@caupulican/pi-ai 0.96.0 → 0.96.2

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.
Files changed (13) hide show
  1. package/dist/models.generated.d.ts +5210 -3786
  2. package/dist/models.generated.d.ts.map +1 -1
  3. package/dist/models.generated.js +1927 -689
  4. package/dist/models.generated.js.map +1 -1
  5. package/dist/providers/provider-runtime.d.ts +1 -1
  6. package/dist/providers/provider-runtime.d.ts.map +1 -1
  7. package/dist/providers/provider-runtime.js +3 -1
  8. package/dist/providers/provider-runtime.js.map +1 -1
  9. package/dist/utils/provider-retry.d.ts +11 -1
  10. package/dist/utils/provider-retry.d.ts.map +1 -1
  11. package/dist/utils/provider-retry.js +128 -27
  12. package/dist/utils/provider-retry.js.map +1 -1
  13. package/package.json +1 -1
package/dist/models.generated.js CHANGED
@@ -815,6 +815,63 @@ export const MODELS = {
815
815
  contextWindow: 1000000,
816
816
  maxTokens: 128000,
817
817
  },
818
+ "global.openai.gpt-5.6-luna": {
819
+ id: "global.openai.gpt-5.6-luna",
820
+ name: "GPT-5.6 Luna (Global)",
821
+ api: "bedrock-converse-stream",
822
+ provider: "amazon-bedrock",
823
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
824
+ reasoning: true,
825
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
826
+ input: ["text", "image"],
827
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
828
+ cost: {
829
+ input: 0.22,
830
+ output: 1.32,
831
+ cacheRead: 0.022,
832
+ cacheWrite: 0.275,
833
+ },
834
+ contextWindow: 1050000,
835
+ maxTokens: 128000,
836
+ },
837
+ "global.openai.gpt-5.6-sol": {
838
+ id: "global.openai.gpt-5.6-sol",
839
+ name: "GPT-5.6 Sol (Global)",
840
+ api: "bedrock-converse-stream",
841
+ provider: "amazon-bedrock",
842
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
843
+ reasoning: true,
844
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
845
+ input: ["text", "image"],
846
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
847
+ cost: {
848
+ input: 5.5,
849
+ output: 33,
850
+ cacheRead: 0.55,
851
+ cacheWrite: 6.875,
852
+ },
853
+ contextWindow: 1050000,
854
+ maxTokens: 128000,
855
+ },
856
+ "global.openai.gpt-5.6-terra": {
857
+ id: "global.openai.gpt-5.6-terra",
858
+ name: "GPT-5.6 Terra (Global)",
859
+ api: "bedrock-converse-stream",
860
+ provider: "amazon-bedrock",
861
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
862
+ reasoning: true,
863
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
864
+ input: ["text", "image"],
865
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
866
+ cost: {
867
+ input: 2.2,
868
+ output: 13.2,
869
+ cacheRead: 0.22,
870
+ cacheWrite: 2.75,
871
+ },
872
+ contextWindow: 1050000,
873
+ maxTokens: 128000,
874
+ },
818
875
  "google.gemma-3-27b-it": {
819
876
  id: "google.gemma-3-27b-it",
820
877
  name: "Google Gemma 3 27B Instruct",
@@ -2008,6 +2065,24 @@ export const MODELS = {
2008
2065
  contextWindow: 1000000,
2009
2066
  maxTokens: 131072,
2010
2067
  },
2068
+ "xai.grok-4.6": {
2069
+ id: "xai.grok-4.6",
2070
+ name: "Grok 4.6",
2071
+ api: "bedrock-converse-stream",
2072
+ provider: "amazon-bedrock",
2073
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2074
+ reasoning: true,
2075
+ input: ["text", "image"],
2076
+ supportedImageMimeTypes: ["image/jpeg", "image/jpg", "image/png", "image/gif", "image/webp"],
2077
+ cost: {
2078
+ input: 2.2,
2079
+ output: 6.6,
2080
+ cacheRead: 0.55,
2081
