@elyracode/ai 0.9.9 → 0.9.10

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: 1048576,
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,
@@ -3912,8 +3948,45 @@ export const MODELS = {
3912
3948
  contextWindow: 262000,
3913
3949
  maxTokens: 262000,
3914
3950
  },
3951
+ "accounts/fireworks/routers/kimi-k2p7-code-fast": {
3952
+ id: "accounts/fireworks/routers/kimi-k2p7-code-fast",
3953
+ name: "Kimi K2.7 Code Fast",
3954
+ api: "anthropic-messages",
3955
+ provider: "fireworks",
3956
+ baseUrl: "https://api.fireworks.ai/inference",
3957
+ reasoning: true,
3958
+ input: ["text", "image"],
3959
+ cost: {
3960
+ input: 1.9,
3961
+ output: 8,
3962
+ cacheRead: 0.38,
3963
+ cacheWrite: 0,
3964
+ },
3965
+ contextWindow: 262000,
3966
+ maxTokens: 262000,
3967
+ },
3915
3968
  },
3916
3969
  "github-copilot": {
3970
+ "claude-fable-5": {
3971
+ id: "claude-fable-5",
3972
+ name: "Claude Fable 5",
3973
+ api: "openai-completions",
3974
+ provider: "github-copilot",
3975
+ baseUrl: "https://api.individual.githubcopilot.com",
3976
+ 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" },
3977
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3978
+ reasoning: true,
3979
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
3980
+ input: ["text", "image"],
3981
+ cost: {
3982
+ input: 10,
3983
+ output: 50,
3984
+ cacheRead: 1,
3985
+ cacheWrite: 12.5,
3986
+ },
3987
+ contextWindow: 1000000,
3988
+ maxTokens: 128000,
3989
+ },
3917
3990
  "claude-haiku-4.5": {
3918
3991
  id: "claude-haiku-4.5",
3919
3992
  name: "Claude Haiku 4.5 (latest)",
@@ -4311,25 +4384,6 @@ export const MODELS = {
4311
4384
  contextWindow: 400000,
4312
4385
  maxTokens: 128000,
4313
4386
  },
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
4387
  },
4334
4388
  "google": {
4335
4389
  "gemini-2.0-flash": {
@@ -4613,6 +4667,42 @@ export const MODELS = {
4613
4667
  contextWindow: 262144,
4614
4668
  maxTokens: 32768,
4615
4669
  },
4670
+ "gemma-4-E2B-it": {
4671
+ id: "gemma-4-E2B-it",
4672
+ name: "Gemma 4 E2B IT",
4673
+ api: "google-generative-ai",
4674
+ provider: "google",
4675
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4676
+ reasoning: true,
4677
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4678
+ input: ["text", "image"],
4679
+ cost: {
4680
+ input: 0,
4681
+ output: 0,
4682
+ cacheRead: 0,
4683
+ cacheWrite: 0,
4684
+ },
4685
+ contextWindow: 131072,
4686
+ maxTokens: 8192,
4687
+ },
4688
+ "gemma-4-E4B-it": {
4689
+ id: "gemma-4-E4B-it",
4690
+ name: "Gemma 4 E4B IT",
4691
+ api: "google-generative-ai",
4692
+ provider: "google",
4693
+ baseUrl: "https://generativelanguage.googleapis.com/v1beta",
4694
+ reasoning: true,
4695
+ thinkingLevelMap: { "off": null, "minimal": "MINIMAL", "low": null, "medium": null, "high": "HIGH" },
4696
+ input: ["text", "image"],
4697
+ cost: {
4698
+ input: 0,
4699
+ output: 0,
4700
+ cacheRead: 0,
4701
+ cacheWrite: 0,
4702
+ },
4703
+ contextWindow: 131072,
4704
+ maxTokens: 8192,
4705
+ },
4616
4706
  },
4617
4707
  "google-vertex": {
4618
4708
  "gemini-1.5-flash": {
@@ -5344,6 +5434,24 @@ export const MODELS = {
5344
5434
  },
5345
5435
  },
5346
5436
  "kimi-coding": {
5437
+ "k2p7": {
5438
+ id: "k2p7",
5439
+ name: "Kimi K2.7 Code",
5440
+ api: "anthropic-messages",
5441
+ provider: "kimi-coding",
5442
+ baseUrl: "https://api.kimi.com/coding",
5443
+ headers: { "User-Agent": "KimiCLI/1.5" },
5444
+ reasoning: true,
5445
+ input: ["text", "image"],
5446
+ cost: {
5447
+ input: 0,
5448
+ output: 0,
5449
+ cacheRead: 0,
5450
+ cacheWrite: 0,
5451
+ },
5452
+ contextWindow: 262144,
5453
+ maxTokens: 32768,
5454
+ },
5347
5455
  "kimi-for-coding": {
5348
5456
  id: "kimi-for-coding",
5349
5457
  name: "Kimi For Coding",
@@ -6126,6 +6234,42 @@ export const MODELS = {
6126
6234
  contextWindow: 262144,
6127
6235
  maxTokens: 262144,
6128
6236
  },
6237
+ "kimi-k2.7-code": {
6238
+ id: "kimi-k2.7-code",
6239
+ name: "Kimi K2.7 Code",
6240
+ api: "openai-completions",
6241
+ provider: "moonshotai",
6242
+ baseUrl: "https://api.moonshot.ai/v1",
6243
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6244
+ reasoning: true,
6245
+ input: ["text", "image"],
6246
+ cost: {
6247
+ input: 0.95,
6248
+ output: 4,
6249
+ cacheRead: 0.19,
6250
+ cacheWrite: 0,
6251
+ },
6252
+ contextWindow: 262144,
6253
+ maxTokens: 262144,
6254
+ },
6255
+ "kimi-k2.7-code-highspeed": {
6256
+ id: "kimi-k2.7-code-highspeed",
6257
+ name: "Kimi K2.7 Code HighSpeed",
6258
+ api: "openai-completions",
6259
+ provider: "moonshotai",
6260
+ baseUrl: "https://api.moonshot.ai/v1",
6261
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6262
+ reasoning: true,
6263
+ input: ["text", "image"],
6264
+ cost: {
