@elyracode/ai 0.9.0 → 0.9.2

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.
@@ -391,8 +391,8 @@ export const MODELS = {
391
391
  thinkingLevelMap: { "xhigh": "max" },
392
392
  input: ["text", "image"],
393
393
  cost: {
394
- input: 5,
395
- output: 25,
394
+ input: 5.5,
395
+ output: 27.5,
396
396
  cacheRead: 0.5,
397
397
  cacheWrite: 6.25,
398
398
  },
@@ -409,10 +409,10 @@ export const MODELS = {
409
409
  thinkingLevelMap: { "xhigh": "xhigh" },
410
410
  input: ["text", "image"],
411
411
  cost: {
412
- input: 5,
413
- output: 25,
414
- cacheRead: 0.5,
415
- cacheWrite: 6.25,
412
+ input: 5.5,
413
+ output: 27.5,
414
+ cacheRead: 0.55,
415
+ cacheWrite: 6.875,
416
416
  },
417
417
  contextWindow: 1000000,
418
418
  maxTokens: 128000,
@@ -427,10 +427,10 @@ export const MODELS = {
427
427
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
428
428
  input: ["text", "image"],
429
429
  cost: {
430
- input: 5,
431
- output: 25,
432
- cacheRead: 0.5,
433
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
434
434
  },
435
435
  contextWindow: 1000000,
436
436
  maxTokens: 128000,
@@ -444,10 +444,10 @@ export const MODELS = {
444
444
  reasoning: true,
445
445
  input: ["text", "image"],
446
446
  cost: {
447
- input: 3,
448
- output: 15,
449
- cacheRead: 0.3,
450
- cacheWrite: 3.75,
447
+ input: 3.3,
448
+ output: 16.5,
449
+ cacheRead: 0.33,
450
+ cacheWrite: 4.125,
451
451
  },
452
452
  contextWindow: 200000,
453
453
  maxTokens: 64000,
@@ -461,10 +461,10 @@ export const MODELS = {
461
461
  reasoning: true,
462
462
  input: ["text", "image"],
463
463
  cost: {
464
- input: 3,
465
- output: 15,
466
- cacheRead: 0.3,
467
- cacheWrite: 3.75,
464
+ input: 3.3,
465
+ output: 16.5,
466
+ cacheRead: 0.33,
467
+ cacheWrite: 4.125,
468
468
  },
469
469
  contextWindow: 1000000,
470
470
  maxTokens: 64000,
@@ -1086,6 +1086,59 @@ export const MODELS = {
1086
1086
  contextWindow: 262144,
1087
1087
  maxTokens: 131072,
1088
1088
  },
1089
+ "openai.gpt-5.4": {
1090
+ id: "openai.gpt-5.4",
1091
+ name: "GPT-5.4",
1092
+ api: "bedrock-converse-stream",
1093
+ provider: "amazon-bedrock",
1094
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1095
+ reasoning: true,
1096
+ thinkingLevelMap: { "xhigh": "xhigh" },
1097
+ input: ["text", "image"],
1098
+ cost: {
1099
+ input: 2.75,
1100
+ output: 16.5,
1101
+ cacheRead: 0.275,
1102
+ cacheWrite: 0,
1103
+ },
1104
+ contextWindow: 272000,
1105
+ maxTokens: 128000,
1106
+ },
1107
+ "openai.gpt-5.5": {
1108
+ id: "openai.gpt-5.5",
1109
+ name: "GPT-5.5",
1110
+ api: "bedrock-converse-stream",
1111
+ provider: "amazon-bedrock",
1112
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1113
+ reasoning: true,
1114
+ thinkingLevelMap: { "xhigh": "xhigh" },
1115
+ input: ["text", "image"],
1116
+ cost: {
1117
+ input: 5.5,
1118
+ output: 33,
1119
+ cacheRead: 0.55,
1120
+ cacheWrite: 0,
1121
+ },
1122
+ contextWindow: 272000,
1123
+ maxTokens: 128000,
1124
+ },
1125
+ "openai.gpt-oss-120b": {
1126
+ id: "openai.gpt-oss-120b",
1127
+ name: "gpt-oss-120b",
1128
+ api: "bedrock-converse-stream",
1129
+ provider: "amazon-bedrock",
1130
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1131
+ reasoning: false,
1132
+ input: ["text"],
1133
+ cost: {
1134
+ input: 0.15,
1135
+ output: 0.6,
1136
+ cacheRead: 0,
1137
+ cacheWrite: 0,
1138
+ },
1139
+ contextWindow: 128000,
1140
+ maxTokens: 16384,
1141
+ },
1089
1142
  "openai.gpt-oss-120b-1:0": {
1090
1143
  id: "openai.gpt-oss-120b-1:0",
1091
1144
  name: "gpt-oss-120b",
@@ -1103,6 +1156,23 @@ export const MODELS = {
1103
1156
  contextWindow: 128000,
1104
1157
  maxTokens: 16384,
1105
1158
  },
1159
+ "openai.gpt-oss-20b": {
1160
+ id: "openai.gpt-oss-20b",
1161
+ name: "gpt-oss-20b",
1162
+ api: "bedrock-converse-stream",
1163
+ provider: "amazon-bedrock",
1164
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1165
+ reasoning: false,
1166
+ input: ["text"],
1167
+ cost: {
1168
+ input: 0.07,
1169
+ output: 0.3,
1170
+ cacheRead: 0,
1171
+ cacheWrite: 0,
1172
+ },
1173
+ contextWindow: 128000,
1174
+ maxTokens: 16384,
1175
+ },
1106
1176
  "openai.gpt-oss-20b-1:0": {
1107
1177
  id: "openai.gpt-oss-20b-1:0",
1108
1178
  name: "gpt-oss-20b",
@@ -3321,24 +3391,6 @@ export const MODELS = {
3321
3391
  contextWindow: 128000,
3322
3392
  maxTokens: 128000,
3323
3393
  },
3324
- "@cf/moonshotai/kimi-k2.5": {
3325
- id: "@cf/moonshotai/kimi-k2.5",
3326
- name: "Kimi K2.5",
3327
- api: "openai-completions",
3328
- provider: "cloudflare-workers-ai",
3329
- baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3330
- compat: { "sendSessionAffinityHeaders": true },
3331
- reasoning: true,
3332
- input: ["text", "image"],
3333
- cost: {
3334
- input: 0.6,
3335
- output: 3,
3336
- cacheRead: 0.1,
3337
- cacheWrite: 0,
3338
- },
3339
- contextWindow: 256000,
3340
- maxTokens: 256000,
3341
- },
3342
3394
  "@cf/moonshotai/kimi-k2.6": {
3343
3395
  id: "@cf/moonshotai/kimi-k2.6",
3344
3396
  name: "Kimi K2.6",
@@ -3656,8 +3708,8 @@ export const MODELS = {
3656
3708
  cacheRead: 0.1,
3657
3709
  cacheWrite: 0,
3658
3710
  },
3659
- contextWindow: 128000,
3660
- maxTokens: 8192,
3711
+ contextWindow: 262144,
3712
+ maxTokens: 65536,
3661
3713
  },
3662
3714
  "accounts/fireworks/routers/glm-5p1-fast": {
3663
3715
  id: "accounts/fireworks/routers/glm-5p1-fast",
@@ -3676,6 +3728,23 @@ export const MODELS = {
3676
3728
  contextWindow: 202800,
3677
3729
  maxTokens: 131072,
3678
3730
  },
3731
+ "accounts/fireworks/routers/kimi-k2p6-fast": {
3732
+ id: "accounts/fireworks/routers/kimi-k2p6-fast",
3733
+ name: "Kimi K2.6 Fast",
3734
+ api: "anthropic-messages",
3735
+ provider: "fireworks",
3736
+ baseUrl: "https://api.fireworks.ai/inference",
3737
+ reasoning: true,
3738
+ input: ["text", "image"],
3739
+ cost: {
3740
+ input: 2,
3741
+ output: 8,
3742
