex AGI: Nex-N2-Pro (free)",
10490
+ "moonshotai/kimi-k2.7-code": {
10491
+ id: "moonshotai/kimi-k2.7-code",
10492
+ name: "MoonshotAI: Kimi K2.7 Code",
9866
10493
  api: "openai-completions",
9867
10494
  provider: "openrouter",
9868
10495
  baseUrl: "https://openrouter.ai/api/v1",
9869
10496
  reasoning: true,
9870
10497
  input: ["text", "image"],
9871
10498
  cost: {
9872
- input: 0,
9873
- output: 0,
9874
- cacheRead: 0,
10499
+ input: 0.74,
10500
+ output: 3.5,
10501
+ cacheRead: 0.15,
9875
10502
  cacheWrite: 0,
9876
10503
  },
9877
10504
  contextWindow: 262144,
9878
- maxTokens: 262144,
10505
+ maxTokens: 16384,
9879
10506
  },
9880
10507
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
9881
10508
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -9954,13 +10581,13 @@ export const MODELS = {
9954
10581
  reasoning: true,
9955
10582
  input: ["text"],
9956
10583
  cost: {
9957
- input: 0.09,
9958
- output: 0.44999999999999996,
10584
+ input: 0.08499999999999999,
10585
+ output: 0.39999999999999997,
9959
10586
  cacheRead: 0,
9960
10587
  cacheWrite: 0,
9961
10588
  },
9962
10589
  contextWindow: 1000000,
9963
- maxTokens: 4096,
10590
+ maxTokens: 16384,
9964
10591
  },
9965
10592
  "nvidia/nemotron-3-super-120b-a12b:free": {
9966
10593
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -9989,8 +10616,8 @@ export const MODELS = {
9989
10616
  input: ["text"],
9990
10617
  cost: {
9991
10618
  input: 0.5,
9992
- output: 2.5,
9993
- cacheRead: 0.15,
10619
+ output: 2.2,
10620
+ cacheRead: 0.09999999999999999,
9994
10621
  cacheWrite: 0,
9995
10622
  },
9996
10623
  contextWindow: 1000000,
@@ -10030,23 +10657,6 @@ export const MODELS = {
10030
10657
  contextWindow: 128000,
10031
10658
  maxTokens: 128000,
10032
10659
  },
10033
- "nvidia/nemotron-nano-9b-v2": {
10034
- id: "nvidia/nemotron-nano-9b-v2",
10035
- name: "NVIDIA: Nemotron Nano 9B V2",
10036
- api: "openai-completions",
10037
- provider: "openrouter",
10038
- baseUrl: "https://openrouter.ai/api/v1",
10039
- reasoning: true,
10040
- input: ["text"],
10041
- cost: {
10042
- input: 0.04,
10043
- output: 0.16,
10044
- cacheRead: 0,
10045
- cacheWrite: 0,
10046
- },
10047
- contextWindow: 131072,
10048
- maxTokens: 16384,
10049
- },
10050
10660
  "nvidia/nemotron-nano-9b-v2:free": {
10051
10661
  id: "nvidia/nemotron-nano-9b-v2:free",
10052
10662
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -10663,13 +11273,13 @@ export const MODELS = {
10663
11273
  reasoning: true,
10664
11274
  input: ["text"],
10665
11275
  cost: {
10666
- input: 0.039,
10667
- output: 0.18,
11276
+ input: 0.03,
11277
+ output: 0.15,
10668
11278
  cacheRead: 0,
10669
11279
  cacheWrite: 0,
10670
11280
  },
10671
11281
  contextWindow: 131072,
10672
- maxTokens: 4096,
11282
+ maxTokens: 131072,
10673
11283
  },
10674
11284
  "openai/gpt-oss-120b:free": {
10675
11285
  id: "openai/gpt-oss-120b:free",
@@ -10720,7 +11330,7 @@ export const MODELS = {
10720
11330
  cacheWrite: 0,
10721
11331
  },
10722
11332
  contextWindow: 131072,
10723
- maxTokens: 8192,
11333
+ maxTokens: 32768,
10724
11334
  },
10725
11335
  "openai/gpt-oss-safeguard-20b": {
10726
11336
  id: "openai/gpt-oss-safeguard-20b",
@@ -10733,7 +11343,7 @@ export const MODELS = {
10733
11343
  cost: {
10734
11344
  input: 0.075,
10735
11345
  output: 0.3,
10736
- cacheRead: 0.037,
11346
+ cacheRead: 0.0375,
10737
11347
  cacheWrite: 0,
10738
11348
  },
10739
11349
  contextWindow: 131072,
@@ -10943,6 +11553,23 @@ export const MODELS = {
10943
11553
  contextWindow: 1048756,
10944
11554
  maxTokens: 262144,
10945
11555
  },
11556
+ "poolside/laguna-m.1": {
11557
+ id: "poolside/laguna-m.1",
11558
+ name: "Poolside: Laguna M.1",
11559
+ api: "openai-completions",
11560
+ provider: "openrouter",
11561
+ baseUrl: "https://openrouter.ai/api/v1",
11562
+ reasoning: true,
11563
+ input: ["text"],
11564
+ cost: {
11565
+ input: 0.19999999999999998,
11566
+ output: 0.39999999999999997,
11567
+ cacheRead: 0.09999999999999999,
11568
+ cacheWrite: 0,
11569
+ },
11570
+ contextWindow: 262144,
11571
+ maxTokens: 32768,
11572
+ },
10946
11573
  "poolside/laguna-m.1:free": {
10947
11574
  id: "poolside/laguna-m.1:free",
10948
11575
  name: "Poolside: Laguna M.1 (free)",
@@ -10960,39 +11587,39 @@ export const MODELS = {
10960
11587
  contextWindow: 262144,
10961
11588
  maxTokens: 32768,
10962
11589
  },
10963
- "poolside/laguna-xs.2:free": {
10964
- id: "poolside/laguna-xs.2:free",
10965
- name: "Poolside: Laguna XS.2 (free)",
11590
+ "poolside/laguna-xs.2": {
11591
+ id: "poolside/laguna-xs.2",
11592
+ name: "Poolside: Laguna XS.2",
10966
11593
  api: "openai-completions",
10967
11594
  provider: "openrouter",
10968
11595
  baseUrl: "https://openrouter.ai/api/v1",
10969
11596
  reasoning: true,
10970
11597
  input: ["text"],
10971
11598
  cost: {
10972
- input: 0,
10973
- output: 0,
10974
- cacheRead: 0,
11599
+ input: 0.09999999999999999,
11600
+ output: 0.19999999999999998,
11601
+ cacheRead: 0.049999999999999996,
10975
11602
  cacheWrite: 0,
10976
11603
  },
10977
11604
  contextWindow: 262144,
10978
11605
  maxTokens: 32768,
10979
11606
  },
10980
- "prime-intellect/intellect-3": {
10981
- id: "prime-intellect/intellect-3",
10982
- name: "Prime Intellect: INTELLECT-3",
11607
+ "poolside/laguna-xs.2:free": {
11608
+ id: "poolside/laguna-xs.2:free",
11609
+ name: "Poolside: Laguna XS.2 (free)",
10983
11610
  api: "openai-completions",
10984
11611
  provider: "openrouter",
10985
11612
  baseUrl: "https://openrouter.ai/api/v1",
10986
11613
  reasoning: true,
10987
11614
  input: ["text"],
10988
11615
  cost: {
10989
- input: 0.19999999999999998,
10990
- output: 1.1,