6265
+ input: 1.9,
6266
+ output: 8,
6267
+ cacheRead: 0.38,
6268
+ cacheWrite: 0,
6269
+ },
6270
+ contextWindow: 262144,
6271
+ maxTokens: 262144,
6272
+ },
6129
6273
  },
6130
6274
  "moonshotai-cn": {
6131
6275
  "kimi-k2-0711-preview": {
@@ -6254,6 +6398,42 @@ export const MODELS = {
6254
6398
  contextWindow: 262144,
6255
6399
  maxTokens: 262144,
6256
6400
  },
6401
+ "kimi-k2.7-code": {
6402
+ id: "kimi-k2.7-code",
6403
+ name: "Kimi K2.7 Code",
6404
+ api: "openai-completions",
6405
+ provider: "moonshotai-cn",
6406
+ baseUrl: "https://api.moonshot.cn/v1",
6407
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6408
+ reasoning: true,
6409
+ input: ["text", "image"],
6410
+ cost: {
6411
+ input: 0.95,
6412
+ output: 4,
6413
+ cacheRead: 0.19,
6414
+ cacheWrite: 0,
6415
+ },
6416
+ contextWindow: 262144,
6417
+ maxTokens: 262144,
6418
+ },
6419
+ "kimi-k2.7-code-highspeed": {
6420
+ id: "kimi-k2.7-code-highspeed",
6421
+ name: "Kimi K2.7 Code HighSpeed",
6422
+ api: "openai-completions",
6423
+ provider: "moonshotai-cn",
6424
+ baseUrl: "https://api.moonshot.cn/v1",
6425
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
6426
+ reasoning: true,
6427
+ input: ["text", "image"],
6428
+ cost: {
6429
+ input: 1.9,
6430
+ output: 8,
6431
+ cacheRead: 0.38,
6432
+ cacheWrite: 0,
6433
+ },
6434
+ contextWindow: 262144,
6435
+ maxTokens: 262144,
6436
+ },
6257
6437
  },
6258
6438
  "openai": {
6259
6439
  "gpt-4": {
@@ -7125,24 +7305,6 @@ export const MODELS = {
7125
7305
  contextWindow: 200000,
7126
7306
  maxTokens: 32000,
7127
7307
  },
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
7308
  "claude-haiku-4-5": {
7147
7309
  id: "claude-haiku-4-5",
7148
7310
  name: "Claude Haiku 4.5",
@@ -7942,26 +8104,26 @@ export const MODELS = {
7942
8104
  contextWindow: 1000000,
7943
8105
  maxTokens: 384000,
7944
8106
  },
7945
- "glm-5": {
7946
- id: "glm-5",
7947
- name: "GLM-5",
8107
+ "glm-5.1": {
8108
+ id: "glm-5.1",
8109
+ name: "GLM-5.1",
7948
8110
  api: "openai-completions",
7949
8111
  provider: "opencode-go",
7950
8112
  baseUrl: "https://opencode.ai/zen/go/v1",
7951
8113
  reasoning: true,
7952
8114
  input: ["text"],
7953
8115
  cost: {
7954
- input: 1,
7955
- output: 3.2,
7956
- cacheRead: 0.2,
8116
+ input: 1.4,
8117
+ output: 4.4,
8118
+ cacheRead: 0.26,
7957
8119
  cacheWrite: 0,
7958
8120
  },
7959
8121
  contextWindow: 202752,
7960
8122
  maxTokens: 32768,
7961
8123
  },
7962
- "glm-5.1": {
7963
- id: "glm-5.1",
7964
- name: "GLM-5.1",
8124
+ "glm-5.2": {
8125
+ id: "glm-5.2",
8126
+ name: "GLM-5.2",
7965
8127
  api: "openai-completions",
7966
8128
  provider: "opencode-go",
7967
8129
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -7973,29 +8135,29 @@ export const MODELS = {
7973
8135
  cacheRead: 0.26,
7974
8136
  cacheWrite: 0,
7975
8137
  },
7976
- contextWindow: 202752,
7977
- maxTokens: 32768,
8138
+ contextWindow: 1000000,
8139
+ maxTokens: 131072,
7978
8140
  },
7979
- "kimi-k2.5": {
7980
- id: "kimi-k2.5",
7981
- name: "Kimi K2.5",
8141
+ "kimi-k2.6": {
8142
+ id: "kimi-k2.6",
8143
+ name: "Kimi K2.6",
7982
8144
  api: "openai-completions",
7983
8145
  provider: "opencode-go",
7984
8146
  baseUrl: "https://opencode.ai/zen/go/v1",
7985
8147
  reasoning: true,
7986
8148
  input: ["text", "image"],
7987
8149
  cost: {
7988
- input: 0.6,
7989
- output: 3,
7990
- cacheRead: 0.1,
8150
+ input: 0.95,
8151
+ output: 4,
8152
+ cacheRead: 0.16,
7991
8153
  cacheWrite: 0,
7992
8154
  },
7993
8155
  contextWindow: 262144,
7994
8156
  maxTokens: 65536,
7995
8157
  },
7996
- "kimi-k2.6": {
7997
- id: "kimi-k2.6",
7998
- name: "Kimi K2.6",
8158
+ "kimi-k2.7-code": {
8159
+ id: "kimi-k2.7-code",
8160
+ name: "Kimi K2.7 Code",
7999
8161
  api: "openai-completions",
8000
8162
  provider: "opencode-go",
8001
8163
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8004,11 +8166,11 @@ export const MODELS = {
8004
8166
  cost: {
8005
8167
  input: 0.95,
8006
8168
  output: 4,
8007
- cacheRead: 0.16,
8169
+ cacheRead: 0.19,
8008
8170
  cacheWrite: 0,
8009
8171
  },
8010
8172
  contextWindow: 262144,
8011
- maxTokens: 65536,
8173
+ maxTokens: 262144,
8012
8174
  },
8013
8175
  "mimo-v2.5": {
8014
8176
  id: "mimo-v2.5",
@@ -8044,23 +8206,6 @@ export const MODELS = {
8044
8206
  contextWindow: 1048576,
8045
8207
  maxTokens: 128000,
8046
8208
  },
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
8209
  "minimax-m2.7": {
8065
8210
  id: "minimax-m2.7",
8066
8211
  name: "MiniMax M2.7",
@@ -8080,16 +8225,16 @@ export const MODELS = {
8080
8225
  },
8081
8226
  "minimax-m3": {
8082
8227
  id: "minimax-m3",
8083
- name: "MiniMax M3",
8228
+ name: "MiniMax M3 (3x usage)",
8084
8229
  api: "anthropic-messages",