+ cacheRead: 0.3,
3743
+ cacheWrite: 0,
3744
+ },
3745
+ contextWindow: 262000,
3746
+ maxTokens: 262000,
3747
+ },
3679
3748
  "accounts/fireworks/routers/kimi-k2p6-turbo": {
3680
3749
  id: "accounts/fireworks/routers/kimi-k2p6-turbo",
3681
3750
  name: "Kimi K2.6 Turbo",
@@ -3697,7 +3766,7 @@ export const MODELS = {
3697
3766
  "github-copilot": {
3698
3767
  "claude-haiku-4.5": {
3699
3768
  id: "claude-haiku-4.5",
3700
- name: "Claude Haiku 4.5",
3769
+ name: "Claude Haiku 4.5 (latest)",
3701
3770
  api: "anthropic-messages",
3702
3771
  provider: "github-copilot",
3703
3772
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3706,17 +3775,17 @@ export const MODELS = {
3706
3775
  reasoning: true,
3707
3776
  input: ["text", "image"],
3708
3777
  cost: {
3709
- input: 0,
3710
- output: 0,
3711
- cacheRead: 0,
3712
- cacheWrite: 0,
3778
+ input: 1,
3779
+ output: 5,
3780
+ cacheRead: 0.1,
3781
+ cacheWrite: 1.25,
3713
3782
  },
3714
3783
  contextWindow: 200000,
3715
3784
  maxTokens: 64000,
3716
3785
  },
3717
3786
  "claude-opus-4.5": {
3718
3787
  id: "claude-opus-4.5",
3719
- name: "Claude Opus 4.5",
3788
+ name: "Claude Opus 4.5 (latest)",
3720
3789
  api: "anthropic-messages",
3721
3790
  provider: "github-copilot",
3722
3791
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3724,10 +3793,10 @@ export const MODELS = {
3724
3793
  reasoning: true,
3725
3794
  input: ["text", "image"],
3726
3795
  cost: {
3727
- input: 0,
3728
- output: 0,
3729
- cacheRead: 0,
3730
- cacheWrite: 0,
3796
+ input: 5,
3797
+ output: 25,
3798
+ cacheRead: 0.5,
3799
+ cacheWrite: 6.25,
3731
3800
  },
3732
3801
  contextWindow: 200000,
3733
3802
  maxTokens: 32000,
@@ -3743,10 +3812,10 @@ export const MODELS = {
3743
3812
  thinkingLevelMap: { "xhigh": "max" },
3744
3813
  input: ["text", "image"],
3745
3814
  cost: {
3746
- input: 0,
3747
- output: 0,
3748
- cacheRead: 0,
3749
- cacheWrite: 0,
3815
+ input: 5,
3816
+ output: 25,
3817
+ cacheRead: 0.5,
3818
+ cacheWrite: 6.25,
3750
3819
  },
3751
3820
  contextWindow: 1000000,
3752
3821
  maxTokens: 32000,
@@ -3762,10 +3831,10 @@ export const MODELS = {
3762
3831
  thinkingLevelMap: { "xhigh": "xhigh" },
3763
3832
  input: ["text", "image"],
3764
3833
  cost: {
3765
- input: 0,
3766
- output: 0,
3767
- cacheRead: 0,
3768
- cacheWrite: 0,
3834
+ input: 5,
3835
+ output: 25,
3836
+ cacheRead: 0.5,
3837
+ cacheWrite: 6.25,
3769
3838
  },
3770
3839
  contextWindow: 200000,
3771
3840
  maxTokens: 32000,
@@ -3781,17 +3850,36 @@ export const MODELS = {
3781
3850
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
3782
3851
  input: ["text", "image"],
3783
3852
  cost: {
3784
- input: 0,
3785
- output: 0,
3786
- cacheRead: 0,
3787
- cacheWrite: 0,
3853
+ input: 5,
3854
+ output: 25,
3855
+ cacheRead: 0.5,
3856
+ cacheWrite: 6.25,
3788
3857
  },
3789
3858
  contextWindow: 200000,
3790
3859
  maxTokens: 64000,
3791
3860
  },
3861
+ "claude-sonnet-4": {
3862
+ id: "claude-sonnet-4",
3863
+ name: "Claude Sonnet 4 (latest)",
3864
+ api: "anthropic-messages",
3865
+ provider: "github-copilot",
3866
+ baseUrl: "https://api.individual.githubcopilot.com",
3867
+ 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" },
3868
+ compat: { "supportsEagerToolInputStreaming": false },
3869
+ reasoning: true,
3870
+ input: ["text", "image"],
3871
+ cost: {
3872
+ input: 3,
3873
+ output: 15,
3874
+ cacheRead: 0.3,
3875
+ cacheWrite: 3.75,
3876
+ },
3877
+ contextWindow: 216000,
3878
+ maxTokens: 16000,
3879
+ },
3792
3880
  "claude-sonnet-4.5": {
3793
3881
  id: "claude-sonnet-4.5",
3794
- name: "Claude Sonnet 4.5",
3882
+ name: "Claude Sonnet 4.5 (latest)",
3795
3883
  api: "anthropic-messages",
3796
3884
  provider: "github-copilot",
3797
3885
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3800,10 +3888,10 @@ export const MODELS = {
3800
3888
  reasoning: true,
3801
3889
  input: ["text", "image"],
3802
3890
  cost: {
3803
- input: 0,
3804
- output: 0,
3805
- cacheRead: 0,
3806
- cacheWrite: 0,
3891
+ input: 3,
3892
+ output: 15,
3893
+ cacheRead: 0.3,
3894
+ cacheWrite: 3.75,
3807
3895
  },
3808
3896
  contextWindow: 200000,
3809
3897
  maxTokens: 32000,
@@ -3818,10 +3906,10 @@ export const MODELS = {
3818
3906
  reasoning: true,
3819
3907
  input: ["text", "image"],
3820
3908
  cost: {
3821
- input: 0,
3822
- output: 0,
3823
- cacheRead: 0,
3824
- cacheWrite: 0,
3909
+ input: 3,
3910
+ output: 15,
3911
+ cacheRead: 0.3,
3912
+ cacheWrite: 3.75,
3825
3913
  },
3826
3914
  contextWindow: 1000000,
3827
3915
  maxTokens: 32000,
@@ -3834,12 +3922,12 @@ export const MODELS = {
3834
3922
  baseUrl: "https://api.individual.githubcopilot.com",
3835
3923
  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" },
3836
3924
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3837
- reasoning: false,
3925
+ reasoning: true,
3838
3926
  input: ["text", "image"],
3839
3927
  cost: {
3840
- input: 0,
3841
- output: 0,
3842
- cacheRead: 0,
3928
+ input: 1.25,
3929
+ output: 10,
3930
+ cacheRead: 0.125,
3843
3931
  cacheWrite: 0,
3844
3932
  },
3845
3933
  contextWindow: 128000,
@@ -3847,7 +3935,7 @@ export const MODELS = {
3847
3935
  },
3848
3936
  "gemini-3-flash-preview": {
3849
3937
  id: "gemini-3-flash-preview",
3850
- name: "Gemini 3 Flash",
3938
+ name: "Gemini 3 Flash Preview",
3851
3939
  api: "openai-completions",
3852
3940
  provider: "github-copilot",
3853
3941
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3856,9 +3944,9 @@ export const MODELS = {
3856
3944
  reasoning: true,
3857
3945
  input: ["text", "image"],
3858
3946
  cost: {
3859
- input: 0,
3860
- output: 0,
3861
- cacheRead: 0,
3947
+ input: 0.5,
3948
+ output: 3,
3949
+ cacheRead: 0.05,
3862
3950
  cacheWrite: 0,
3863
3951
  },
3864
3952
  contextWindow: 128000,
@@ -3875,9 +3963,9 @@ export const MODELS = {
3875
3963
  reasoning: true,
3876
3964