11616
+ input: 0,
11617
+ output: 0,
10991
11618
  cacheRead: 0,
10992
11619
  cacheWrite: 0,
10993
11620
  },
10994
- contextWindow: 131072,
10995
- maxTokens: 131072,
11621
+ contextWindow: 262144,
11622
+ maxTokens: 32768,
10996
11623
  },
10997
11624
  "qwen/qwen-2.5-72b-instruct": {
10998
11625
  id: "qwen/qwen-2.5-72b-instruct",
@@ -11011,6 +11638,23 @@ export const MODELS = {
11011
11638
  contextWindow: 131072,
11012
11639
  maxTokens: 16384,
11013
11640
  },
11641
+ "qwen/qwen-2.5-7b-instruct": {
11642
+ id: "qwen/qwen-2.5-7b-instruct",
11643
+ name: "Qwen: Qwen2.5 7B Instruct",
11644
+ api: "openai-completions",
11645
+ provider: "openrouter",
11646
+ baseUrl: "https://openrouter.ai/api/v1",
11647
+ reasoning: false,
11648
+ input: ["text"],
11649
+ cost: {
11650
+ input: 0.04,
11651
+ output: 0.09999999999999999,
11652
+ cacheRead: 0,
11653
+ cacheWrite: 0,
11654
+ },
11655
+ contextWindow: 131072,
11656
+ maxTokens: 32768,
11657
+ },
11014
11658
  "qwen/qwen-plus": {
11015
11659
  id: "qwen/qwen-plus",
11016
11660
  name: "Qwen: Qwen-Plus",
@@ -11570,7 +12214,7 @@ export const MODELS = {
11570
12214
  cacheWrite: 0,
11571
12215
  },
11572
12216
  contextWindow: 262144,
11573
- maxTokens: 262144,
12217
+ maxTokens: 81920,
11574
12218
  },
11575
12219
  "qwen/qwen3.5-397b-a17b": {
11576
12220
  id: "qwen/qwen3.5-397b-a17b",
@@ -11581,13 +12225,13 @@ export const MODELS = {
11581
12225
  reasoning: true,
11582
12226
  input: ["text", "image"],
11583
12227
  cost: {
11584
- input: 0.39,
11585
- output: 2.34,
12228
+ input: 0.385,
12229
+ output: 2.4499999999999997,
11586
12230
  cacheRead: 0,
11587
12231
  cacheWrite: 0,
11588
12232
  },
11589
- contextWindow: 262144,
11590
- maxTokens: 65536,
12233
+ contextWindow: 256000,
12234
+ maxTokens: 4096,
11591
12235
  },
11592
12236
  "qwen/qwen3.5-9b": {
11593
12237
  id: "qwen/qwen3.5-9b",
@@ -11666,13 +12310,13 @@ export const MODELS = {
11666
12310
  reasoning: true,
11667
12311
  input: ["text", "image"],
11668
12312
  cost: {
11669
- input: 0.28900000000000003,
11670
- output: 2.4,
12313
+ input: 0.28850000000000003,
12314
+ output: 2.65,
11671
12315
  cacheRead: 0,
11672
12316
  cacheWrite: 0,
11673
12317
  },
11674
12318
  contextWindow: 262144,
11675
- maxTokens: 131072,
12319
+ maxTokens: 262140,
11676
12320
  },
11677
12321
  "qwen/qwen3.6-35b-a3b": {
11678
12322
  id: "qwen/qwen3.6-35b-a3b",
@@ -11683,9 +12327,9 @@ export const MODELS = {
11683
12327
  reasoning: true,
11684
12328
  input: ["text", "image"],
11685
12329
  cost: {
11686
- input: 0.15,
12330
+ input: 0.14,
11687
12331
  output: 1,
11688
- cacheRead: 0.049999999999999996,
12332
+ cacheRead: 0,
11689
12333
  cacheWrite: 0,
11690
12334
  },
11691
12335
  contextWindow: 262144,
@@ -11768,10 +12412,10 @@ export const MODELS = {
11768
12412
  reasoning: true,
11769
12413
  input: ["text", "image"],
11770
12414
  cost: {
11771
- input: 0.39999999999999997,
11772
- output: 1.5999999999999999,
11773
- cacheRead: 0.08,
11774
- cacheWrite: 0.5,
12415
+ input: 0.32,
12416
+ output: 1.28,
12417
+ cacheRead: 0.064,
12418
+ cacheWrite: 0.39999999999999997,
11775
12419
  },
11776
12420
  contextWindow: 1000000,
11777
12421
  maxTokens: 65536,
@@ -11810,6 +12454,23 @@ export const MODELS = {
11810
12454
  contextWindow: 256000,
11811
12455
  maxTokens: 128000,
11812
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
+ },
11813
12474
  "sao10k/l3.1-euryale-70b": {
11814
12475
  id: "sao10k/l3.1-euryale-70b",
11815
12476
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -11861,23 +12522,6 @@ export const MODELS = {
11861
12522
  contextWindow: 256000,
11862
12523
  maxTokens: 256000,
11863
12524
  },
11864
- "thedrummer/rocinante-12b": {
11865
- id: "thedrummer/rocinante-12b",
11866
- name: "TheDrummer: Rocinante 12B",
11867
- api: "openai-completions",
11868
- provider: "openrouter",
11869
- baseUrl: "https://openrouter.ai/api/v1",
11870
- reasoning: false,
11871
- input: ["text"],
11872
- cost: {
11873
- input: 0.16999999999999998,
11874
- output: 0.43,
11875
- cacheRead: 0,
11876
- cacheWrite: 0,
11877
- },
11878
- contextWindow: 32768,
11879
- maxTokens: 32768,
11880
- },
11881
12525
  "thedrummer/unslopnemo-12b": {
11882
12526
  id: "thedrummer/unslopnemo-12b",
11883
12527
  name: "TheDrummer: UnslopNemo 12B",
@@ -11972,13 +12616,13 @@ export const MODELS = {
11972
12616
  reasoning: true,
11973
12617
  input: ["text", "image"],
11974
12618
  cost: {
11975
- input: 0.14,
12619
+ input: 0.105,
11976
12620
  output: 0.28,
11977
- cacheRead: 0.0028,
12621
+ cacheRead: 0,
11978
12622
  cacheWrite: 0,
11979
12623
  },
11980
12624
  contextWindow: 1048576,
11981
- maxTokens: 131072,
12625
+ maxTokens: 4096,
11982
12626
  },
11983
12627
  "xiaomi/mimo-v2.5-pro": {
11984
12628
  id: "xiaomi/mimo-v2.5-pro",
@@ -12023,13 +12667,13 @@ export const MODELS = {
12023
12667
  reasoning: true,
12024
12668
  input: ["text"],
12025
12669
  cost: {
12026
- input: 0.125,
12670
+ input: 0.13,
12027
12671
  output: 0.85,
12028
- cacheRead: 0.06,
12672
+ cacheRead: 0.024999999999999998,
12029
12673
  cacheWrite: 0,
12030
12674
  },
12031
12675
  contextWindow: 131072,
12032
- maxTokens: 131070,
12676
+ maxTokens: 98304,
12033
12677
  },
12034
12678
  "z-ai/glm-4.5v": {
12035
12679
  id: "z-ai/glm-4.5v",
@@ -12167,6 +12811,40 @@ export const MODELS = {
12167
12811
  contextWindow: 202752,
12168
12812
  maxTokens: 4096,
12169
12813
  },
12814
+ "z-ai/glm-5.2": {
12815
+ id: "z-ai/glm-5.2",
12816
+ name: "Z.ai: GLM 5.2",
12817
+ api: "openai-completions",
12818
+ provider: "openrouter",
12819
+ baseUrl: "https://openrouter.ai/api/v1",
12820
+ reasoning: true,