8085
8230
  provider: "opencode-go",
8086
8231
  baseUrl: "https://opencode.ai/zen/go",
8087
8232
  reasoning: true,
8088
8233
  input: ["text", "image"],
8089
8234
  cost: {
8090
- input: 0.3,
8091
- output: 1.2,
8092
- cacheRead: 0.06,
8235
+ input: 0.1,
8236
+ output: 0.4,
8237
+ cacheRead: 0.02,
8093
8238
  cacheWrite: 0,
8094
8239
  },
8095
8240
  contextWindow: 512000,
@@ -8683,6 +8828,23 @@ export const MODELS = {
8683
8828
  contextWindow: 128000,
8684
8829
  maxTokens: 4000,
8685
8830
  },
8831
+ "cohere/north-mini-code:free": {
8832
+ id: "cohere/north-mini-code:free",
8833
+ name: "Cohere: North Mini Code (free)",
8834
+ api: "openai-completions",
8835
+ provider: "openrouter",
8836
+ baseUrl: "https://openrouter.ai/api/v1",
8837
+ reasoning: true,
8838
+ input: ["text"],
8839
+ cost: {
8840
+ input: 0,
8841
+ output: 0,
8842
+ cacheRead: 0,
8843
+ cacheWrite: 0,
8844
+ },
8845
+ contextWindow: 256000,
8846
+ maxTokens: 64000,
8847
+ },
8686
8848
  "deepseek/deepseek-chat": {
8687
8849
  id: "deepseek/deepseek-chat",
8688
8850
  name: "DeepSeek: DeepSeek V3",
@@ -8714,7 +8876,7 @@ export const MODELS = {
8714
8876
  cacheRead: 0.135,
8715
8877
  cacheWrite: 0,
8716
8878
  },
8717
- contextWindow: 131072,
8879
+ contextWindow: 163840,
8718
8880
  maxTokens: 16384,
8719
8881
  },
8720
8882
  "deepseek/deepseek-chat-v3.1": {
@@ -8830,13 +8992,13 @@ export const MODELS = {
8830
8992
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8831
8993
  input: ["text"],
8832
8994
  cost: {
8833
- input: 0.0983,
8834
- output: 0.1966,
8835
- cacheRead: 0.019700000000000002,
8995
+ input: 0.09,
8996
+ output: 0.18,
8997
+ cacheRead: 0.02,
8836
8998
  cacheWrite: 0,
8837
8999
  },
8838
9000
  contextWindow: 1048576,
8839
- maxTokens: 131072,
9001
+ maxTokens: 65536,
8840
9002
  },
8841
9003
  "deepseek/deepseek-v4-pro": {
8842
9004
  id: "deepseek/deepseek-v4-pro",
@@ -8991,7 +9153,24 @@ export const MODELS = {
8991
9153
  cacheWrite: 0.08333333333333334,
8992
9154
  },
8993
9155
  contextWindow: 1048576,
8994
- maxTokens: 65536,
9156
+ maxTokens: 65535,
9157
+ },
9158
+ "google/gemini-3-pro-image": {
9159
+ id: "google/gemini-3-pro-image",
9160
+ name: "Google: Nano Banana Pro (Gemini 3 Pro Image)",
9161
+ api: "openai-completions",
9162
+ provider: "openrouter",
9163
+ baseUrl: "https://openrouter.ai/api/v1",
9164
+ reasoning: true,
9165
+ input: ["text", "image"],
9166
+ cost: {
9167
+ input: 2,
9168
+ output: 12,
9169
+ cacheRead: 0.19999999999999998,
9170
+ cacheWrite: 0.375,
9171
+ },
9172
+ contextWindow: 65536,
9173
+ maxTokens: 32768,
8995
9174
  },
8996
9175
  "google/gemini-3.1-flash-lite": {
8997
9176
  id: "google/gemini-3.1-flash-lite",
@@ -9156,12 +9335,12 @@ export const MODELS = {
9156
9335
  input: ["text", "image"],
9157
9336
  cost: {
9158
9337
  input: 0.12,
9159
- output: 0.36,
9338
+ output: 0.35,
9160
9339
  cacheRead: 0.09,
9161
9340
  cacheWrite: 0,
9162
9341
  },
9163
9342
  contextWindow: 262144,
9164
- maxTokens: 8192,
9343
+ maxTokens: 262144,
9165
9344
  },
9166
9345
  "google/gemma-4-31b-it:free": {
9167
9346
  id: "google/gemma-4-31b-it:free",
@@ -9178,7 +9357,7 @@ export const MODELS = {
9178
9357
  cacheWrite: 0,
9179
9358
  },
9180
9359
  contextWindow: 262144,
9181
- maxTokens: 32768,
9360
+ maxTokens: 8192,
9182
9361
  },
9183
9362
  "ibm-granite/granite-4.1-8b": {
9184
9363
  id: "ibm-granite/granite-4.1-8b",
@@ -9282,6 +9461,23 @@ export const MODELS = {
9282
9461
  contextWindow: 256000,
9283
9462
  maxTokens: 80000,
9284
9463
  },
9464
+ "liquid/lfm-2.5-1.2b-thinking:free": {
9465
+ id: "liquid/lfm-2.5-1.2b-thinking:free",
9466
+ name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
9467
+ api: "openai-completions",
9468
+ provider: "openrouter",
9469
+ baseUrl: "https://openrouter.ai/api/v1",
9470
+ reasoning: true,
9471
+ input: ["text"],
9472
+ cost: {
9473
+ input: 0,
9474
+ output: 0,
9475
+ cacheRead: 0,
9476
+ cacheWrite: 0,
9477
+ },
9478
+ contextWindow: 32768,
9479
+ maxTokens: 4096,
9480
+ },
9285
9481
  "meta-llama/llama-3.1-70b-instruct": {
9286
9482
  id: "meta-llama/llama-3.1-70b-instruct",
9287
9483
  name: "Meta: Llama 3.1 70B Instruct",
@@ -9461,9 +9657,9 @@ export const MODELS = {
9461
9657
  reasoning: true,
9462
9658
  input: ["text"],
9463
9659
  cost: {
9464
- input: 0.27,
9465
- output: 1.08,
9466
- cacheRead: 0.054,
9660
+ input: 0.25,
9661
+ output: 1,
9662
+ cacheRead: 0.049999999999999996,
9467
9663
  cacheWrite: 0,
9468
9664
  },
9469
9665
  contextWindow: 204800,
@@ -9852,13 +10048,30 @@ export const MODELS = {
9852
10048
  reasoning: true,
9853
10049
  input: ["text", "image"],
9854
10050
  cost: {
9855
- input: 0.6799999999999999,
9856
- output: 3.41,