+ cacheWrite: 0,
2082
+ },
2083
+ contextWindow: 500000,
2084
+ maxTokens: 500000,
2085
+ },
2011
2086
  "zai.glm-4.7": {
2012
2087
  id: "zai.glm-4.7",
2013
2088
  name: "GLM-4.7",
@@ -2983,127 +3058,8 @@ export const MODELS = {
2983
3058
  contextWindow: 131072,
2984
3059
  maxTokens: 40960,
2985
3060
  },
2986
- "zai-glm-4.7": {
2987
- id: "zai-glm-4.7",
2988
- name: "Z.AI GLM-4.7",
2989
- api: "openai-completions",
2990
- provider: "cerebras",
2991
- baseUrl: "https://api.cerebras.ai/v1",
2992
- reasoning: true,
2993
- input: ["text"],
2994
- cost: {
2995
- input: 2.25,
2996
- output: 2.75,
2997
- cacheRead: 2.25,
2998
- cacheWrite: 0,
2999
- },
3000
- contextWindow: 131072,
3001
- maxTokens: 40960,
3002
- },
3003
3061
  },
3004
3062
  "cloudflare-ai-gateway": {
3005
- "claude-3-5-haiku": {
3006
- id: "claude-3-5-haiku",
3007
- name: "Claude Haiku 3.5 (latest)",
3008
- api: "anthropic-messages",
3009
- provider: "cloudflare-ai-gateway",
3010
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3011
- reasoning: false,
3012
- input: ["text", "image"],
3013
- cost: {
3014
- input: 0.8,
3015
- output: 4,
3016
- cacheRead: 0.08,
3017
- cacheWrite: 1,
3018
- },
3019
- contextWindow: 200000,
3020
- maxTokens: 8192,
3021
- },
3022
- "claude-3-haiku": {
3023
- id: "claude-3-haiku",
3024
- name: "Claude Haiku 3",
3025
- api: "anthropic-messages",
3026
- provider: "cloudflare-ai-gateway",
3027
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3028
- reasoning: false,
3029
- input: ["text", "image"],
3030
- cost: {
3031
- input: 0.25,
3032
- output: 1.25,
3033
- cacheRead: 0.03,
3034
- cacheWrite: 0.3,
3035
- },
3036
- contextWindow: 200000,
3037
- maxTokens: 4096,
3038
- },
3039
- "claude-3-opus": {
3040
- id: "claude-3-opus",
3041
- name: "Claude Opus 3",
3042
- api: "anthropic-messages",
3043
- provider: "cloudflare-ai-gateway",
3044
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3045
- reasoning: false,
3046
- input: ["text", "image"],
3047
- cost: {
3048
- input: 15,
3049
- output: 75,
3050
- cacheRead: 1.5,
3051
- cacheWrite: 18.75,
3052
- },
3053
- contextWindow: 200000,
3054
- maxTokens: 4096,
3055
- },
3056
- "claude-3-sonnet": {
3057
- id: "claude-3-sonnet",
3058
- name: "Claude Sonnet 3",
3059
- api: "anthropic-messages",
3060
- provider: "cloudflare-ai-gateway",
3061
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3062
- reasoning: false,
3063
- input: ["text", "image"],
3064
- cost: {
3065
- input: 3,
3066
- output: 15,
3067
- cacheRead: 0.3,
3068
- cacheWrite: 0.3,
3069
- },
3070
- contextWindow: 200000,
3071
- maxTokens: 4096,
3072
- },
3073
- "claude-3.5-haiku": {
3074
- id: "claude-3.5-haiku",
3075
- name: "Claude Haiku 3.5 (latest)",
3076
- api: "anthropic-messages",
3077
- provider: "cloudflare-ai-gateway",
3078
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3079
- reasoning: false,
3080
- input: ["text", "image"],
3081
- cost: {
3082
- input: 0.8,
3083
- output: 4,
3084
- cacheRead: 0.08,
3085
- cacheWrite: 1,
3086
- },
3087
- contextWindow: 200000,
3088
- maxTokens: 8192,
3089
- },
3090
- "claude-3.5-sonnet": {
3091
- id: "claude-3.5-sonnet",
3092
- name: "Claude Sonnet 3.5 v2",
3093
- api: "anthropic-messages",
3094
- provider: "cloudflare-ai-gateway",
3095
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3096
- reasoning: false,
3097
- input: ["text", "image"],
3098
- cost: {
3099
- input: 3,
3100
- output: 15,
3101