  input: ["text", "image"],
3877
3965
  cost: {
3878
- input: 0,
3879
- output: 0,
3880
- cacheRead: 0,
3966
+ input: 2,
3967
+ output: 12,
3968
+ cacheRead: 0.2,
3881
3969
  cacheWrite: 0,
3882
3970
  },
3883
3971
  contextWindow: 200000,
@@ -3894,9 +3982,9 @@ export const MODELS = {
3894
3982
  reasoning: true,
3895
3983
  input: ["text", "image"],
3896
3984
  cost: {
3897
- input: 0,
3898
- output: 0,
3899
- cacheRead: 0,
3985
+ input: 1.5,
3986
+ output: 9,
3987
+ cacheRead: 0.15,
3900
3988
  cacheWrite: 0,
3901
3989
  },
3902
3990
  contextWindow: 200000,
@@ -3913,36 +4001,17 @@ export const MODELS = {
3913
4001
  reasoning: false,
3914
4002
  input: ["text", "image"],
3915
4003
  cost: {
3916
- input: 0,
3917
- output: 0,
3918
- cacheRead: 0,
4004
+ input: 2,
4005
+ output: 8,
4006
+ cacheRead: 0.5,
3919
4007
  cacheWrite: 0,
3920
4008
  },
3921
4009
  contextWindow: 128000,
3922
4010
  maxTokens: 16384,
3923
4011
  },
3924
- "gpt-4o": {
3925
- id: "gpt-4o",
3926
- name: "GPT-4o",
3927
- api: "openai-completions",
3928
- provider: "github-copilot",
3929
- baseUrl: "https://api.individual.githubcopilot.com",
3930
- 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" },
3931
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
3932
- reasoning: false,
3933
- input: ["text", "image"],
3934
- cost: {
3935
- input: 0,
3936
- output: 0,
3937
- cacheRead: 0,
3938
- cacheWrite: 0,
3939
- },
3940
- contextWindow: 128000,
3941
- maxTokens: 4096,
3942
- },
3943
4012
  "gpt-5-mini": {
3944
4013
  id: "gpt-5-mini",
3945
- name: "GPT-5-mini",
4014
+ name: "GPT-5 Mini",
3946
4015
  api: "openai-responses",
3947
4016
  provider: "github-copilot",
3948
4017
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3951,9 +4020,9 @@ export const MODELS = {
3951
4020
  thinkingLevelMap: { "off": null },
3952
4021
  input: ["text", "image"],
3953
4022
  cost: {
3954
- input: 0,
3955
- output: 0,
3956
- cacheRead: 0,
4023
+ input: 0.25,
4024
+ output: 2,
4025
+ cacheRead: 0.025,
3957
4026
  cacheWrite: 0,
3958
4027
  },
3959
4028
  contextWindow: 264000,
@@ -3970,9 +4039,9 @@ export const MODELS = {
3970
4039
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3971
4040
  input: ["text", "image"],
3972
4041
  cost: {
3973
- input: 0,
3974
- output: 0,
3975
- cacheRead: 0,
4042
+ input: 1.75,
4043
+ output: 14,
4044
+ cacheRead: 0.175,
3976
4045
  cacheWrite: 0,
3977
4046
  },
3978
4047
  contextWindow: 400000,
@@ -3980,7 +4049,7 @@ export const MODELS = {
3980
4049
  },
3981
4050
  "gpt-5.2-codex": {
3982
4051
  id: "gpt-5.2-codex",
3983
- name: "GPT-5.2-Codex",
4052
+ name: "GPT-5.2 Codex",
3984
4053
  api: "openai-responses",
3985
4054
  provider: "github-copilot",
3986
4055
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -3989,17 +4058,36 @@ export const MODELS = {
3989
4058
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3990
4059
  input: ["text", "image"],
3991
4060
  cost: {
3992
- input: 0,
3993
- output: 0,
3994
- cacheRead: 0,
4061
+ input: 1.75,
4062
+ output: 14,
4063
+ cacheRead: 0.175,
4064
+ cacheWrite: 0,
4065
+ },
4066
+ contextWindow: 400000,
4067
+ maxTokens: 128000,
4068
+ },
4069
+ "gpt-5.3-codex": {
4070
+ id: "gpt-5.3-codex",
4071
+ name: "GPT-5.3 Codex",
4072
+ api: "openai-responses",
4073
+ provider: "github-copilot",
4074
+ baseUrl: "https://api.individual.githubcopilot.com",
4075
+ 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" },
4076
+ reasoning: true,
4077
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4078
+ input: ["text", "image"],
4079
+ cost: {
4080
+ input: 1.75,
4081
+ output: 14,
4082
+ cacheRead: 0.175,
3995
4083
  cacheWrite: 0,
3996
4084
  },
3997
4085
  contextWindow: 400000,
3998
4086
  maxTokens: 128000,
3999
4087
  },
4000
- "gpt-5.3-codex": {
4001
- id: "gpt-5.3-codex",
4002
- name: "GPT-5.3-Codex",
4088
+ "gpt-5.4": {
4089
+ id: "gpt-5.4",
4090
+ name: "GPT-5.4",
4003
4091
  api: "openai-responses",
4004
4092
  provider: "github-copilot",
4005
4093
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4008,17 +4096,17 @@ export const MODELS = {
4008
4096
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4009
4097
  input: ["text", "image"],
4010
4098
  cost: {
4011
- input: 0,
4012
- output: 0,
4013
- cacheRead: 0,
4099
+ input: 2.5,
4100
+ output: 15,
4101
+ cacheRead: 0.25,
4014
4102
  cacheWrite: 0,
4015
4103
  },
4016
4104
  contextWindow: 400000,
4017
4105
  maxTokens: 128000,
4018
4106
  },
4019
- "gpt-5.4": {
4020
- id: "gpt-5.4",
4021
- name: "GPT-5.4",
4107
+ "gpt-5.4-mini": {
4108
+ id: "gpt-5.4-mini",
4109
+ name: "GPT-5.4 mini",
4022
4110
  api: "openai-responses",
4023
4111
  provider: "github-copilot",
4024
4112
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4027,17 +4115,17 @@ export const MODELS = {
4027
4115
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4028
4116
  input: ["text", "image"],
4029
4117
  cost: {
4030
- input: 0,
4031
- output: 0,
4032
- cacheRead: 0,
4118
+ input: 0.75,
4119
+ output: 4.5,
4120
+ cacheRead: 0.075,
4033
4121
  cacheWrite: 0,
4034
4122
  },
4035
4123
  contextWindow: 400000,
4036
4124
  maxTokens: 128000,
4037
4125
  },
4038
- "gpt-5.4-mini": {
4039
- id: "gpt-5.4-mini",
4040
- name: "GPT-5.4 Mini",
4126
+ "gpt-5.4-nano": {
4127
+ id: "gpt-5.4-nano",
4128
+ name: "GPT-5.4 nano",
4041
4129
  api: "openai-responses",
4042
4130
  provider: "github-copilot",
4043
4131
  baseUrl: "https://api.individual.githubcopilot.com",
@@ -4046,9 +4134,9 @@ export const MODELS = {
4046
4134
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4047
4135
  input: ["text", "image"],
4048
4136
  cost: {
4049
- input: 0,
4050
- output: 0,
4051
- cacheRead: 0,
4137
+ input: 0.2,
4138
+ output: 1.25,
4139
+ cacheRead: 0.02,
4052
4140
  cacheWrite: 0,
4053
4141
  },
4054
4142
  contextWindow: 400000,
@@ -4065,32 +4153,32 @@ export const MODELS = {
4065