9857
- cacheRead: 0.33999999999999997,
10051
+ input: 0.67,
10052
+ output: 3.5,
10053
+ cacheRead: 0.19999999999999998,
10054
+ cacheWrite: 0,
10055
+ },
10056
+ contextWindow: 262144,
10057
+ maxTokens: 262144,
10058
+ },
10059
+ "moonshotai/kimi-k2.7-code": {
10060
+ id: "moonshotai/kimi-k2.7-code",
10061
+ name: "MoonshotAI: Kimi K2.7 Code",
10062
+ api: "openai-completions",
10063
+ provider: "openrouter",
10064
+ baseUrl: "https://openrouter.ai/api/v1",
10065
+ reasoning: true,
10066
+ input: ["text", "image"],
10067
+ cost: {
10068
+ input: 0.74,
10069
+ output: 3.5,
10070
+ cacheRead: 0.15,
9858
10071
  cacheWrite: 0,
9859
10072
  },
9860
10073
  contextWindow: 262144,
9861
- maxTokens: 262142,
10074
+ maxTokens: 16384,
9862
10075
  },
9863
10076
  "nex-agi/nex-n2-pro:free": {
9864
10077
  id: "nex-agi/nex-n2-pro:free",
@@ -9989,8 +10202,8 @@ export const MODELS = {
9989
10202
  input: ["text"],
9990
10203
  cost: {
9991
10204
  input: 0.5,
9992
- output: 2.5,
9993
- cacheRead: 0.15,
10205
+ output: 2.2,
10206
+ cacheRead: 0.09999999999999999,
9994
10207
  cacheWrite: 0,
9995
10208
  },
9996
10209
  contextWindow: 1000000,
@@ -10030,23 +10243,6 @@ export const MODELS = {
10030
10243
  contextWindow: 128000,
10031
10244
  maxTokens: 128000,
10032
10245
  },
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
10246
  "nvidia/nemotron-nano-9b-v2:free": {
10051
10247
  id: "nvidia/nemotron-nano-9b-v2:free",
10052
10248
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -10720,7 +10916,7 @@ export const MODELS = {
10720
10916
  cacheWrite: 0,
10721
10917
  },
10722
10918
  contextWindow: 131072,
10723
- maxTokens: 8192,
10919
+ maxTokens: 32768,
10724
10920
  },
10725
10921
  "openai/gpt-oss-safeguard-20b": {
10726
10922
  id: "openai/gpt-oss-safeguard-20b",
@@ -10733,7 +10929,7 @@ export const MODELS = {
10733
10929
  cost: {
10734
10930
  input: 0.075,
10735
10931
  output: 0.3,
10736
- cacheRead: 0.037,
10932
+ cacheRead: 0.0375,
10737
10933
  cacheWrite: 0,
10738
10934
  },
10739
10935
  contextWindow: 131072,
@@ -10943,6 +11139,23 @@ export const MODELS = {
10943
11139
  contextWindow: 1048756,
10944
11140
  maxTokens: 262144,
10945
11141
  },
11142
+ "poolside/laguna-m.1": {
11143
+ id: "poolside/laguna-m.1",
11144
+ name: "Poolside: Laguna M.1",
11145
+ api: "openai-completions",
11146
+ provider: "openrouter",
11147
+ baseUrl: "https://openrouter.ai/api/v1",
11148
+ reasoning: true,
11149
+ input: ["text"],
11150
+ cost: {
11151
+ input: 0.19999999999999998,
11152
+ output: 0.39999999999999997,
11153
+ cacheRead: 0.09999999999999999,
11154
+ cacheWrite: 0,
11155
+ },
11156
+ contextWindow: 262144,
11157
+ maxTokens: 32768,
11158
+ },
10946
11159
  "poolside/laguna-m.1:free": {
10947
11160
  id: "poolside/laguna-m.1:free",
10948
11161
  name: "Poolside: Laguna M.1 (free)",
@@ -10960,6 +11173,23 @@ export const MODELS = {
10960
11173
  contextWindow: 262144,
10961
11174
  maxTokens: 32768,
10962
11175
  },
11176
+ "poolside/laguna-xs.2": {
11177
+ id: "poolside/laguna-xs.2",
11178
+ name: "Poolside: Laguna XS.2",
11179
+ api: "openai-completions",
11180
+ provider: "openrouter",
11181
+ baseUrl: "https://openrouter.ai/api/v1",
11182
+ reasoning: true,
11183
+ input: ["text"],
11184
+ cost: {
11185
+ input: 0.09999999999999999,
11186
+ output: 0.19999999999999998,
11187
+ cacheRead: 0.049999999999999996,
11188
+ cacheWrite: 0,
11189
+ },
11190
+ contextWindow: 262144,
11191
+ maxTokens: 32768,
11192
+ },
10963
11193
  "poolside/laguna-xs.2:free": {
10964
11194
  id: "poolside/laguna-xs.2:free",
10965
11195
  name: "Poolside: Laguna XS.2 (free)",
@@ -11011,6 +11241,23 @@ export const MODELS = {
11011
11241
  contextWindow: 131072,
11012
11242
  maxTokens: 16384,
11013
11243
  },
11244
+ "qwen/qwen-2.5-7b-instruct": {
11245
+ id: "qwen/qwen-2.5-7b-instruct",
11246
+ name: "Qwen: Qwen2.5 7B Instruct",
11247
+ api: "openai-completions",
11248
+ provider: "openrouter",
11249
+ baseUrl: "https://openrouter.ai/api/v1",
11250
+ reasoning: false,
11251
+ input: ["text"],
11252
+ cost: {
11253
+ input: 0.04,
11254
+ output: 0.09999999999999999,
11255
+ cacheRead: 0,
11256
+ cacheWrite: 0,
11257
+ },
11258
+ contextWindow: 131072,
11259
+ maxTokens: 32768,
11260
+ },
11014
11261
  "qwen/qwen-plus": {
11015
11262
  id: "qwen/qwen-plus",
11016
11263
  name: "Qwen: Qwen-Plus",
@@ -11566,7 +11813,7 @@ export const MODELS = {
11566
11813
  cost: {
11567
11814
  input: 0.14,
11568
11815
  output: 1,
11569
- cacheRead: 0.049999999999999996,
11816
+ cacheRead: 0,
11570
11817
  cacheWrite: 0,
11571
11818
  },
11572
11819
  contextWindow: 262144,
@@ -11581,13 +11828,13 @@ export const MODELS = {
11581
11828
  reasoning: true,
11582