12821
+ input: ["text"],
12822
+ cost: {
12823
+ input: 0.95,
12824
+ output: 3,
12825
+ cacheRead: 0.18,
12826
+ cacheWrite: 0,
12827
+ },
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,
12846
+ maxTokens: 131072,
12847
+ },
12170
12848
  "~anthropic/claude-fable-latest": {
12171
12849
  id: "~anthropic/claude-fable-latest",
12172
12850
  name: "Anthropic: Claude Fable Latest",
@@ -12279,13 +12957,13 @@ export const MODELS = {
12279
12957
  reasoning: true,
12280
12958
  input: ["text", "image"],
12281
12959
  cost: {
12282
- input: 0.6799999999999999,
12960
+ input: 0.66,
12283
12961
  output: 3.41,
12284
- cacheRead: 0.33999999999999997,
12962
+ cacheRead: 0.144,
12285
12963
  cacheWrite: 0,
12286
12964
  },
12287
12965
  contextWindow: 262144,
12288
- maxTokens: 262142,
12966
+ maxTokens: 262144,
12289
12967
  },
12290
12968
  "~openai/gpt-latest": {
12291
12969
  id: "~openai/gpt-latest",
@@ -12323,9 +13001,9 @@ export const MODELS = {
12323
13001
  },
12324
13002
  },
12325
13003
  "together": {
12326
- "MiniMaxAI/MiniMax-M2.5": {
12327
- id: "MiniMaxAI/MiniMax-M2.5",
12328
- name: "MiniMax-M2.5",
13004
+ "MiniMaxAI/MiniMax-M2.7": {
13005
+ id: "MiniMaxAI/MiniMax-M2.7",
13006
+ name: "MiniMax-M2.7",
12329
13007
  api: "openai-completions",
12330
13008
  provider: "together",
12331
13009
  baseUrl: "https://api.together.ai/v1",
@@ -12339,50 +13017,49 @@ export const MODELS = {
12339
13017
  cacheRead: 0.06,
12340
13018
  cacheWrite: 0,
12341
13019
  },
12342
- contextWindow: 204800,
13020
+ contextWindow: 202752,
12343
13021
  maxTokens: 131072,
12344
13022
  },
12345
- "MiniMaxAI/MiniMax-M2.7": {
12346
- id: "MiniMaxAI/MiniMax-M2.7",
12347
- name: "MiniMax-M2.7",
13023
+ "MiniMaxAI/MiniMax-M3": {
13024
+ id: "MiniMaxAI/MiniMax-M3",
13025
+ name: "MiniMax-M3",
12348
13026
  api: "openai-completions",
12349
13027
  provider: "together",
12350
13028
  baseUrl: "https://api.together.ai/v1",
12351
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
13029
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12352
13030
  reasoning: true,
12353
- thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null },
12354
- input: ["text"],
13031
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13032
+ input: ["text", "image"],
12355
13033
  cost: {
12356
13034
  input: 0.3,
12357
13035
  output: 1.2,
12358
13036
  cacheRead: 0.06,
12359
13037
  cacheWrite: 0,
12360
13038
  },
12361
- contextWindow: 202752,
12362
- maxTokens: 131072,
13039
+ contextWindow: 524288,
13040
+ maxTokens: 250000,
12363
13041
  },
12364
- "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12365
- id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12366
- name: "Qwen3 235B A22B Instruct 2507 FP8",
13042
+ "Qwen/Qwen2.5-7B-Instruct-Turbo": {
13043
+ id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
13044
+ name: "Qwen 2.5 7B Instruct Turbo",
12367
13045
  api: "openai-completions",
12368
13046
  provider: "together",
12369
13047
  baseUrl: "https://api.together.ai/v1",
12370
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12371
- reasoning: true,
12372
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13048
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
13049
+ reasoning: false,
12373
13050
  input: ["text"],
12374
13051
  cost: {
12375
- input: 0.2,
12376
- output: 0.6,
13052
+ input: 0.3,
13053
+ output: 0.3,
12377
13054
  cacheRead: 0,
12378
13055
  cacheWrite: 0,
12379
13056
  },
12380
- contextWindow: 262144,
12381
- maxTokens: 262144,
13057
+ contextWindow: 32768,
13058
+ maxTokens: 32768,
12382
13059
  },
12383
- "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {
12384
- id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8",
12385
- name: "Qwen3 Coder 480B A35B Instruct",
13060
+ "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
13061
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
13062
+ name: "Qwen3 235B A22B Instruct 2507 FP8",
12386
13063
  api: "openai-completions",
12387
13064
  provider: "together",
12388
13065
  baseUrl: "https://api.together.ai/v1",
@@ -12390,36 +13067,36 @@ export const MODELS = {
12390
13067
  reasoning: false,
12391
13068
  input: ["text"],
12392
13069
  cost: {
12393
- input: 2,
12394
- output: 2,
13070
+ input: 0.2,
13071
+ output: 0.6,
12395
13072
  cacheRead: 0,
12396
13073
  cacheWrite: 0,
12397
13074
  },
12398
13075
  contextWindow: 262144,
12399
13076
  maxTokens: 262144,
12400
13077
  },
12401
- "Qwen/Qwen3-Coder-Next-FP8": {
12402
- id: "Qwen/Qwen3-Coder-Next-FP8",
12403
- name: "Qwen3 Coder Next FP8",
13078
+ "Qwen/Qwen3.5-397B-A17B": {
13079
+ id: "Qwen/Qwen3.5-397B-A17B",
13080
+ name: "Qwen3.5 397B A17B",
12404
13081
  api: "openai-completions",
12405
13082
  provider: "together",
12406
13083
  baseUrl: "https://api.together.ai/v1",
12407
13084
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12408
13085
  reasoning: true,
12409
13086
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12410
- input: ["text"],
13087
+ input: ["text", "image"],
12411
13088
  cost: {
12412
- input: 0.5,
12413
- output: 1.2,
13089
+ input: 0.6,
13090
+ output: 3.6,
12414
13091
  cacheRead: 0,
12415
13092
  cacheWrite: 0,
12416
13093
  },
12417
13094
  contextWindow: 262144,
12418
- maxTokens: 262144,
13095
+ maxTokens: 130000,
12419
13096
  },
12420
- "Qwen/Qwen3.5-397B-A17B": {
12421
- id: "Qwen/Qwen3.5-397B-A17B",
12422
- name: "Qwen3.5 397B A17B",
13097
+ "Qwen/Qwen3.5-9B": {
13098
+ id: "Qwen/Qwen3.5-9B",
13099
+ name: "Qwen3.5 9B",
12423
13100
  api: "openai-completions",
12424
13101
  provider: "together",
12425
13102
  baseUrl: "https://api.together.ai/v1",