4153
  thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
4066
4154
  input: ["text", "image"],
4067
4155
  cost: {
4068
- input: 0,
4069
- output: 0,
4070
- cacheRead: 0,
4156
+ input: 5,
4157
+ output: 30,
4158
+ cacheRead: 0.5,
4071
4159
  cacheWrite: 0,
4072
4160
  },
4073
4161
  contextWindow: 400000,
4074
4162
  maxTokens: 128000,
4075
4163
  },
4076
- "grok-code-fast-1": {
4077
- id: "grok-code-fast-1",
4078
- name: "Grok Code Fast 1",
4164
+ "raptor-mini": {
4165
+ id: "raptor-mini",
4166
+ name: "Raptor mini",
4079
4167
  api: "openai-completions",
4080
4168
  provider: "github-copilot",
4081
4169
  baseUrl: "https://api.individual.githubcopilot.com",
4082
4170
  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" },
4083
4171
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4084
4172
  reasoning: true,
4085
- input: ["text"],
4173
+ input: ["text", "image"],
4086
4174
  cost: {
4087
- input: 0,
4088
- output: 0,
4089
- cacheRead: 0,
4175
+ input: 0.25,
4176
+ output: 2,
4177
+ cacheRead: 0.025,
4090
4178
  cacheWrite: 0,
4091
4179
  },
4092
- contextWindow: 128000,
4093
- maxTokens: 64000,
4180
+ contextWindow: 400000,
4181
+ maxTokens: 128000,
4094
4182
  },
4095
4183
  },
4096
4184
  "google": {
@@ -5437,6 +5525,23 @@ export const MODELS = {
5437
5525
  contextWindow: 262144,
5438
5526
  maxTokens: 262144,
5439
5527
  },
5528
+ "devstral-latest": {
5529
+ id: "devstral-latest",
5530
+ name: "Devstral 2",
5531
+ api: "mistral-conversations",
5532
+ provider: "mistral",
5533
+ baseUrl: "https://api.mistral.ai",
5534
+ reasoning: false,
5535
+ input: ["text"],
5536
+ cost: {
5537
+ input: 0.4,
5538
+ output: 2,
5539
+ cacheRead: 0,
5540
+ cacheWrite: 0,
5541
+ },
5542
+ contextWindow: 262144,
5543
+ maxTokens: 262144,
5544
+ },
5440
5545
  "devstral-medium-2507": {
5441
5546
  id: "devstral-medium-2507",
5442
5547
  name: "Devstral Medium",
@@ -5715,11 +5820,11 @@ export const MODELS = {
5715
5820
  api: "mistral-conversations",
5716
5821
  provider: "mistral",
5717
5822
  baseUrl: "https://api.mistral.ai",
5718
- reasoning: true,
5823
+ reasoning: false,
5719
5824
  input: ["text", "image"],
5720
5825
  cost: {
5721
- input: 1.5,
5722
- output: 7.5,
5826
+ input: 0.4,
5827
+ output: 2,
5723
5828
  cacheRead: 0,
5724
5829
  cacheWrite: 0,
5725
5830
  },
@@ -5811,6 +5916,23 @@ export const MODELS = {
5811
5916
  contextWindow: 8000,
5812
5917
  maxTokens: 8000,
5813
5918
  },
5919
+ "open-mistral-nemo": {
5920
+ id: "open-mistral-nemo",
5921
+ name: "Open Mistral Nemo",
5922
+ api: "mistral-conversations",
5923
+ provider: "mistral",
5924
+ baseUrl: "https://api.mistral.ai",
5925
+ reasoning: false,
5926
+ input: ["text"],
5927
+ cost: {
5928
+ input: 0.15,
5929
+ output: 0.15,
5930
+ cacheRead: 0,
5931
+ cacheWrite: 0,
5932
+ },
5933
+ contextWindow: 128000,
5934
+ maxTokens: 128000,
5935
+ },
5814
5936
  "open-mixtral-8x22b": {
5815
5937
  id: "open-mixtral-8x22b",
5816
5938
  name: "Mixtral 8x22B",
@@ -7162,6 +7284,25 @@ export const MODELS = {
7162
7284
  contextWindow: 1000000,
7163
7285
  maxTokens: 64000,
7164
7286
  },
7287
+ "deepseek-v4-flash": {
7288
+ id: "deepseek-v4-flash",
7289
+ name: "DeepSeek V4 Flash",
7290
+ api: "openai-completions",
7291
+ provider: "opencode",
7292
+ baseUrl: "https://opencode.ai/zen/v1",
7293
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7294
+ reasoning: true,
7295
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7296
+ input: ["text"],
7297
+ cost: {
7298
+ input: 0.14,
7299
+ output: 0.28,
7300
+ cacheRead: 0.03,
7301
+ cacheWrite: 0,
7302
+ },
7303
+ contextWindow: 1000000,
7304
+ maxTokens: 384000,
7305
+ },
7165
7306
  "deepseek-v4-flash-free": {
7166
7307
  id: "deepseek-v4-flash-free",
7167
7308
  name: "DeepSeek V4 Flash Free",
@@ -7622,8 +7763,8 @@ export const MODELS = {
7622
7763
  cacheRead: 0,
7623
7764
  cacheWrite: 0,
7624
7765
  },
7625
- contextWindow: 1000000,
7626
- maxTokens: 128000,
7766
+ contextWindow: 200000,
7767
+ maxTokens: 32000,
7627
7768
  },
7628
7769
  "minimax-m2.5": {
7629
7770
  id: "minimax-m2.5",
@@ -7659,9 +7800,26 @@ export const MODELS = {
7659
7800
  contextWindow: 204800,
7660
7801
  maxTokens: 131072,
7661
7802
  },
7662
- "nemotron-3-super-free": {
7663
- id: "nemotron-3-super-free",
7664
- name: "Nemotron 3 Super Free",
7803
+ "minimax-m3-free": {
7804
+ id: "minimax-m3-free",
7805
+ name: "MiniMax M3 Free",
7806
+ api: "anthropic-messages",
7807
+ provider: "opencode",
7808
+ baseUrl: "https://opencode.ai/zen",
7809
+ reasoning: true,
7810
+ input: ["text", "image"],
7811
+ cost: {
7812
+ input: 0,
7813
+ output: 0,
7814
+ cacheRead: 0,
7815
+ cacheWrite: 0,
7816
+ },
7817
+ contextWindow: 200000,
7818
+ maxTokens: 32000,
7819
+ },
7820
+ "nemotron-3-ultra-free": {
7821
+ id: "nemotron-3-ultra-free",
7822
+ name: "Nemotron 3 Ultra Free",
7665
7823
  api: "openai-completions",
7666
7824
  provider: "opencode",
7667
7825
  baseUrl: "https://opencode.ai/zen/v1",
@@ -7673,7 +7831,7 @@ export const MODELS = {
7673
7831
  cacheRead: 0,
7674
7832
  cacheWrite: 0,
7675
7833
  },
7676
- contextWindow: 204800,
7834
+ contextWindow: 1000000,
7677
7835
  maxTokens: 128000,
7678
7836
  },
7679
7837
  "qwen3.5-plus": {
@@ -7886,6 +8044,23 @@ export const MODELS = {
7886
8044
  contextWindow: 204800,
7887
8045
  maxTokens: 131072,
7888
8046
  },
8047
+ "minimax-m3": {
8048
+ id: "minimax-m3",
8049
+ name: "MiniMax M3",
8050
+ api: "anthropic-messages",
8051
+ provider: "opencode-go",
8052
+ baseUrl: "https://opencode.ai/zen/go",
8053
+ reasoning: true,
8054
+ input: ["text", "image"],
8055
+ cost: {
8056
+ input: 0.6,
8057
+ output: 2.4,
8058
+ cacheRead: 0.12,
8059
+ cacheWrite: 0,
8060
+ },
8061
+ contextWindow: 512000,
8062
+ maxTokens: 131072,
8063
+ },
7889
8064
  "qwen3.6-plus": {
7890
8065
  id: "qwen3.6-plus",
7891
8066