11829
  input: ["text", "image"],
11583
11830
  cost: {
11584
- input: 0.39,
11585
- output: 2.34,
11831
+ input: 0.385,
11832
+ output: 2.4499999999999997,
11586
11833
  cacheRead: 0,
11587
11834
  cacheWrite: 0,
11588
11835
  },
11589
- contextWindow: 262144,
11590
- maxTokens: 65536,
11836
+ contextWindow: 256000,
11837
+ maxTokens: 4096,
11591
11838
  },
11592
11839
  "qwen/qwen3.5-9b": {
11593
11840
  id: "qwen/qwen3.5-9b",
@@ -11666,13 +11913,13 @@ export const MODELS = {
11666
11913
  reasoning: true,
11667
11914
  input: ["text", "image"],
11668
11915
  cost: {
11669
- input: 0.28900000000000003,
11670
- output: 2.4,
11916
+ input: 0.28850000000000003,
11917
+ output: 3.17,
11671
11918
  cacheRead: 0,
11672
11919
  cacheWrite: 0,
11673
11920
  },
11674
11921
  contextWindow: 262144,
11675
- maxTokens: 131072,
11922
+ maxTokens: 262140,
11676
11923
  },
11677
11924
  "qwen/qwen3.6-35b-a3b": {
11678
11925
  id: "qwen/qwen3.6-35b-a3b",
@@ -11683,9 +11930,9 @@ export const MODELS = {
11683
11930
  reasoning: true,
11684
11931
  input: ["text", "image"],
11685
11932
  cost: {
11686
- input: 0.15,
11933
+ input: 0.14,
11687
11934
  output: 1,
11688
- cacheRead: 0.049999999999999996,
11935
+ cacheRead: 0,
11689
11936
  cacheWrite: 0,
11690
11937
  },
11691
11938
  contextWindow: 262144,
@@ -11768,10 +12015,10 @@ export const MODELS = {
11768
12015
  reasoning: true,
11769
12016
  input: ["text", "image"],
11770
12017
  cost: {
11771
- input: 0.39999999999999997,
11772
- output: 1.5999999999999999,
11773
- cacheRead: 0.08,
11774
- cacheWrite: 0.5,
12018
+ input: 0.32,
12019
+ output: 1.28,
12020
+ cacheRead: 0.064,
12021
+ cacheWrite: 0.39999999999999997,
11775
12022
  },
11776
12023
  contextWindow: 1000000,
11777
12024
  maxTokens: 65536,
@@ -12023,13 +12270,13 @@ export const MODELS = {
12023
12270
  reasoning: true,
12024
12271
  input: ["text"],
12025
12272
  cost: {
12026
- input: 0.125,
12273
+ input: 0.13,
12027
12274
  output: 0.85,
12028
- cacheRead: 0.06,
12275
+ cacheRead: 0.024999999999999998,
12029
12276
  cacheWrite: 0,
12030
12277
  },
12031
12278
  contextWindow: 131072,
12032
- maxTokens: 131070,
12279
+ maxTokens: 98304,
12033
12280
  },
12034
12281
  "z-ai/glm-4.5v": {
12035
12282
  id: "z-ai/glm-4.5v",
@@ -12167,6 +12414,23 @@ export const MODELS = {
12167
12414
  contextWindow: 202752,
12168
12415
  maxTokens: 4096,
12169
12416
  },
12417
+ "z-ai/glm-5.2": {
12418
+ id: "z-ai/glm-5.2",
12419
+ name: "Z.ai: GLM 5.2",
12420
+ api: "openai-completions",
12421
+ provider: "openrouter",
12422
+ baseUrl: "https://openrouter.ai/api/v1",
12423
+ reasoning: true,
12424
+ input: ["text"],
12425
+ cost: {
12426
+ input: 1.2,
12427
+ output: 4.1,
12428
+ cacheRead: 0.19999999999999998,
12429
+ cacheWrite: 0,
12430
+ },
12431
+ contextWindow: 1048576,
12432
+ maxTokens: 131072,
12433
+ },
12170
12434
  "~anthropic/claude-fable-latest": {
12171
12435
  id: "~anthropic/claude-fable-latest",
12172
12436
  name: "Anthropic: Claude Fable Latest",
@@ -12279,13 +12543,13 @@ export const MODELS = {
12279
12543
  reasoning: true,
12280
12544
  input: ["text", "image"],
12281
12545
  cost: {
12282
- input: 0.6799999999999999,
12283
- output: 3.41,
12284
- cacheRead: 0.33999999999999997,
12546
+ input: 0.67,
12547
+ output: 3.5,
12548
+ cacheRead: 0.19999999999999998,
12285
12549
  cacheWrite: 0,
12286
12550
  },
12287
12551
  contextWindow: 262144,
12288
- maxTokens: 262142,
12552
+ maxTokens: 262144,
12289
12553
  },
12290
12554
  "~openai/gpt-latest": {
12291
12555
  id: "~openai/gpt-latest",
@@ -12323,9 +12587,9 @@ export const MODELS = {
12323
12587
  },
12324
12588
  },
12325
12589
  "together": {
12326
- "MiniMaxAI/MiniMax-M2.5": {
12327
- id: "MiniMaxAI/MiniMax-M2.5",
12328
- name: "MiniMax-M2.5",
12590
+ "MiniMaxAI/MiniMax-M2.7": {
12591
+ id: "MiniMaxAI/MiniMax-M2.7",
12592
+ name: "MiniMax-M2.7",
12329
12593
  api: "openai-completions",
12330
12594
  provider: "together",
12331
12595
  baseUrl: "https://api.together.ai/v1",
@@ -12339,50 +12603,49 @@ export const MODELS = {
12339
12603
  cacheRead: 0.06,
12340
12604
  cacheWrite: 0,
12341
12605
  },
12342
- contextWindow: 204800,
12606
+ contextWindow: 202752,
12343
12607
  maxTokens: 131072,
12344
12608
  },
12345
- "MiniMaxAI/MiniMax-M2.7": {
12346
- id: "MiniMaxAI/MiniMax-M2.7",
12347
- name: "MiniMax-M2.7",
12609
+ "MiniMaxAI/MiniMax-M3": {
12610
+ id: "MiniMaxAI/MiniMax-M3",
12611
+ name: "MiniMax-M3",
12348
12612
  api: "openai-completions",
12349
12613
  provider: "together",
12350
12614
  baseUrl: "https://api.together.ai/v1",
12351
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12615
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12352
12616