@@ -12428,13 +13105,13 @@ export const MODELS = {
12428
13105
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12429
13106
  input: ["text", "image"],
12430
13107
  cost: {
12431
- input: 0.6,
12432
- output: 3.6,
13108
+ input: 0.17,
13109
+ output: 0.25,
12433
13110
  cacheRead: 0,
12434
13111
  cacheWrite: 0,
12435
13112
  },
12436
13113
  contextWindow: 262144,
12437
- maxTokens: 130000,
13114
+ maxTokens: 65536,
12438
13115
  },
12439
13116
  "Qwen/Qwen3.6-Plus": {
12440
13117
  id: "Qwen/Qwen3.6-Plus",
@@ -12461,57 +13138,18 @@ export const MODELS = {
12461
13138
  api: "openai-completions",
12462
13139
  provider: "together",
12463
13140
  baseUrl: "https://api.together.ai/v1",
12464
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12465
- reasoning: true,
12466
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13141
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
13142
+ reasoning: false,
12467
13143
  input: ["text"],
12468
13144
  cost: {
12469
- input: 2.5,
12470
- output: 7.5,
13145
+ input: 1.25,
13146
+ output: 3.75,
12471
13147
  cacheRead: 0,
12472
13148
  cacheWrite: 0,
12473
13149
  },
12474
13150
  contextWindow: 1000000,
12475
13151
  maxTokens: 500000,
12476
13152
  },
12477
- "deepseek-ai/DeepSeek-V3": {
12478
- id: "deepseek-ai/DeepSeek-V3",
12479
- name: "DeepSeek-V3",
12480
- api: "openai-completions",
12481
- provider: "together",
12482
- baseUrl: "https://api.together.ai/v1",
12483
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12484
- reasoning: true,
12485
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12486
- input: ["text"],
12487
- cost: {
12488
- input: 1.25,
12489
- output: 1.25,
12490
- cacheRead: 0,
12491
- cacheWrite: 0,
12492
- },
12493
- contextWindow: 131072,
12494
- maxTokens: 131072,
12495
- },
12496
- "deepseek-ai/DeepSeek-V3-1": {
12497
- id: "deepseek-ai/DeepSeek-V3-1",
12498
- name: "DeepSeek V3.1",
12499
- api: "openai-completions",
12500
- provider: "together",
12501
- baseUrl: "https://api.together.ai/v1",
12502
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12503
- reasoning: true,
12504
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12505
- input: ["text"],
12506
- cost: {
12507
- input: 0.6,
12508
- output: 1.7,
12509
- cacheRead: 0,
12510
- cacheWrite: 0,
12511
- },
12512
- contextWindow: 131072,
12513
- maxTokens: 131072,
12514
- },
12515
13153
  "deepseek-ai/DeepSeek-V4-Pro": {
12516
13154
  id: "deepseek-ai/DeepSeek-V4-Pro",
12517
13155
  name: "DeepSeek V4 Pro",
@@ -12523,8 +13161,8 @@ export const MODELS = {
12523
13161
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": null },
12524
13162
  input: ["text"],
12525
13163
  cost: {
12526
- input: 2.1,
12527
- output: 4.4,
13164
+ input: 1.74,
13165
+ output: 3.48,
12528
13166
  cacheRead: 0.2,
12529
13167
  cacheWrite: 0,
12530
13168
  },
@@ -12560,8 +13198,8 @@ export const MODELS = {
12560
13198
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12561
13199
  input: ["text", "image"],
12562
13200
  cost: {
12563
- input: 0.2,
12564
- output: 0.5,
13201
+ input: 0.39,
13202
+ output: 0.97,
12565
13203
  cacheRead: 0,
12566
13204
  cacheWrite: 0,
12567
13205
  },
@@ -12586,9 +13224,9 @@ export const MODELS = {
12586
13224
  contextWindow: 131072,
12587
13225
  maxTokens: 131072,
12588
13226
  },
12589
- "moonshotai/Kimi-K2.5": {
12590
- id: "moonshotai/Kimi-K2.5",
12591
- name: "Kimi K2.5",
13227
+ "moonshotai/Kimi-K2.6": {
13228
+ id: "moonshotai/Kimi-K2.6",
13229
+ name: "Kimi K2.6",
12592
13230
  api: "openai-completions",
12593
13231
  provider: "together",
12594
13232
  baseUrl: "https://api.together.ai/v1",
@@ -12597,32 +13235,32 @@ export const MODELS = {
12597
13235
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12598
13236
  input: ["text", "image"],
12599
13237
  cost: {
12600
- input: 0.5,
12601
- output: 2.8,
12602
- cacheRead: 0,
13238
+ input: 1.2,
13239
+ output: 4.5,
13240
+ cacheRead: 0.2,
12603
13241
  cacheWrite: 0,
12604
13242
  },
12605
13243
  contextWindow: 262144,
12606
- maxTokens: 262144,
13244
+ maxTokens: 131000,
12607
13245
  },
12608
- "moonshotai/Kimi-K2.6": {
12609
- id: "moonshotai/Kimi-K2.6",
12610
- name: "Kimi K2.6",
13246
+ "moonshotai/Kimi-K2.7-Code": {
13247
+ id: "moonshotai/Kimi-K2.7-Code",
13248
+ name: "Kimi K2.7 Code",
12611
13249
  api: "openai-completions",
12612
13250
  provider: "together",
12613
13251
  baseUrl: "https://api.together.ai/v1",
12614
13252
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12615
13253
  reasoning: true,
12616
13254
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12617
- input: ["text", "image"],
13255
+ input: ["text"],
12618
13256
  cost: {
12619
- input: 1.2,
12620
- output: 4.5,
12621
- cacheRead: 0.2,
13257
+ input: 0.95,
13258
+ output: 4,
13259
+ cacheRead: 0.19,
12622
13260
  cacheWrite: 0,
12623
13261
  },
12624
13262
  contextWindow: 262144,
12625
- maxTokens: 131000,
13263
+ maxTokens: 131072,
12626
13264
  },
12627
13265
  "nvidia/nemotron-3-ultra-550b-a55b": {
12628
13266
  id: "nvidia/nemotron-3-ultra-550b-a55b",
@@ -12659,12 +13297,69 @@ export const MODELS = {
12659
13297
  cacheRead: 0,
12660
13298
  cacheWrite: 0,
12661
13299
  },
12662
- contextWindow: 131072,
13300
+ contextWindow: 131072,
13301
+ maxTokens: 131072,
13302
+ },
13303
+ "openai/gpt-oss-20b": {
13304
+ id: "openai/gpt-oss-20b",
13305
+ name: "GPT OSS 20B",
13306
+ api: "openai-completions",
13307
+ provider: "together",
13308
+ baseUrl: "https://api.together.ai/v1",
13309
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": true, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "openai" },
13310
+ reasoning: true,
13311