  name: "Qwen3.6 Plus",
@@ -7901,7 +8076,7 @@ export const MODELS = {
7901
8076
  cacheRead: 0.05,
7902
8077
  cacheWrite: 0.625,
7903
8078
  },
7904
- contextWindow: 262144,
8079
+ contextWindow: 1000000,
7905
8080
  maxTokens: 65536,
7906
8081
  },
7907
8082
  "qwen3.7-max": {
@@ -7921,6 +8096,23 @@ export const MODELS = {
7921
8096
  contextWindow: 1000000,
7922
8097
  maxTokens: 65536,
7923
8098
  },
8099
+ "qwen3.7-plus": {
8100
+ id: "qwen3.7-plus",
8101
+ name: "Qwen3.7 Plus",
8102
+ api: "anthropic-messages",
8103
+ provider: "opencode-go",
8104
+ baseUrl: "https://opencode.ai/zen/go",
8105
+ reasoning: true,
8106
+ input: ["text", "image"],
8107
+ cost: {
8108
+ input: 0.4,
8109
+ output: 1.6,
8110
+ cacheRead: 0.04,
8111
+ cacheWrite: 0.5,
8112
+ },
8113
+ contextWindow: 1000000,
8114
+ maxTokens: 65536,
8115
+ },
7924
8116
  },
7925
8117
  "openrouter": {
7926
8118
  "ai21/jamba-large-1.7": {
@@ -8465,8 +8657,8 @@ export const MODELS = {
8465
8657
  reasoning: false,
8466
8658
  input: ["text"],
8467
8659
  cost: {
8468
- input: 0.2288,
8469
- output: 0.9144,
8660
+ input: 0.20020000000000002,
8661
+ output: 0.8000999999999999,
8470
8662
  cacheRead: 0,
8471
8663
  cacheWrite: 0,
8472
8664
  },
@@ -8567,13 +8759,13 @@ export const MODELS = {
8567
8759
  reasoning: true,
8568
8760
  input: ["text"],
8569
8761
  cost: {
8570
- input: 0.252,
8571
- output: 0.378,
8572
- cacheRead: 0.0252,
8762
+ input: 0.2288,
8763
+ output: 0.3432,
8764
+ cacheRead: 0,
8573
8765
  cacheWrite: 0,
8574
8766
  },
8575
8767
  contextWindow: 131072,
8576
- maxTokens: 65536,
8768
+ maxTokens: 64000,
8577
8769
  },
8578
8770
  "deepseek/deepseek-v3.2-exp": {
8579
8771
  id: "deepseek/deepseek-v3.2-exp",
@@ -8611,25 +8803,6 @@ export const MODELS = {
8611
8803
  contextWindow: 1048576,
8612
8804
  maxTokens: 131072,
8613
8805
  },
8614
- "deepseek/deepseek-v4-flash:free": {
8615
- id: "deepseek/deepseek-v4-flash:free",
8616
- name: "DeepSeek: DeepSeek V4 Flash (free)",
8617
- api: "openai-completions",
8618
- provider: "openrouter",
8619
- baseUrl: "https://openrouter.ai/api/v1",
8620
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8621
- reasoning: true,
8622
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8623
- input: ["text"],
8624
- cost: {
8625
- input: 0,
8626
- output: 0,
8627
- cacheRead: 0,
8628
- cacheWrite: 0,
8629
- },
8630
- contextWindow: 1048576,
8631
- maxTokens: 384000,
8632
- },
8633
8806
  "deepseek/deepseek-v4-pro": {
8634
8807
  id: "deepseek/deepseek-v4-pro",
8635
8808
  name: "DeepSeek: DeepSeek V4 Pro",
@@ -8666,40 +8839,6 @@ export const MODELS = {
8666
8839
  contextWindow: 32768,
8667
8840
  maxTokens: 4096,
8668
8841
  },
8669
- "google/gemini-2.0-flash-001": {
8670
- id: "google/gemini-2.0-flash-001",
8671
- name: "Google: Gemini 2.0 Flash",
8672
- api: "openai-completions",
8673
- provider: "openrouter",
8674
- baseUrl: "https://openrouter.ai/api/v1",
8675
- reasoning: false,
8676
- input: ["text", "image"],
8677
- cost: {
8678
- input: 0.09999999999999999,
8679
- output: 0.39999999999999997,
8680
- cacheRead: 0.024999999999999998,
8681
- cacheWrite: 0.08333333333333334,
8682
- },
8683
- contextWindow: 1048576,
8684
- maxTokens: 8192,
8685
- },
8686
- "google/gemini-2.0-flash-lite-001": {
8687
- id: "google/gemini-2.0-flash-lite-001",
8688
- name: "Google: Gemini 2.0 Flash Lite",
8689
- api: "openai-completions",
8690
- provider: "openrouter",
8691
- baseUrl: "https://openrouter.ai/api/v1",
8692
- reasoning: false,
8693
- input: ["text", "image"],
8694
- cost: {
8695
- input: 0.075,
8696
- output: 0.3,
8697
- cacheRead: 0,
8698
- cacheWrite: 0,
8699
- },
8700
- contextWindow: 1048576,
8701
- maxTokens: 8192,
8702
- },
8703
8842
  "google/gemini-2.5-flash": {
8704
8843
  id: "google/gemini-2.5-flash",
8705
8844
  name: "Google: Gemini 2.5 Flash",
@@ -8982,12 +9121,12 @@ export const MODELS = {
8982
9121
  input: ["text", "image"],
8983
9122
  cost: {
8984
9123
  input: 0.12,
8985
- output: 0.37,
8986
- cacheRead: 0,
9124
+ output: 0.36,
9125
+ cacheRead: 0.09,
8987
9126
  cacheWrite: 0,
8988
9127
  },
8989
9128
  contextWindow: 262144,
8990
- maxTokens: 16384,
9129
+ maxTokens: 8192,
8991
9130
  },
8992
9131
  "google/gemma-4-31b-it:free": {
8993
9132
  id: "google/gemma-4-31b-it:free",
@@ -9135,7 +9274,7 @@ export const MODELS = {
9135
9274
  input: ["text"],
9136
9275
  cost: {
9137
9276
  input: 0.02,
9138
- output: 0.049999999999999996,
9277
+ output: 0.03,
9139
9278
  cacheRead: 0,
9140
9279
  cacheWrite: 0,
9141
9280
  },
@@ -9287,13 +9426,30 @@ export const MODELS = {
9287
9426
  reasoning: true,
9288
9427
  input: ["text"],
9289
9428
  cost: {
9290
- input: 0.26,
9429
+ input: 0.27899999999999997,
9291
9430
  output: 1.2,
9292
9431
  cacheRead: 0,
9293
9432
  cacheWrite: 0,
9294
9433
  },
9295
9434
  contextWindow: 204800,
9296
- maxTokens: 4096,
9435
+ maxTokens: 196608,
9436
+ },
9437
+ "minimax/minimax-m3": {
9438
+ id: "minimax/minimax-m3",
9439
+ name: "MiniMax: MiniMax M3",
9440
+ api: "openai-completions",
9441
+ provider: "openrouter",
9442
+ baseUrl: "https://openrouter.ai/api/v1",
9443
+ reasoning: true,
9444
+ input: ["text", "image"],
9445
+ cost: {
9446
+ input: 0.3,
9447
+ output: 1.2,
9448
+ cacheRead: 0.06,
9449
+ cacheWrite: 0,
9450
+ },
9451
+ contextWindow: 1048576,
9452
+ maxTokens: 512000,
9297
9453
  },
9298
9454
  "mistralai/codestral-2508": {
9299
9455
  id: "mistralai/codestral-2508",
@@ -9803,7 +9959,41 @@ export const MODELS = {
9803
9959
  cacheWrite: 0,
9804
9960
  },
9805
9961
  contextWindow: 1000000,
9806
- maxTokens: 262144,
9962
+ maxTokens: 262144,
9963
+ },
9964
+ "nvidia/nemotron-3-ultra-550b-a55b": {
9965
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
9966
+ name: "NVIDIA: Nemotron 3 Ultra",
9967
+ api: "openai-completions",
9968
+ provider: "openrouter",
9969