  reasoning: true,
12353
- thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null },
12354
- input: ["text"],
12617
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12618
+ input: ["text", "image"],
12355
12619
  cost: {
12356
12620
  input: 0.3,
12357
12621
  output: 1.2,
12358
12622
  cacheRead: 0.06,
12359
12623
  cacheWrite: 0,
12360
12624
  },
12361
- contextWindow: 202752,
12362
- maxTokens: 131072,
12625
+ contextWindow: 524288,
12626
+ maxTokens: 250000,
12363
12627
  },
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",
12628
+ "Qwen/Qwen2.5-7B-Instruct-Turbo": {
12629
+ id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
12630
+ name: "Qwen 2.5 7B Instruct Turbo",
12367
12631
  api: "openai-completions",
12368
12632
  provider: "together",
12369
12633
  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 },
12634
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12635
+ reasoning: false,
12373
12636
  input: ["text"],
12374
12637
  cost: {
12375
- input: 0.2,
12376
- output: 0.6,
12638
+ input: 0.3,
12639
+ output: 0.3,
12377
12640
  cacheRead: 0,
12378
12641
  cacheWrite: 0,
12379
12642
  },
12380
- contextWindow: 262144,
12381
- maxTokens: 262144,
12643
+ contextWindow: 32768,
12644
+ maxTokens: 32768,
12382
12645
  },
12383
- "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {
12384
- id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8",
12385
- name: "Qwen3 Coder 480B A35B Instruct",
12646
+ "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12647
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12648
+ name: "Qwen3 235B A22B Instruct 2507 FP8",
12386
12649
  api: "openai-completions",
12387
12650
  provider: "together",
12388
12651
  baseUrl: "https://api.together.ai/v1",
@@ -12390,36 +12653,36 @@ export const MODELS = {
12390
12653
  reasoning: false,
12391
12654
  input: ["text"],
12392
12655
  cost: {
12393
- input: 2,
12394
- output: 2,
12656
+ input: 0.2,
12657
+ output: 0.6,
12395
12658
  cacheRead: 0,
12396
12659
  cacheWrite: 0,
12397
12660
  },
12398
12661
  contextWindow: 262144,
12399
12662
  maxTokens: 262144,
12400
12663
  },
12401
- "Qwen/Qwen3-Coder-Next-FP8": {
12402
- id: "Qwen/Qwen3-Coder-Next-FP8",
12403
- name: "Qwen3 Coder Next FP8",
12664
+ "Qwen/Qwen3.5-397B-A17B": {
12665
+ id: "Qwen/Qwen3.5-397B-A17B",
12666
+ name: "Qwen3.5 397B A17B",
12404
12667
  api: "openai-completions",
12405
12668
  provider: "together",
12406
12669
  baseUrl: "https://api.together.ai/v1",
12407
12670
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12408
12671
  reasoning: true,
12409
12672
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12410
- input: ["text"],
12673
+ input: ["text", "image"],
12411
12674
  cost: {
12412
- input: 0.5,
12413
- output: 1.2,
12675
+ input: 0.6,
12676
+ output: 3.6,
12414
12677
  cacheRead: 0,
12415
12678
  cacheWrite: 0,
12416
12679
  },
12417
12680
  contextWindow: 262144,
12418
- maxTokens: 262144,
12681
+ maxTokens: 130000,
12419
12682
  },
12420
- "Qwen/Qwen3.5-397B-A17B": {
12421
- id: "Qwen/Qwen3.5-397B-A17B",
12422
- name: "Qwen3.5 397B A17B",
12683
+ "Qwen/Qwen3.5-9B": {
12684
+ id: "Qwen/Qwen3.5-9B",
12685
+ name: "Qwen3.5 9B",
12423
12686
  api: "openai-completions",
12424
12687
  provider: "together",
12425
12688
  baseUrl: "https://api.together.ai/v1",
@@ -12428,13 +12691,13 @@ export const MODELS = {
12428
12691
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12429
12692
  input: ["text", "image"],
12430
12693
  cost: {
12431
- input: 0.6,
12432
- output: 3.6,
12694
+ input: 0.17,
12695
+ output: 0.25,
12433
12696
  cacheRead: 0,
12434
12697
  cacheWrite: 0,
12435
12698
  },
12436
12699
  contextWindow: 262144,
12437
- maxTokens: 130000,
12700
+ maxTokens: 65536,
12438
12701
  },
12439
12702
  "Qwen/Qwen3.6-Plus": {
12440
12703
  id: "Qwen/Qwen3.6-Plus",
@@ -12461,57 +12724,18 @@ export const MODELS = {
12461
12724
  api: "openai-completions",
12462
12725
  provider: "together",
12463
12726
  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 },
12727
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12728
+ reasoning: false,
12467
12729
  input: ["text"],
12468
12730
  cost: {
12469
- input: 2.5,
12470
- output: 7.5,
12731
+ input: 1.25,
12732
+ output: 3.75,
12471
12733
  cacheRead: 0,
12472
12734
  cacheWrite: 0,
12473
12735
  },
12474
12736
  contextWindow: 1000000,
12475
12737
  maxTokens: 500000,
12476
12738
  },
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
12739
  "deepseek-ai/DeepSeek-V4-Pro": {
12516
12740
  id: "deepseek-ai/DeepSeek-V4-Pro",
12517
12741
  name: "DeepSeek V4 Pro",
@@ -12523,8 +12747,8 @@ export const MODELS = {
12523
12747
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": null },