+ thinkingLevelMap: { "off": null, "minimal": null },
13312
+ input: ["text"],
13313
+ cost: {
13314
+ input: 0.05,
13315
+ output: 0.2,
13316
+ cacheRead: 0,
13317
+ cacheWrite: 0,
13318
+ },
13319
+ contextWindow: 131072,
13320
+ maxTokens: 131072,
13321
+ },
13322
+ "zai-org/GLM-5": {
13323
+ id: "zai-org/GLM-5",
13324
+ name: "GLM-5",
13325
+ api: "openai-completions",
13326
+ provider: "together",
13327
+ baseUrl: "https://api.together.ai/v1",
13328
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13329
+ reasoning: true,
13330
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13331
+ input: ["text"],
13332
+ cost: {
13333
+ input: 1,
13334
+ output: 3.2,
13335
+ cacheRead: 0,
13336
+ cacheWrite: 0,
13337
+ },
13338
+ contextWindow: 202752,
13339
+ maxTokens: 131072,
13340
+ },
13341
+ "zai-org/GLM-5.1": {
13342
+ id: "zai-org/GLM-5.1",
13343
+ name: "GLM-5.1",
13344
+ api: "openai-completions",
13345
+ provider: "together",
13346
+ baseUrl: "https://api.together.ai/v1",
13347
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13348
+ reasoning: true,
13349
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13350
+ input: ["text"],
13351
+ cost: {
13352
+ input: 1.4,
13353
+ output: 4.4,
13354
+ cacheRead: 0,
13355
+ cacheWrite: 0,
13356
+ },
13357
+ contextWindow: 202752,
12663
13358
  maxTokens: 131072,
12664
13359
  },
12665
- "zai-org/GLM-5.1": {
12666
- id: "zai-org/GLM-5.1",
12667
- name: "GLM-5.1",
13360
+ "zai-org/GLM-5.2": {
13361
+ id: "zai-org/GLM-5.2",
13362
+ name: "GLM-5.2",
12668
13363
  api: "openai-completions",
12669
13364
  provider: "together",
12670
13365
  baseUrl: "https://api.together.ai/v1",
@@ -12675,11 +13370,11 @@ export const MODELS = {
12675
13370
  cost: {
12676
13371
  input: 1.4,
12677
13372
  output: 4.4,
12678
- cacheRead: 0,
13373
+ cacheRead: 0.26,
12679
13374
  cacheWrite: 0,
12680
13375
  },
12681
- contextWindow: 202752,
12682
- maxTokens: 131072,
13376
+ contextWindow: 262144,
13377
+ maxTokens: 164000,
12683
13378
  },
12684
13379
  },
12685
13380
  "vercel-ai-gateway": {
@@ -12791,7 +13486,7 @@ export const MODELS = {
12791
13486
  api: "anthropic-messages",
12792
13487
  provider: "vercel-ai-gateway",
12793
13488
  baseUrl: "https://ai-gateway.vercel.sh",
12794
- reasoning: true,
13489
+ reasoning: false,
12795
13490
  input: ["text"],
12796
13491
  cost: {
12797
13492
  input: 1.5,
@@ -12808,7 +13503,7 @@ export const MODELS = {
12808
13503
  api: "anthropic-messages",
12809
13504
  provider: "vercel-ai-gateway",
12810
13505
  baseUrl: "https://ai-gateway.vercel.sh",
12811
- reasoning: true,
13506
+ reasoning: false,
12812
13507
  input: ["text"],
12813
13508
  cost: {
12814
13509
  input: 0.15,
@@ -12825,7 +13520,7 @@ export const MODELS = {
12825
13520
  api: "anthropic-messages",
12826
13521
  provider: "vercel-ai-gateway",
12827
13522
  baseUrl: "https://ai-gateway.vercel.sh",
12828
- reasoning: true,
13523
+ reasoning: false,
12829
13524
  input: ["text"],
12830
13525
  cost: {
12831
13526
  input: 0.5,
@@ -12938,6 +13633,40 @@ export const MODELS = {
12938
13633
  contextWindow: 131072,
12939
13634
  maxTokens: 32768,
12940
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
+ },
12941
13670
  "alibaba/qwen3-vl-thinking": {
12942
13671
  id: "alibaba/qwen3-vl-thinking",
12943
13672
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13057,6 +13786,74 @@ export const MODELS = {
13057
13786
  contextWindow: 1000000,
13058
13787
  maxTokens: 64000,
13059
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
+ },
13060
13857
  "anthropic/claude-3-haiku": {
13061
13858
  id: "anthropic/claude-3-haiku",
13062
13859
  name: "Claude 3 Haiku",
@@ -13074,24 +13871,6 @@ export const MODELS = {
13074
13871
  contextWindow: 200000,
13075
13872
  maxTokens: 4096,
13076
13873
  },
13077
- "anthropic/claude-fable-5": {
13078
- id: "anthropic/claude-fable-5",
13079
- name: "Claude Fable 5",
13080
- api: "anthropic-messages",
13081
- provider: "vercel-ai-gateway",
13082
- baseUrl: "https://ai-gateway.vercel.sh",
13083
- reasoning: true,
13084
- thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
13085
- input: ["text", "image"],
13086
- cost: {
13087
- input: 10,
13088
- output: 50,
13089
- cacheRead: 1,
13090
- cacheWrite: 12.5,
13091
- },
13092
- contextWindow: 1000000,
13093
- maxTokens: 128000,
13094
- },
13095
13874
  "anthropic/claude-haiku-4.5": {
13096
13875
  id: "anthropic/claude-haiku-4.5",
13097
13876
  name: "Claude Haiku 4.5",
@@ -13299,6 +14078,23 @@ export const MODELS = {
13299
14078
  contextWindow: 262100,
13300
14079
  maxTokens: 80000,
13301
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
+ },
13302
14098
  "bytedance/seed-1.6": {
13303
14099
  id: "bytedance/seed-1.6",
13304
14100
  name: "Seed 1.6",
@@ -13306,7 +14102,7 @@ export const MODELS = {
13306
14102
  provider: "vercel-ai-gateway",
13307
14103
  baseUrl: "https://ai-gateway.vercel.sh",
13308
14104
  reasoning: true,
13309
- input: ["text"],
14105
+ input: ["text", "image"],
13310
14106
  cost: {
13311
14107
  input: 0.25,
13312
14108
  output: 2,
@@ -13316,6 +14112,23 @@ export const MODELS = {
13316
14112
  contextWindow: 256000,
13317
14113
  maxTokens: 32000,
13318
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
+ },
13319
14132
  "cohere/command-a": {
13320
14133
  id: "cohere/command-a",
13321
14134
  name: "Command A",
@@ -13407,8 +14220,8 @@ export const MODELS = {
13407
14220
  api: "anthropic-messages",
13408
14221
  provider: "vercel-ai-gateway",
13409
14222
  baseUrl: "https://ai-gateway.vercel.sh",
13410
- reasoning: true,
13411
- input: ["text", "image"],
14223
+ reasoning: false,
14224
+ input: ["text"],
13412