+ baseUrl: "https://openrouter.ai/api/v1",
9970
+ reasoning: true,
9971
+ input: ["text"],
9972
+ cost: {
9973
+ input: 0.5,
9974
+ output: 2.5,
9975
+ cacheRead: 0.15,
9976
+ cacheWrite: 0,
9977
+ },
9978
+ contextWindow: 1000000,
9979
+ maxTokens: 16384,
9980
+ },
9981
+ "nvidia/nemotron-3-ultra-550b-a55b:free": {
9982
+ id: "nvidia/nemotron-3-ultra-550b-a55b:free",
9983
+ name: "NVIDIA: Nemotron 3 Ultra (free)",
9984
+ api: "openai-completions",
9985
+ provider: "openrouter",
9986
+ baseUrl: "https://openrouter.ai/api/v1",
9987
+ reasoning: true,
9988
+ input: ["text"],
9989
+ cost: {
9990
+ input: 0,
9991
+ output: 0,
9992
+ cacheRead: 0,
9993
+ cacheWrite: 0,
9994
+ },
9995
+ contextWindow: 1000000,
9996
+ maxTokens: 65536,
9807
9997
  },
9808
9998
  "nvidia/nemotron-nano-12b-v2-vl:free": {
9809
9999
  id: "nvidia/nemotron-nano-12b-v2-vl:free",
@@ -9924,23 +10114,6 @@ export const MODELS = {
9924
10114
  contextWindow: 8191,
9925
10115
  maxTokens: 4096,
9926
10116
  },
9927
- "openai/gpt-4-0314": {
9928
- id: "openai/gpt-4-0314",
9929
- name: "OpenAI: GPT-4 (older v0314)",
9930
- api: "openai-completions",
9931
- provider: "openrouter",
9932
- baseUrl: "https://openrouter.ai/api/v1",
9933
- reasoning: false,
9934
- input: ["text"],
9935
- cost: {
9936
- input: 30,
9937
- output: 60,
9938
- cacheRead: 0,
9939
- cacheWrite: 0,
9940
- },
9941
- contextWindow: 8191,
9942
- maxTokens: 4096,
9943
- },
9944
10117
  "openai/gpt-4-1106-preview": {
9945
10118
  id: "openai/gpt-4-1106-preview",
9946
10119
  name: "OpenAI: GPT-4 Turbo (older v1106)",
@@ -10971,7 +11144,7 @@ export const MODELS = {
10971
11144
  cacheWrite: 0,
10972
11145
  },
10973
11146
  contextWindow: 131072,
10974
- maxTokens: 20000,
11147
+ maxTokens: 16384,
10975
11148
  },
10976
11149
  "qwen/qwen3-30b-a3b-instruct-2507": {
10977
11150
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -10982,13 +11155,13 @@ export const MODELS = {
10982
11155
  reasoning: false,
10983
11156
  input: ["text"],
10984
11157
  cost: {
10985
- input: 0.09,
10986
- output: 0.3,
11158
+ input: 0.04815,
11159
+ output: 0.19305,
10987
11160
  cacheRead: 0,
10988
11161
  cacheWrite: 0,
10989
11162
  },
10990
- contextWindow: 262144,
10991
- maxTokens: 262144,
11163
+ contextWindow: 131072,
11164
+ maxTokens: 32000,
10992
11165
  },
10993
11166
  "qwen/qwen3-30b-a3b-thinking-2507": {
10994
11167
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -11585,6 +11758,23 @@ export const MODELS = {
11585
11758
  contextWindow: 1000000,
11586
11759
  maxTokens: 65536,
11587
11760
  },
11761
+ "qwen/qwen3.7-plus": {
11762
+ id: "qwen/qwen3.7-plus",
11763
+ name: "Qwen: Qwen3.7 Plus",
11764
+ api: "openai-completions",
11765
+ provider: "openrouter",
11766
+ baseUrl: "https://openrouter.ai/api/v1",
11767
+ reasoning: true,
11768
+ input: ["text", "image"],
11769
+ cost: {
11770
+ input: 0.39999999999999997,
11771
+ output: 1.5999999999999999,
11772
+ cacheRead: 0.08,
11773
+ cacheWrite: 0.5,
11774
+ },
11775
+ contextWindow: 1000000,
11776
+ maxTokens: 65536,
11777
+ },
11588
11778
  "rekaai/reka-edge": {
11589
11779
  id: "rekaai/reka-edge",
11590
11780
  name: "Reka Edge",
@@ -11619,23 +11809,6 @@ export const MODELS = {
11619
11809
  contextWindow: 256000,
11620
11810
  maxTokens: 128000,
11621
11811
  },
11622
- "sao10k/l3-euryale-70b": {
11623
- id: "sao10k/l3-euryale-70b",
11624
- name: "Sao10k: Llama 3 Euryale 70B v2.1",
11625
- api: "openai-completions",
11626
- provider: "openrouter",
11627
- baseUrl: "https://openrouter.ai/api/v1",
11628
- reasoning: false,
11629
- input: ["text"],
11630
- cost: {
11631
- input: 1.48,
11632
- output: 1.48,
11633
- cacheRead: 0,
11634
- cacheWrite: 0,
11635
- },
11636
- contextWindow: 8192,
11637
- maxTokens: 8192,
11638
- },
11639
11812
  "sao10k/l3.1-euryale-70b": {
11640
11813
  id: "sao10k/l3.1-euryale-70b",
11641
11814
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -11721,6 +11894,23 @@ export const MODELS = {
11721
11894
  contextWindow: 32768,
11722
11895
  maxTokens: 32768,
11723
11896
  },
11897
+ "upstage/solar-pro-3": {
11898
+ id: "upstage/solar-pro-3",
11899
+ name: "Upstage: Solar Pro 3",
11900
+ api: "openai-completions",
11901
+ provider: "openrouter",
11902
+ baseUrl: "https://openrouter.ai/api/v1",
11903
+ reasoning: true,
11904
+ input: ["text"],
11905
+ cost: {
11906
+ input: 0.15,
11907
+ output: 0.6,
11908
+ cacheRead: 0.015,
11909
+ cacheWrite: 0,
11910
+ },
11911
+ contextWindow: 128000,
11912
+ maxTokens: 4096,
11913
+ },
11724
11914
  "x-ai/grok-4.20": {
11725
11915
  id: "x-ai/grok-4.20",
11726
11916
  name: "xAI: Grok 4.20",
@@ -12318,7 +12508,7 @@ export const MODELS = {
12318
12508
  },
12319
12509
  "deepseek-ai/DeepSeek-V3": {
12320
12510
  id: "deepseek-ai/DeepSeek-V3",
12321
- name: "DeepSeek V3",
12511
+ name: "DeepSeek-V3",
12322
12512
  api: "openai-completions",
12323
12513
  provider: "together",
12324
12514
  baseUrl: "https://api.together.ai/v1",
@@ -12466,6 +12656,25 @@ export const MODELS = {
12466
12656
  contextWindow: 262144,
12467
12657
  maxTokens: 131000,
12468
12658
  },
12659
+ "nvidia/nemotron-3-ultra-550b-a55b": {
12660
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
12661
+ name: "Nemotron 3 Ultra 550B A55B",
12662
+ api: "openai-completions",
12663
+ provider: "together",
12664
+ baseUrl: "https://api.together.ai/v1",
12665
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12666
+ reasoning: true,
12667
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12668
+ input: ["text"],
12669
+ cost: {
12670
+ input: 0.6,
12671
+ output: 3.6,
12672
+ cacheRead: 0.2,
12673
+ cacheWrite: 0,
12674
+ },
12675
+ contextWindow: 512300,
12676
+ maxTokens: 512300,
12677
+ },
12469
12678
  "openai/gpt-oss-120b": {
12470
12679
  id: "openai/gpt-oss-120b",
12471
12680
  name: "GPT OSS 120B",
@@ -12529,7 +12738,7 @@ export const MODELS = {