12524
12748
  input: ["text"],
12525
12749
  cost: {
12526
- input: 2.1,
12527
- output: 4.4,
12750
+ input: 1.74,
12751
+ output: 3.48,
12528
12752
  cacheRead: 0.2,
12529
12753
  cacheWrite: 0,
12530
12754
  },
@@ -12560,8 +12784,8 @@ export const MODELS = {
12560
12784
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12561
12785
  input: ["text", "image"],
12562
12786
  cost: {
12563
- input: 0.2,
12564
- output: 0.5,
12787
+ input: 0.39,
12788
+ output: 0.97,
12565
12789
  cacheRead: 0,
12566
12790
  cacheWrite: 0,
12567
12791
  },
@@ -12586,9 +12810,9 @@ export const MODELS = {
12586
12810
  contextWindow: 131072,
12587
12811
  maxTokens: 131072,
12588
12812
  },
12589
- "moonshotai/Kimi-K2.5": {
12590
- id: "moonshotai/Kimi-K2.5",
12591
- name: "Kimi K2.5",
12813
+ "moonshotai/Kimi-K2.6": {
12814
+ id: "moonshotai/Kimi-K2.6",
12815
+ name: "Kimi K2.6",
12592
12816
  api: "openai-completions",
12593
12817
  provider: "together",
12594
12818
  baseUrl: "https://api.together.ai/v1",
@@ -12597,32 +12821,32 @@ export const MODELS = {
12597
12821
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12598
12822
  input: ["text", "image"],
12599
12823
  cost: {
12600
- input: 0.5,
12601
- output: 2.8,
12602
- cacheRead: 0,
12824
+ input: 1.2,
12825
+ output: 4.5,
12826
+ cacheRead: 0.2,
12603
12827
  cacheWrite: 0,
12604
12828
  },
12605
12829
  contextWindow: 262144,
12606
- maxTokens: 262144,
12830
+ maxTokens: 131000,
12607
12831
  },
12608
- "moonshotai/Kimi-K2.6": {
12609
- id: "moonshotai/Kimi-K2.6",
12610
- name: "Kimi K2.6",
12832
+ "moonshotai/Kimi-K2.7-Code": {
12833
+ id: "moonshotai/Kimi-K2.7-Code",
12834
+ name: "Kimi K2.7 Code",
12611
12835
  api: "openai-completions",
12612
12836
  provider: "together",
12613
12837
  baseUrl: "https://api.together.ai/v1",
12614
12838
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12615
12839
  reasoning: true,
12616
12840
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12617
- input: ["text", "image"],
12841
+ input: ["text"],
12618
12842
  cost: {
12619
- input: 1.2,
12620
- output: 4.5,
12621
- cacheRead: 0.2,
12843
+ input: 0.95,
12844
+ output: 4,
12845
+ cacheRead: 0.19,
12622
12846
  cacheWrite: 0,
12623
12847
  },
12624
12848
  contextWindow: 262144,
12625
- maxTokens: 131000,
12849
+ maxTokens: 131072,
12626
12850
  },
12627
12851
  "nvidia/nemotron-3-ultra-550b-a55b": {
12628
12852
  id: "nvidia/nemotron-3-ultra-550b-a55b",
@@ -12662,6 +12886,44 @@ export const MODELS = {
12662
12886
  contextWindow: 131072,
12663
12887
  maxTokens: 131072,
12664
12888
  },
12889
+ "openai/gpt-oss-20b": {
12890
+ id: "openai/gpt-oss-20b",
12891
+ name: "GPT OSS 20B",
12892
+ api: "openai-completions",
12893
+ provider: "together",
12894
+ baseUrl: "https://api.together.ai/v1",
12895
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": true, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "openai" },
12896
+ reasoning: true,
12897
+ thinkingLevelMap: { "off": null, "minimal": null },
12898
+ input: ["text"],
12899
+ cost: {
12900
+ input: 0.05,
12901
+ output: 0.2,
12902
+ cacheRead: 0,
12903
+ cacheWrite: 0,
12904
+ },
12905
+ contextWindow: 131072,
12906
+ maxTokens: 131072,
12907
+ },
12908
+ "zai-org/GLM-5": {
12909
+ id: "zai-org/GLM-5",
12910
+ name: "GLM-5",
12911
+ api: "openai-completions",
12912
+ provider: "together",
12913
+ baseUrl: "https://api.together.ai/v1",
12914
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12915
+ reasoning: true,
12916
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12917
+ input: ["text"],
12918
+ cost: {
12919
+ input: 1,
12920
+ output: 3.2,
12921
+ cacheRead: 0,
12922
+ cacheWrite: 0,
12923
+ },
12924
+ contextWindow: 202752,
12925
+ maxTokens: 131072,
12926
+ },
12665
12927
  "zai-org/GLM-5.1": {
12666
12928
  id: "zai-org/GLM-5.1",
12667
12929
  name: "GLM-5.1",
@@ -13074,24 +13336,6 @@ export const MODELS = {
13074
13336
  contextWindow: 200000,
13075
13337
  maxTokens: 4096,
13076
13338
  },
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
13339
  "anthropic/claude-haiku-4.5": {
13096
13340
  id: "anthropic/claude-haiku-4.5",
13097
13341
  name: "Claude Haiku 4.5",
@@ -14124,13 +14368,13 @@ export const MODELS = {
14124
14368
  reasoning: false,
14125
14369
  input: ["text"],
14126
14370
  cost: {
14127
- input: 0.02,
14128
- output: 0.04,
14371
+ input: 0.15,
14372
+ output: 0.15,
14129
14373
  cacheRead: 0,
14130
14374
  cacheWrite: 0,
14131
14375
  },
14132
- contextWindow: 131072,
14133
- maxTokens: 131072,
14376
+ contextWindow: 128000,
14377
+ maxTokens: 128000,
14134
14378
  },
14135
14379
  "mistral/mistral-small": {
14136
14380
  id: "mistral/mistral-small",