14225
  cost: {
13413
14226
  input: 0.28,
13414
14227
  output: 0.42,
@@ -13425,7 +14238,7 @@ export const MODELS = {
13425
14238
  provider: "vercel-ai-gateway",
13426
14239
  baseUrl: "https://ai-gateway.vercel.sh",
13427
14240
  reasoning: true,
13428
- input: ["text", "image"],
14241
+ input: ["text"],
13429
14242
  cost: {
13430
14243
  input: 0.62,
13431
14244
  output: 1.85,
@@ -13442,7 +14255,7 @@ export const MODELS = {
13442
14255
  provider: "vercel-ai-gateway",
13443
14256
  baseUrl: "https://ai-gateway.vercel.sh",
13444
14257
  reasoning: true,
13445
- input: ["text", "image"],
14258
+ input: ["text"],
13446
14259
  cost: {
13447
14260
  input: 0.14,
13448
14261
  output: 0.28,
@@ -13645,7 +14458,7 @@ export const MODELS = {
13645
14458
  api: "anthropic-messages",
13646
14459
  provider: "vercel-ai-gateway",
13647
14460
  baseUrl: "https://ai-gateway.vercel.sh",
13648
- reasoning: false,
14461
+ reasoning: true,
13649
14462
  input: ["text", "image"],
13650
14463
  cost: {
13651
14464
  input: 0.14,
@@ -13690,6 +14503,40 @@ export const MODELS = {
13690
14503
  contextWindow: 32000,
13691
14504
  maxTokens: 16384,
13692
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
+ },
13693
14540
  "kwaipilot/kat-coder-pro-v2": {
13694
14541
  id: "kwaipilot/kat-coder-pro-v2",
13695
14542
  name: "Kat Coder Pro V2",
@@ -13724,6 +14571,23 @@ export const MODELS = {
13724
14571
  contextWindow: 128000,
13725
14572
  maxTokens: 100000,
13726
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
+ },
13727
14591
  "meta/llama-3.1-70b": {
13728
14592
  id: "meta/llama-3.1-70b",
13729
14593
  name: "Llama 3.1 70B Instruct",
@@ -13993,54 +14857,105 @@ export const MODELS = {
13993
14857
  cacheRead: 0,
13994
14858
  cacheWrite: 0,
13995
14859
  },
13996
- contextWindow: 128000,
13997
- maxTokens: 4000,
14860
+ contextWindow: 128000,
14861
+ maxTokens: 4000,
14862
+ },
14863
+ "mistral/devstral-2": {
14864
+ id: "mistral/devstral-2",
14865
+ name: "Devstral 2",
14866
+ api: "anthropic-messages",
14867
+ provider: "vercel-ai-gateway",
14868
+ baseUrl: "https://ai-gateway.vercel.sh",
14869
+ reasoning: false,
14870
+ input: ["text"],
14871
+ cost: {
14872
+ input: 0.39999999999999997,
14873
+ output: 2,
14874
+ cacheRead: 0,
14875
+ cacheWrite: 0,
14876
+ },
14877
+ contextWindow: 256000,
14878
+ maxTokens: 256000,
14879
+ },
14880
+ "mistral/devstral-small": {
14881
+ id: "mistral/devstral-small",
14882
+ name: "Devstral Small 1.1",
14883
+ api: "anthropic-messages",
14884
+ provider: "vercel-ai-gateway",
14885
+ baseUrl: "https://ai-gateway.vercel.sh",
14886
+ reasoning: false,
14887
+ input: ["text"],
14888
+ cost: {
14889
+ input: 0.09999999999999999,
14890
+ output: 0.3,
14891
+ cacheRead: 0,
14892
+ cacheWrite: 0,
14893
+ },
14894
+ contextWindow: 128000,
14895
+ maxTokens: 64000,
14896
+ },
14897
+ "mistral/devstral-small-2": {
14898
+ id: "mistral/devstral-small-2",
14899
+ name: "Devstral Small 2",
14900
+ api: "anthropic-messages",
14901
+ provider: "vercel-ai-gateway",
14902
+ baseUrl: "https://ai-gateway.vercel.sh",
14903
+ reasoning: false,
14904
+ input: ["text", "image"],
14905
+ cost: {
14906
+ input: 0.09999999999999999,
14907
+ output: 0.3,
14908
+ cacheRead: 0,
14909
+ cacheWrite: 0,
14910
+ },
14911
+ contextWindow: 256000,
14912
+ maxTokens: 256000,
13998
14913
  },
13999
- "mistral/devstral-2": {
14000
- id: "mistral/devstral-2",
14001
- name: "Devstral 2",
14914
+ "mistral/magistral-medium": {
14915
+ id: "mistral/magistral-medium",
14916
+ name: "Magistral Medium 2509",
14002
14917
  api: "anthropic-messages",
14003
14918
  provider: "vercel-ai-gateway",
14004
14919
  baseUrl: "https://ai-gateway.vercel.sh",
14005
- reasoning: false,
14006
- input: ["text"],
14920
+ reasoning: true,
14921
+ input: ["text", "image"],
14007
14922
  cost: {
14008
- input: 0.39999999999999997,
14009
- output: 2,
14923
+ input: 2,
14924
+ output: 5,
14010
14925
  cacheRead: 0,
14011
14926
  cacheWrite: 0,
14012
14927
  },
14013
- contextWindow: 256000,
14014
- maxTokens: 256000,
14928
+ contextWindow: 128000,
14929
+ maxTokens: 64000,
14015
14930
  },
14016
- "mistral/devstral-small": {
14017
- id: "mistral/devstral-small",
14018
- name: "Devstral Small 1.1",
14931
+ "mistral/magistral-small": {
14932
+ id: "mistral/magistral-small",
14933
+ name: "Magistral Small 2509",
14019
14934
  api: "anthropic-messages",
14020
14935
  provider: "vercel-ai-gateway",
14021
14936
  baseUrl: "https://ai-gateway.vercel.sh",
14022
- reasoning: false,
14023
- input: ["text"],
14937
+ reasoning: true,
14938
+ input: ["text", "image"],
14024
14939
  cost: {
14025
- input: 0.09999999999999999,
14026
- output: 0.3,
14940
+ input: 0.5,
14941
+ output: 1.5,
14027
14942
  cacheRead: 0,
14028
14943
  cacheWrite: 0,
14029
14944
  },
14030
14945
  contextWindow: 128000,
14031
14946
  maxTokens: 64000,
14032
14947
  },
14033
- "mistral/devstral-small-2": {
14034
- id: "mistral/devstral-small-2",
14035
- name: "Devstral Small 2",
14948
+ "mistral/ministral-14b": {
14949
+ id: "mistral/ministral-14b",
14950
+ name: "Ministral 14B",
14036
14951
  api: "anthropic-messages",
14037
14952
  provider: "vercel-ai-gateway",
14038
14953
  baseUrl: "https://ai-gateway.vercel.sh",
14039
14954
  reasoning: false,
14040
- input: ["text"],
14955
+ input: ["text", "image"],
14041
14956
  cost: {
14042
- input: 0.09999999999999999,
14043
- output: 0.3,
14957
+ input: 0.19999999999999998,
14958
+ output: 0.19999999999999998,
14044
14959
  cacheRead: 0,
14045
14960
  cacheWrite: 0,
14046
14961
  },
@@ -14081,6 +14996,23 @@ export const MODELS = {
14081
14996
  contextWindow: 128000,
14082
14997