12529
12738
  api: "anthropic-messages",
12530
12739
  provider: "vercel-ai-gateway",
12531
12740
  baseUrl: "https://ai-gateway.vercel.sh",
12532
- reasoning: false,
12741
+ reasoning: true,
12533
12742
  input: ["text"],
12534
12743
  cost: {
12535
12744
  input: 0.22,
@@ -12581,7 +12790,7 @@ export const MODELS = {
12581
12790
  provider: "vercel-ai-gateway",
12582
12791
  baseUrl: "https://ai-gateway.vercel.sh",
12583
12792
  reasoning: true,
12584
- input: ["text", "image"],
12793
+ input: ["text"],
12585
12794
  cost: {
12586
12795
  input: 1.3,
12587
12796
  output: 7.8,
@@ -12614,7 +12823,7 @@ export const MODELS = {
12614
12823
  api: "anthropic-messages",
12615
12824
  provider: "vercel-ai-gateway",
12616
12825
  baseUrl: "https://ai-gateway.vercel.sh",
12617
- reasoning: false,
12826
+ reasoning: true,
12618
12827
  input: ["text"],
12619
12828
  cost: {
12620
12829
  input: 1.5,
@@ -12648,7 +12857,7 @@ export const MODELS = {
12648
12857
  api: "anthropic-messages",
12649
12858
  provider: "vercel-ai-gateway",
12650
12859
  baseUrl: "https://ai-gateway.vercel.sh",
12651
- reasoning: false,
12860
+ reasoning: true,
12652
12861
  input: ["text"],
12653
12862
  cost: {
12654
12863
  input: 0.5,
@@ -12727,6 +12936,40 @@ export const MODELS = {
12727
12936
  contextWindow: 256000,
12728
12937
  maxTokens: 65536,
12729
12938
  },
12939
+ "alibaba/qwen3-next-80b-a3b-instruct": {
12940
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
12941
+ name: "Qwen3 Next 80B A3B Instruct",
12942
+ api: "anthropic-messages",
12943
+ provider: "vercel-ai-gateway",
12944
+ baseUrl: "https://ai-gateway.vercel.sh",
12945
+ reasoning: false,
12946
+ input: ["text"],
12947
+ cost: {
12948
+ input: 0.15,
12949
+ output: 1.2,
12950
+ cacheRead: 0,
12951
+ cacheWrite: 0,
12952
+ },
12953
+ contextWindow: 131072,
12954
+ maxTokens: 32768,
12955
+ },
12956
+ "alibaba/qwen3-next-80b-a3b-thinking": {
12957
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
12958
+ name: "Qwen3 Next 80B A3B Thinking",
12959
+ api: "anthropic-messages",
12960
+ provider: "vercel-ai-gateway",
12961
+ baseUrl: "https://ai-gateway.vercel.sh",
12962
+ reasoning: true,
12963
+ input: ["text"],
12964
+ cost: {
12965
+ input: 0.15,
12966
+ output: 1.2,
12967
+ cacheRead: 0,
12968
+ cacheWrite: 0,
12969
+ },
12970
+ contextWindow: 131072,
12971
+ maxTokens: 32768,
12972
+ },
12730
12973
  "alibaba/qwen3-vl-thinking": {
12731
12974
  id: "alibaba/qwen3-vl-thinking",
12732
12975
  name: "Qwen3 VL 235B A22B Thinking",
@@ -12819,7 +13062,7 @@ export const MODELS = {
12819
13062
  provider: "vercel-ai-gateway",
12820
13063
  baseUrl: "https://ai-gateway.vercel.sh",
12821
13064
  reasoning: true,
12822
- input: ["text", "image"],
13065
+ input: ["text"],
12823
13066
  cost: {
12824
13067
  input: 1.25,
12825
13068
  output: 3.75,
@@ -12829,6 +13072,23 @@ export const MODELS = {
12829
13072
  contextWindow: 991000,
12830
13073
  maxTokens: 64000,
12831
13074
  },
13075
+ "alibaba/qwen3.7-plus": {
13076
+ id: "alibaba/qwen3.7-plus",
13077
+ name: "Qwen 3.7 Plus",
13078
+ api: "anthropic-messages",
13079
+ provider: "vercel-ai-gateway",
13080
+ baseUrl: "https://ai-gateway.vercel.sh",
13081
+ reasoning: true,
13082
+ input: ["text", "image"],
13083
+ cost: {
13084
+ input: 0.39999999999999997,
13085
+ output: 1.5999999999999999,
13086
+ cacheRead: 0.08,
13087
+ cacheWrite: 0.5,
13088
+ },
13089
+ contextWindow: 1000000,
13090
+ maxTokens: 64000,
13091
+ },
12832
13092
  "anthropic/claude-3-haiku": {
12833
13093
  id: "anthropic/claude-3-haiku",
12834
13094
  name: "Claude 3 Haiku",
@@ -13113,17 +13373,17 @@ export const MODELS = {
13113
13373
  reasoning: false,
13114
13374
  input: ["text"],
13115
13375
  cost: {
13116
- input: 0.77,
13117
- output: 0.77,
13118
- cacheRead: 0,
13376
+ input: 0.27,
13377
+ output: 1.12,
13378
+ cacheRead: 0.135,
13119
13379
  cacheWrite: 0,
13120
13380
  },
13121
13381
  contextWindow: 163840,
13122
- maxTokens: 16384,
13382
+ maxTokens: 163840,
13123
13383
  },
13124
13384
  "deepseek/deepseek-v3.1": {
13125
13385
  id: "deepseek/deepseek-v3.1",
13126
- name: "DeepSeek-V3.1",
13386
+ name: "DeepSeek V3.1",
13127
13387
  api: "anthropic-messages",
13128
13388
  provider: "vercel-ai-gateway",
13129
13389
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13161,8 +13421,8 @@ export const MODELS = {
13161
13421
  api: "anthropic-messages",
13162
13422
  provider: "vercel-ai-gateway",
13163
13423
  baseUrl: "https://ai-gateway.vercel.sh",
13164
- reasoning: false,
13165
- input: ["text"],
13424
+ reasoning: true,
13425
+ input: ["text", "image"],
13166
13426
  cost: {
13167
13427
  input: 0.28,
13168
13428
  output: 0.42,
@@ -13178,8 +13438,8 @@ export const MODELS = {
13178
13438
  api: "anthropic-messages",
13179
13439
  provider: "vercel-ai-gateway",
13180
13440
  baseUrl: "https://ai-gateway.vercel.sh",
13181
- reasoning: false,
13182
- input: ["text"],
13441
+ reasoning: true,
13442
+ input: ["text", "image"],
13183
13443
  cost: {
13184
13444
  input: 0.62,
13185
13445
  output: 1.85,
@@ -13196,7 +13456,7 @@ export const MODELS = {
13196
13456
  provider: "vercel-ai-gateway",
13197
13457
  baseUrl: "https://ai-gateway.vercel.sh",
13198
13458
  reasoning: true,
13199
- input: ["text"],
13459
+ input: ["text", "image"],
13200
13460
  cost: {
13201
13461
  input: 0.14,
13202
13462
  output: 0.28,
@@ -13416,12 +13676,12 @@ export const MODELS = {
13416
13676
  api: "anthropic-messages",
13417
13677
  provider: "vercel-ai-gateway",
13418
13678
  baseUrl: "https://ai-gateway.vercel.sh",
13419
- reasoning: false,
13679
+ reasoning: true,
13420
13680
  input: ["text", "image"],
13421
13681
  cost: {
13422
- input: 0.13,
13423
- output: 0.39999999999999997,
13424
- cacheRead: 0,
13682
+ input: 0.15,
13683
+ output: 0.6,
13684
+ cacheRead: 0.015,