@@ -14217,73 +14461,73 @@ export const MODELS = {
14217
14461
  contextWindow: 262114,
14218
14462
  maxTokens: 262114,
14219
14463
  },
14220
- "moonshotai/kimi-k2-thinking-turbo": {
14221
- id: "moonshotai/kimi-k2-thinking-turbo",
14222
- name: "Kimi K2 Thinking Turbo",
14464
+ "moonshotai/kimi-k2.5": {
14465
+ id: "moonshotai/kimi-k2.5",
14466
+ name: "Kimi K2.5",
14223
14467
  api: "anthropic-messages",
14224
14468
  provider: "vercel-ai-gateway",
14225
14469
  baseUrl: "https://ai-gateway.vercel.sh",
14226
14470
  reasoning: true,
14227
- input: ["text"],
14471
+ input: ["text", "image"],
14228
14472
  cost: {
14229
- input: 1.15,
14230
- output: 8,
14231
- cacheRead: 0.15,
14473
+ input: 0.6,
14474
+ output: 3,
14475
+ cacheRead: 0.09999999999999999,
14232
14476
  cacheWrite: 0,
14233
14477
  },
14234
14478
  contextWindow: 262114,
14235
14479
  maxTokens: 262114,
14236
14480
  },
14237
- "moonshotai/kimi-k2-turbo": {
14238
- id: "moonshotai/kimi-k2-turbo",
14239
- name: "Kimi K2 Turbo",
14481
+ "moonshotai/kimi-k2.6": {
14482
+ id: "moonshotai/kimi-k2.6",
14483
+ name: "Kimi K2.6",
14240
14484
  api: "anthropic-messages",
14241
14485
  provider: "vercel-ai-gateway",
14242
14486
  baseUrl: "https://ai-gateway.vercel.sh",
14243
- reasoning: false,
14244
- input: ["text"],
14487
+ reasoning: true,
14488
+ input: ["text", "image"],
14245
14489
  cost: {
14246
- input: 1.15,
14247
- output: 8,
14248
- cacheRead: 0.15,
14490
+ input: 0.95,
14491
+ output: 4,
14492
+ cacheRead: 0.16,
14249
14493
  cacheWrite: 0,
14250
14494
  },
14251
- contextWindow: 256000,
14252
- maxTokens: 16384,
14495
+ contextWindow: 262000,
14496
+ maxTokens: 262000,
14253
14497
  },
14254
- "moonshotai/kimi-k2.5": {
14255
- id: "moonshotai/kimi-k2.5",
14256
- name: "Kimi K2.5",
14498
+ "moonshotai/kimi-k2.7-code": {
14499
+ id: "moonshotai/kimi-k2.7-code",
14500
+ name: "Kimi K2.7 Code",
14257
14501
  api: "anthropic-messages",
14258
14502
  provider: "vercel-ai-gateway",
14259
14503
  baseUrl: "https://ai-gateway.vercel.sh",
14260
14504
  reasoning: true,
14261
14505
  input: ["text", "image"],
14262
14506
  cost: {
14263
- input: 0.6,
14264
- output: 3,
14265
- cacheRead: 0.09999999999999999,
14507
+ input: 0.95,
14508
+ output: 4,
14509
+ cacheRead: 0.19,
14266
14510
  cacheWrite: 0,
14267
14511
  },
14268
- contextWindow: 262114,
14269
- maxTokens: 262114,
14512
+ contextWindow: 256000,
14513
+ maxTokens: 32768,
14270
14514
  },
14271
- "moonshotai/kimi-k2.6": {
14272
- id: "moonshotai/kimi-k2.6",
14273
- name: "Kimi K2.6",
14515
+ "moonshotai/kimi-k2.7-code-highspeed": {
14516
+ id: "moonshotai/kimi-k2.7-code-highspeed",
14517
+ name: "Kimi K2.7 Code High Speed",
14274
14518
  api: "anthropic-messages",
14275
14519
  provider: "vercel-ai-gateway",
14276
14520
  baseUrl: "https://ai-gateway.vercel.sh",
14277
14521
  reasoning: true,
14278
14522
  input: ["text", "image"],
14279
14523
  cost: {
14280
- input: 0.95,
14281
- output: 4,
14282
- cacheRead: 0.16,
14524
+ input: 1.9,
14525
+ output: 8,
14526
+ cacheRead: 0.38,
14283
14527
  cacheWrite: 0,
14284
14528
  },
14285
- contextWindow: 262000,
14286
- maxTokens: 262000,
14529
+ contextWindow: 262144,
14530
+ maxTokens: 32768,
14287
14531
  },
14288
14532
  "nvidia/nemotron-3-super-120b-a12b": {
14289
14533
  id: "nvidia/nemotron-3-super-120b-a12b",
@@ -15470,6 +15714,23 @@ export const MODELS = {
15470
15714
  contextWindow: 202800,
15471
15715
  maxTokens: 64000,
15472
15716
  },
15717
+ "zai/glm-5.2": {
15718
+ id: "zai/glm-5.2",
15719
+ name: "GLM 5.2",
15720
+ api: "anthropic-messages",
15721
+ provider: "vercel-ai-gateway",
15722
+ baseUrl: "https://ai-gateway.vercel.sh",
15723
+ reasoning: true,
15724
+ input: ["text"],
15725
+ cost: {
15726
+ input: 1.5,
15727
+ output: 4.5,
15728
+ cacheRead: 0.3,
15729
+ cacheWrite: 0,
15730
+ },
15731
+ contextWindow: 1000000,
15732
+ maxTokens: 128000,
15733
+ },
15473
15734
  "zai/glm-5v-turbo": {
15474
15735
  id: "zai/glm-5v-turbo",
15475
15736
  name: "GLM 5V Turbo",
@@ -15860,6 +16121,24 @@ export const MODELS = {
15860
16121
  contextWindow: 200000,
15861
16122
  maxTokens: 131072,
15862
16123
  },
16124
+ "glm-5.2": {
16125
+ id: "glm-5.2",
16126
+ name: "GLM-5.2",
16127
+ api: "openai-completions",
16128
+ provider: "zai",
16129
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
16130
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
16131
+ reasoning: true,
16132
+ input: ["text"],
16133
+ cost: {
16134
+ input: 0,
16135
+ output: 0,
16136
+ cacheRead: 0,
16137
+ cacheWrite: 0,
16138
+ },
16139
+ contextWindow: 1000000,
16140
+ maxTokens: 131072,
16141
+ },
15863
16142
  "glm-5v-turbo": {
15864
16143
  id: "glm-5v-turbo",
15865
16144
  name: "GLM-5V-Turbo",