- cacheRead: 0.3,
3102
- cacheWrite: 3.75,
3103
- },
3104
- contextWindow: 200000,
3105
- maxTokens: 8192,
3106
- },
3107
3063
  "claude-fable-5": {
3108
3064
  id: "claude-fable-5",
3109
3065
  name: "Claude Fable 5",
@@ -3140,40 +3096,6 @@ export const MODELS = {
3140
3096
  contextWindow: 200000,
3141
3097
  maxTokens: 64000,
3142
3098
  },
3143
- "claude-opus-4": {
3144
- id: "claude-opus-4",
3145
- name: "Claude Opus 4 (latest)",
3146
- api: "anthropic-messages",
3147
- provider: "cloudflare-ai-gateway",
3148
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3149
- reasoning: true,
3150
- input: ["text", "image"],
3151
- cost: {
3152
- input: 15,
3153
- output: 75,
3154
- cacheRead: 1.5,
3155
- cacheWrite: 18.75,
3156
- },
3157
- contextWindow: 200000,
3158
- maxTokens: 32000,
3159
- },
3160
- "claude-opus-4-1": {
3161
- id: "claude-opus-4-1",
3162
- name: "Claude Opus 4.1 (latest)",
3163
- api: "anthropic-messages",
3164
- provider: "cloudflare-ai-gateway",
3165
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3166
- reasoning: true,
3167
- input: ["text", "image"],
3168
- cost: {
3169
- input: 15,
3170
- output: 75,
3171
- cacheRead: 1.5,
3172
- cacheWrite: 18.75,
3173
- },
3174
- contextWindow: 200000,
3175
- maxTokens: 32000,
3176
- },
3177
3099
  "claude-opus-4-5": {
3178
3100
  id: "claude-opus-4-5",
3179
3101
  name: "Claude Opus 4.5 (latest)",
@@ -3193,7 +3115,7 @@ export const MODELS = {
3193
3115
  },
3194
3116
  "claude-opus-4-6": {
3195
3117
  id: "claude-opus-4-6",
3196
- name: "Claude Opus 4.6 (latest)",
3118
+ name: "Claude Opus 4.6",
3197
3119
  api: "anthropic-messages",
3198
3120
  provider: "cloudflare-ai-gateway",
3199
3121
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
@@ -3267,23 +3189,6 @@ export const MODELS = {
3267
3189
  contextWindow: 1000000,
3268
3190
  maxTokens: 128000,
3269
3191
  },
3270
- "claude-sonnet-4": {
3271
- id: "claude-sonnet-4",
3272
- name: "Claude Sonnet 4 (latest)",
3273
- api: "anthropic-messages",
3274
- provider: "cloudflare-ai-gateway",
3275
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3276
- reasoning: true,
3277
- input: ["text", "image"],
3278
- cost: {
3279
- input: 3,
3280
- output: 15,
3281
- cacheRead: 0.3,
3282
- cacheWrite: 3.75,
3283
- },
3284
- contextWindow: 200000,
3285
- maxTokens: 64000,
3286
- },
3287
3192
  "claude-sonnet-4-5": {
3288
3193
  id: "claude-sonnet-4-5",
3289
3194
  name: "Claude Sonnet 4.5 (latest)",
@@ -3298,7 +3203,7 @@ export const MODELS = {
3298
3203
  cacheRead: 0.3,
3299
3204
  cacheWrite: 3.75,
3300
3205
  },
3301
- contextWindow: 200000,
3206
+ contextWindow: 1000000,
3302
3207
  maxTokens: 64000,
3303
3208
  },
3304
3209
  "claude-sonnet-4-6": {
@@ -3318,7 +3223,7 @@ export const MODELS = {
3318
3223
  cacheWrite: 3.75,
3319
3224
  },
3320
3225
  contextWindow: 1000000,
3321
- maxTokens: 64000,
3226
+ maxTokens: 128000,
3322
3227
  },
3323
3228
  "claude-sonnet-5": {
3324
3229
  id: "claude-sonnet-5",
@@ -3373,61 +3278,166 @@ export const MODELS = {
3373
3278
  contextWindow: 128000,
3374
3279
  maxTokens: 4096,
3375
3280
  },
3376
- "gpt-4o": {
3377
- id: "gpt-4o",
3378
- name: "GPT-4o",
3281
+ "gpt-4.1": {
3282
+ id: "gpt-4.1",
3283
+ name: "GPT-4.1",
3379
3284
  api: "openai-responses",
3380
3285
  provider: "cloudflare-ai-gateway",
3381
3286
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3382
3287
  reasoning: false,
3383
3288