13425
13685
  cacheWrite: 0,
13426
13686
  },
13427
13687
  contextWindow: 262144,
@@ -13723,7 +13983,7 @@ export const MODELS = {
13723
13983
  provider: "vercel-ai-gateway",
13724
13984
  baseUrl: "https://ai-gateway.vercel.sh",
13725
13985
  reasoning: true,
13726
- input: ["text", "image"],
13986
+ input: ["text"],
13727
13987
  cost: {
13728
13988
  input: 0.3,
13729
13989
  output: 1.2,
@@ -13740,7 +14000,7 @@ export const MODELS = {
13740
14000
  provider: "vercel-ai-gateway",
13741
14001
  baseUrl: "https://ai-gateway.vercel.sh",
13742
14002
  reasoning: true,
13743
- input: ["text", "image"],
14003
+ input: ["text"],
13744
14004
  cost: {
13745
14005
  input: 0.6,
13746
14006
  output: 2.4,
@@ -13750,6 +14010,23 @@ export const MODELS = {
13750
14010
  contextWindow: 204800,
13751
14011
  maxTokens: 131100,
13752
14012
  },
14013
+ "minimax/minimax-m3": {
14014
+ id: "minimax/minimax-m3",
14015
+ name: "MiniMax M3",
14016
+ api: "anthropic-messages",
14017
+ provider: "vercel-ai-gateway",
14018
+ baseUrl: "https://ai-gateway.vercel.sh",
14019
+ reasoning: true,
14020
+ input: ["text", "image"],
14021
+ cost: {
14022
+ input: 0.3,
14023
+ output: 1.2,
14024
+ cacheRead: 0.06,
14025
+ cacheWrite: 0,
14026
+ },
14027
+ contextWindow: 1000000,
14028
+ maxTokens: 1000000,
14029
+ },
13753
14030
  "mistral/codestral": {
13754
14031
  id: "mistral/codestral",
13755
14032
  name: "Mistral Codestral",
@@ -13886,6 +14163,23 @@ export const MODELS = {
13886
14163
  contextWindow: 256000,
13887
14164
  maxTokens: 256000,
13888
14165
  },
14166
+ "mistral/mistral-nemo": {
14167
+ id: "mistral/mistral-nemo",
14168
+ name: "Mistral Nemo 12B",
14169
+ api: "anthropic-messages",
14170
+ provider: "vercel-ai-gateway",
14171
+ baseUrl: "https://ai-gateway.vercel.sh",
14172
+ reasoning: false,
14173
+ input: ["text"],
14174
+ cost: {
14175
+ input: 0.02,
14176
+ output: 0.04,
14177
+ cacheRead: 0,
14178
+ cacheWrite: 0,
14179
+ },
14180
+ contextWindow: 131072,
14181
+ maxTokens: 131072,
14182
+ },
13889
14183
  "mistral/mistral-small": {
13890
14184
  id: "mistral/mistral-small",
13891
14185
  name: "Mistral Small",
@@ -14039,6 +14333,40 @@ export const MODELS = {
14039
14333
  contextWindow: 262000,
14040
14334
  maxTokens: 262000,
14041
14335
  },
14336
+ "nvidia/nemotron-3-super-120b-a12b": {
14337
+ id: "nvidia/nemotron-3-super-120b-a12b",
14338
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
14339
+ api: "anthropic-messages",
14340
+ provider: "vercel-ai-gateway",
14341
+ baseUrl: "https://ai-gateway.vercel.sh",
14342
+ reasoning: true,
14343
+ input: ["text"],
14344
+ cost: {
14345
+ input: 0.15,
14346
+ output: 0.65,
14347
+ cacheRead: 0,
14348
+ cacheWrite: 0,
14349
+ },
14350
+ contextWindow: 256000,
14351
+ maxTokens: 32000,
14352
+ },
14353
+ "nvidia/nemotron-3-ultra-550b-a55b": {
14354
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
14355
+ name: "Nemotron 3 Ultra",
14356
+ api: "anthropic-messages",
14357
+ provider: "vercel-ai-gateway",
14358
+ baseUrl: "https://ai-gateway.vercel.sh",
14359
+ reasoning: true,
14360
+ input: ["text"],
14361
+ cost: {
14362
+ input: 0.6,
14363
+ output: 2.4,
14364
+ cacheRead: 0.12,
14365
+ cacheWrite: 0,
14366
+ },
14367
+ contextWindow: 1000000,
14368
+ maxTokens: 65000,
14369
+ },
14042
14370
  "nvidia/nemotron-nano-12b-v2-vl": {
14043
14371
  id: "nvidia/nemotron-nano-12b-v2-vl",
14044
14372
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -14199,7 +14527,7 @@ export const MODELS = {
14199
14527
  provider: "vercel-ai-gateway",
14200
14528
  baseUrl: "https://ai-gateway.vercel.sh",
14201
14529
  reasoning: true,
14202
- input: ["text"],
14530
+ input: ["text", "image"],
14203
14531
  cost: {
14204
14532
  input: 1.25,
14205
14533
  output: 10,
@@ -14561,6 +14889,23 @@ export const MODELS = {
14561
14889
  contextWindow: 1000000,
14562
14890
  maxTokens: 128000,
14563
14891
  },
14892
+ "openai/gpt-oss-120b": {
14893
+ id: "openai/gpt-oss-120b",
14894
+ name: "GPT OSS 120B",
14895
+ api: "anthropic-messages",
14896
+ provider: "vercel-ai-gateway",
14897
+ baseUrl: "https://ai-gateway.vercel.sh",
14898
+ reasoning: true,
14899
+ input: ["text"],
14900
+ cost: {
14901
+ input: 0.35,
14902
+ output: 0.75,
14903
+ cacheRead: 0.25,
14904
+ cacheWrite: 0,
14905
+ },
14906
+ contextWindow: 131072,
14907
+ maxTokens: 131000,
14908
+ },
14564
14909
  "openai/gpt-oss-20b": {
14565
14910
  id: "openai/gpt-oss-20b",
14566
14911
  name: "GPT OSS 20B",
@@ -14731,6 +15076,23 @@ export const MODELS = {
14731
15076
  contextWindow: 200000,
14732
15077
  maxTokens: 8000,
14733
15078
  },
15079
+ "stepfun/step-3.5-flash": {
15080
+ id: "stepfun/step-3.5-flash",
15081
+ name: "StepFun 3.5 Flash",
15082
+ api: "anthropic-messages",
15083
+ provider: "vercel-ai-gateway",
15084
+ baseUrl: "https://ai-gateway.vercel.sh",
15085
+ reasoning: true,
15086
+ input: ["text"],
15087
+ cost: {
15088
+ input: 0.09,
15089
+ output: 0.3,
15090
+ cacheRead: 0,
15091
+ cacheWrite: 0.02,
15092
+ },
15093
+ contextWindow: 262114,
15094
+ maxTokens: 262114,
15095
+ },
14734
15096
  "stepfun/step-3.7-flash": {
14735
15097
  id: "stepfun/step-3.7-flash",
14736
15098
  name: "Step 3.7 Flash",
@@ -14992,7 +15354,7 @@ export const MODELS = {
14992
15354
  api: "anthropic-messages",
14993
15355
  provider: "vercel-ai-gateway",
14994
15356
  baseUrl: "https://ai-gateway.vercel.sh",
14995
- reasoning: false,
15357
+ reasoning: true,
14996
15358
  input: ["text", "image"],
14997
15359
  cost: {
14998
15360
  input: 0.6,
@@ -15146,7 +15508,7 @@ export const MODELS = {
15146
15508
  provider: "vercel-ai-gateway",
15147
15509
  baseUrl: "https://ai-gateway.vercel.sh",
15148
15510
  reasoning: true,
15149
- input: ["text"],
15511
+ input: ["text", "image"],
15150
15512
  cost: {
15151
15513
  input: 1.4,
15152
15514
  output: 4.4,