  maxTokens: 4000,
14083
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
+ },
14084
15016
  "mistral/mistral-medium": {
14085
15017
  id: "mistral/mistral-medium",
14086
15018
  name: "Mistral Medium 3.1",
@@ -14105,7 +15037,7 @@ export const MODELS = {
14105
15037
  provider: "vercel-ai-gateway",
14106
15038
  baseUrl: "https://ai-gateway.vercel.sh",
14107
15039
  reasoning: true,
14108
- input: ["text"],
15040
+ input: ["text", "image"],
14109
15041
  cost: {
14110
15042
  input: 1.5,
14111
15043
  output: 7.5,
@@ -14124,13 +15056,13 @@ export const MODELS = {
14124
15056
  reasoning: false,
14125
15057
  input: ["text"],
14126
15058
  cost: {
14127
- input: 0.02,
14128
- output: 0.04,
15059
+ input: 0.15,
15060
+ output: 0.15,
14129
15061
  cacheRead: 0,
14130
15062
  cacheWrite: 0,
14131
15063
  },
14132
- contextWindow: 131072,
14133
- maxTokens: 131072,
15064
+ contextWindow: 128000,
15065
+ maxTokens: 128000,
14134
15066
  },
14135
15067
  "mistral/mistral-small": {
14136
15068
  id: "mistral/mistral-small",
@@ -14217,73 +15149,90 @@ export const MODELS = {
14217
15149
  contextWindow: 262114,
14218
15150
  maxTokens: 262114,
14219
15151
  },
14220
- "moonshotai/kimi-k2-thinking-turbo": {
14221
- id: "moonshotai/kimi-k2-thinking-turbo",
14222
- name: "Kimi K2 Thinking Turbo",
15152
+ "moonshotai/kimi-k2.5": {
15153
+ id: "moonshotai/kimi-k2.5",
15154
+ name: "Kimi K2.5",
14223
15155
  api: "anthropic-messages",
14224
15156
  provider: "vercel-ai-gateway",
14225
15157
  baseUrl: "https://ai-gateway.vercel.sh",
14226
15158
  reasoning: true,
14227
- input: ["text"],
15159
+ input: ["text", "image"],
14228
15160
  cost: {
14229
- input: 1.15,
14230
- output: 8,
14231
- cacheRead: 0.15,
15161
+ input: 0.6,
15162
+ output: 3,
15163
+ cacheRead: 0.09999999999999999,
14232
15164
  cacheWrite: 0,
14233
15165
  },
14234
15166
  contextWindow: 262114,
14235
15167
  maxTokens: 262114,
14236
15168
  },
14237
- "moonshotai/kimi-k2-turbo": {
14238
- id: "moonshotai/kimi-k2-turbo",
14239
- name: "Kimi K2 Turbo",
15169
+ "moonshotai/kimi-k2.6": {
15170
+ id: "moonshotai/kimi-k2.6",
15171
+ name: "Kimi K2.6",
14240
15172
  api: "anthropic-messages",
14241
15173
  provider: "vercel-ai-gateway",
14242
15174
  baseUrl: "https://ai-gateway.vercel.sh",
14243
- reasoning: false,
14244
- input: ["text"],
15175
+ reasoning: true,
15176
+ input: ["text", "image"],
14245
15177
  cost: {
14246
- input: 1.15,
14247
- output: 8,
14248
- cacheRead: 0.15,
15178
+ input: 0.95,
15179
+ output: 4,
15180
+ cacheRead: 0.16,
14249
15181
  cacheWrite: 0,
14250
15182
  },
14251
- contextWindow: 256000,
14252
- maxTokens: 16384,
15183
+ contextWindow: 262000,
15184
+ maxTokens: 262000,
14253
15185
  },
14254
- "moonshotai/kimi-k2.5": {
14255
- id: "moonshotai/kimi-k2.5",
14256
- name: "Kimi K2.5",
15186
+ "moonshotai/kimi-k2.7-code": {
15187
+ id: "moonshotai/kimi-k2.7-code",
15188
+ name: "Kimi K2.7 Code",
14257
15189
  api: "anthropic-messages",
14258
15190
  provider: "vercel-ai-gateway",
14259
15191
  baseUrl: "https://ai-gateway.vercel.sh",
14260
15192
  reasoning: true,
14261
15193
  input: ["text", "image"],
14262
15194
  cost: {
14263
- input: 0.6,
14264
- output: 3,
14265
- cacheRead: 0.09999999999999999,
15195
+ input: 0.95,
15196
+ output: 4,
15197
+ cacheRead: 0.19,
14266
15198
  cacheWrite: 0,
14267
15199
  },
14268
- contextWindow: 262114,
14269
- maxTokens: 262114,
15200
+ contextWindow: 256000,
15201
+ maxTokens: 32768,
14270
15202
  },
14271
- "moonshotai/kimi-k2.6": {
14272
- id: "moonshotai/kimi-k2.6",
14273
- name: "Kimi K2.6",
15203
+ "moonshotai/kimi-k2.7-code-highspeed": {
15204
+ id: "moonshotai/kimi-k2.7-code-highspeed",
15205
+ name: "Kimi K2.7 Code High Speed",
14274
15206
  api: "anthropic-messages",
14275
15207
  provider: "vercel-ai-gateway",
14276
15208
  baseUrl: "https://ai-gateway.vercel.sh",
14277
15209
  reasoning: true,
14278
15210
  input: ["text", "image"],
14279
15211
  cost: {
14280
- input: 0.95,
14281
- output: 4,
14282
- cacheRead: 0.16,
15212
+ input: 1.9,
15213
+ output: 8,
15214
+ cacheRead: 0.38,
14283
15215
  cacheWrite: 0,
14284
15216
  },
14285
- contextWindow: 262000,
14286
- maxTokens: 262000,
15217
+ contextWindow: 262144,
15218
+ maxTokens: 32768,
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,
14287
15236
  },
14288
15237
  "nvidia/nemotron-3-super-120b-a12b": {
14289
15238
  id: "nvidia/nemotron-3-super-120b-a12b",
@@ -14353,6 +15302,23 @@ export const MODELS = {
14353
15302
  contextWindow: 131072,
14354
15303
  maxTokens: 131072,
14355
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
+ },
14356
15322
  "openai/gpt-4.1": {
14357
15323
  id: "openai/gpt-4.1",
14358
15324
  name: "GPT-4.1",
@@ -14461,7 +15427,7 @@ export const MODELS = {
14461
15427
  api: "anthropic-messages",
14462
15428
  provider: "vercel-ai-gateway",
14463
15429
  baseUrl: "https://ai-gateway.vercel.sh",
14464
- reasoning: true,
15430
+ reasoning: false,
14465
15431
  input: ["text", "image"],
14466
15432
  cost: {
14467
15433
  input: 1.25,
@@ -14597,7 +15563,7 @@ export const MODELS = {
14597
15563
  api: "anthropic-messages",
14598
15564
  provider: "vercel-ai-gateway",
14599
15565
  baseUrl: "https://ai-gateway.vercel.sh",
14600
- reasoning: true,
15566
+ reasoning: false,
14601
15567
  input: ["text", "image"],
14602
15568
  cost: {
14603
15569
  input: 1.25,
@@ -14649,7 +15615,7 @@ export const MODELS = {
14649
15615
  api: "anthropic-messages",
14650
15616
  provider: "vercel-ai-gateway",
14651
15617
  baseUrl: "https://ai-gateway.vercel.sh",
14652
- reasoning: true,
15618
+ reasoning: false,
14653
15619
  thinkingLevelMap: { "xhigh": "xhigh" },
14654
15620
  input: ["text", "image"],
14655
15621
  cost: {
@@ -14703,7 +15669,7 @@ export const MODELS = {
14703
15669
  api: "anthropic-messages",
14704
15670
  provider: "vercel-ai-gateway",
14705
15671
  baseUrl: "https://ai-gateway.vercel.sh",
14706
- reasoning: true,
15672
+ reasoning: false,
14707
15673
  thinkingLevelMap: { "xhigh": "xhigh" },
14708
15674
  input: ["text", "image"],
14709
15675
  cost: {
@@ -14850,13 +15816,13 @@ export const MODELS = {
14850
15816
  reasoning: true,
14851
15817
  input: ["text"],
14852
15818
  cost: {
14853
- input: 0.35,
14854
- output: 0.75,
14855
- cacheRead: 0.25,
15819
+ input: 0.09999999999999999,
15820
+ output: 0.5,
15821
+ cacheRead: 0,
14856
15822
  cacheWrite: 0,
14857
15823
  },
14858
15824
  contextWindow: 131072,
14859
- maxTokens: 131000,
15825
+ maxTokens: 131072,
14860
15826
  },
14861
15827
  "openai/gpt-oss-20b": {
14862
15828
  id: "openai/gpt-oss-20b",
@@ -14994,39 +15960,22 @@ export const MODELS = {
14994
15960
  contextWindow: 200000,
14995
15961
  maxTokens: 100000,
14996
15962
  },
14997
- "perplexity/sonar": {
14998
- id: "perplexity/sonar",
14999
- name: "Sonar",
15000
- api: "anthropic-messages",
15001
- provider: "vercel-ai-gateway",
15002
- baseUrl: "https://ai-gateway.vercel.sh",
15003
- reasoning: false,
15004
- input: ["text", "image"],
15005
- cost: {
15006
- input: 0,
15007
- output: 0,
15008
- cacheRead: 0,
15009
- cacheWrite: 0,
15010
- },
15011
- contextWindow: 127000,
15012
- maxTokens: 8000,
15013
- },
15014
- "perplexity/sonar-pro": {
15015
- id: "perplexity/sonar-pro",
15016
- name: "Sonar Pro",
15963
+ "sakana/fugu-ultra": {
15964
+ id: "sakana/fugu-ultra",
15965
+ name: "Fugu Ultra",
15017
15966
  api: "anthropic-messages",
15018
15967
  provider: "vercel-ai-gateway",
15019
15968
  baseUrl: "https://ai-gateway.vercel.sh",
15020
- reasoning: false,
15969
+ reasoning: true,
15021
15970
  input: ["text", "image"],
15022
15971
  cost: {
15023
- input: 0,
15024
- output: 0,
15025
- cacheRead: 0,
15972
+ input: 5,
15973
+ output: 30,
15974
+ cacheRead: 0.5,
15026
15975
  cacheWrite: 0,
15027
15976
  },
15028
- contextWindow: 200000,
15029
- maxTokens: 8000,
15977
+ contextWindow: 1000000,
15978
+ maxTokens: 1000000,
15030
15979
  },
15031
15980
  "stepfun/step-3.5-flash": {
15032
15981
  id: "stepfun/step-3.5-flash",
@@ -15039,8 +15988,8 @@ export const MODELS = {
15039
15988
  cost: {
15040
15989
  input: 0.09,
15041
15990
  output: 0.3,
15042
- cacheRead: 0,
15043
- cacheWrite: 0.02,
15991
+ cacheRead: 0.02,
15992
+ cacheWrite: 0,
15044
15993
  },
15045
15994
  contextWindow: 262114,
15046
15995
  maxTokens: 262114,
@@ -15256,7 +16205,7 @@ export const MODELS = {
15256
16205
  provider: "vercel-ai-gateway",
15257
16206
  baseUrl: "https://ai-gateway.vercel.sh",
15258
16207
  reasoning: true,
15259
- input: ["text", "image"],
16208
+ input: ["text"],
15260
16209
  cost: {
15261
16210
  input: 0.435,
15262
16211
  output: 0.87,
@@ -15377,13 +16326,13 @@ export const MODELS = {
15377
16326
  reasoning: true,
15378
16327
  input: ["text"],
15379
16328
  cost: {
15380
- input: 2.25,
15381
- output: 2.75,
15382
- cacheRead: 2.25,
16329
+ input: 0.6,
16330
+ output: 2.2,
16331
+ cacheRead: 0.12,
15383
16332
  cacheWrite: 0,
15384
16333
  },
15385
- contextWindow: 131000,
15386
- maxTokens: 40000,
16334
+ contextWindow: 200000,
16335
+ maxTokens: 120000,
15387
16336
  },
15388
16337
  "zai/glm-4.7-flash": {
15389
16338
  id: "zai/glm-4.7-flash",
@@ -15428,8 +16377,8 @@ export const MODELS = {
15428
16377
  reasoning: true,
15429
16378
  input: ["text"],
15430
16379
  cost: {
15431
- input: 1,
15432
- output: 3.1999999999999997,
16380
+ input: 0.95,
16381
+ output: 3.15,
15433
16382
  cacheRead: 0.19999999999999998,
15434
16383
  cacheWrite: 0,
15435
16384
  },
@@ -15460,15 +16409,49 @@ export const MODELS = {
15460
16409
  provider: "vercel-ai-gateway",
15461
16410
  baseUrl: "https://ai-gateway.vercel.sh",
15462
16411
  reasoning: true,
15463
- input: ["text", "image"],
16412
+ input: ["text"],
15464
16413
  cost: {
15465
- input: 1.4,
15466
- output: 4.4,
16414
+ input: 1.3,
16415
+ output: 4.300000000000001,
15467
16416
  cacheRead: 0.26,
15468
16417
  cacheWrite: 0,
15469
16418
  },
15470
- contextWindow: 202800,
15471
- maxTokens: 64000,
16419
+ contextWindow: 202000,
16420
+ maxTokens: 202000,
16421
+ },
16422
+ "zai/glm-5.2": {
16423
+ id: "zai/glm-5.2",
16424
+ name: "GLM 5.2",
16425
+ api: "anthropic-messages",
16426
+ provider: "vercel-ai-gateway",
16427
+ baseUrl: "https://ai-gateway.vercel.sh",
16428
+ reasoning: true,
16429
+ input: ["text"],
16430
+ cost: {
16431
+ input: 1.5,
16432
+ output: 4.5,
16433
+ cacheRead: 0.3,
16434
+ cacheWrite: 0,
16435
+ },
16436
+ contextWindow: 1000000,
16437
+ maxTokens: 128000,
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,
15472
16455
  },
15473
16456
  "zai/glm-5v-turbo": {
15474
16457
  id: "zai/glm-5v-turbo",
@@ -15860,6 +16843,24 @@ export const MODELS = {
15860
16843
  contextWindow: 200000,
15861
16844
  maxTokens: 131072,
15862
16845
  },
16846
+ "glm-5.2": {
16847
+ id: "glm-5.2",
16848
+ name: "GLM-5.2",
16849
+ api: "openai-completions",
16850
+ provider: "zai",
16851
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
16852
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16853
+ reasoning: true,
16854
+ input: ["text"],
16855
+ cost: {
16856
+ input: 0,
16857
+ output: 0,
16858
+ cacheRead: 0,
16859
+ cacheWrite: 0,
16860
+ },
16861
+ contextWindow: 1000000,
16862
+ maxTokens: 131072,
16863
+ },
15863
16864
  "glm-5v-turbo": {
15864
16865
  id: "glm-5v-turbo",
15865
16866
  name: "GLM-5V-Turbo",