  input: ["text", "image"],
3384
3289
  cost: {
3385
- input: 2.5,
3386
- output: 10,
3387
- cacheRead: 1.25,
3290
+ input: 2,
3291
+ output: 8,
3292
+ cacheRead: 0.5,
3388
3293
  cacheWrite: 0,
3389
3294
  },
3390
- contextWindow: 128000,
3391
- maxTokens: 16384,
3295
+ contextWindow: 1047576,
3296
+ maxTokens: 32768,
3392
3297
  },
3393
- "gpt-4o-mini": {
3394
- id: "gpt-4o-mini",
3395
- name: "GPT-4o mini",
3298
+ "gpt-4.1-mini": {
3299
+ id: "gpt-4.1-mini",
3300
+ name: "GPT-4.1 mini",
3396
3301
  api: "openai-responses",
3397
3302
  provider: "cloudflare-ai-gateway",
3398
3303
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3399
3304
  reasoning: false,
3400
3305
  input: ["text", "image"],
3401
3306
  cost: {
3402
- input: 0.15,
3403
- output: 0.6,
3404
- cacheRead: 0.08,
3307
+ input: 0.4,
3308
+ output: 1.6,
3309
+ cacheRead: 0.1,
3405
3310
  cacheWrite: 0,
3406
3311
  },
3407
- contextWindow: 128000,
3408
- maxTokens: 16384,
3312
+ contextWindow: 1047576,
3313
+ maxTokens: 32768,
3409
3314
  },
3410
- "gpt-5.1": {
3411
- id: "gpt-5.1",
3412
- name: "GPT-5.1",
3315
+ "gpt-4.1-nano": {
3316
+ id: "gpt-4.1-nano",
3317
+ name: "GPT-4.1 nano",
3413
3318
  api: "openai-responses",
3414
3319
  provider: "cloudflare-ai-gateway",
3415
3320
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3416
- reasoning: true,
3417
- thinkingLevelMap: { "off": null },
3321
+ reasoning: false,
3418
3322
  input: ["text", "image"],
3419
3323
  cost: {
3420
- input: 1.25,
3421
- output: 10,
3422
- cacheRead: 0.13,
3324
+ input: 0.1,
3325
+ output: 0.4,
3326
+ cacheRead: 0.025,
3423
3327
  cacheWrite: 0,
3424
3328
  },
3425
- contextWindow: 400000,
3329
+ contextWindow: 1047576,
3330
+ maxTokens: 32768,
3331
+ },
3332
+ "gpt-4o": {
3333
+ id: "gpt-4o",
3334
+ name: "GPT-4o",
3335
+ api: "openai-responses",
3336
+ provider: "cloudflare-ai-gateway",
3337
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3338
+ reasoning: false,
3339
+ input: ["text", "image"],
3340
+ cost: {
3341
+ input: 1.25,
3342
+ output: 5,
3343
+ cacheRead: 0.625,
3344
+ cacheWrite: 0,
3345
+ },
3346
+ contextWindow: 128000,
3347
+ maxTokens: 16384,
3348
+ },
3349
+ "gpt-4o-mini": {
3350
+ id: "gpt-4o-mini",
3351
+ name: "GPT-4o mini",
3352
+ api: "openai-responses",
3353
+ provider: "cloudflare-ai-gateway",
3354
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3355
+ reasoning: false,
3356
+ input: ["text", "image"],
3357
+ cost: {
3358
+ input: 0.075,
3359
+ output: 0.3,
3360
+ cacheRead: 0.0375,
3361
+ cacheWrite: 0,
3362
+ },
3363
+ contextWindow: 128000,
3364
+ maxTokens: 16384,
3365
+ },
3366
+ "gpt-5": {
3367
+ id: "gpt-5",
3368
+ name: "GPT-5",
3369
+ api: "openai-responses",
3370
+ provider: "cloudflare-ai-gateway",
3371
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3372
+ reasoning: true,
3373
+ thinkingLevelMap: { "off": null },
3374
+ input: ["text", "image"],
3375
+ cost: {
3376
+ input: 1.25,
3377
+ output: 10,
3378
+ cacheRead: 0.125,
3379
+ cacheWrite: 0,
3380
+ },
3381
+ contextWindow: 400000,
3426
3382
  maxTokens: 128000,
3427
3383
  },
3428
- "gpt-5.1-codex": {
3429
- id: "gpt-5.1-codex",
3430
- name: "GPT-5.1 Codex",
3384
+ "gpt-5-mini": {
3385
+ id: "gpt-5-mini",
3386
+ name: "GPT-5 Mini",
3387
+ api: "openai-responses",
3388
+ provider: "cloudflare-ai-gateway",
3389
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3390
+ reasoning: true,
3391
+ thinkingLevelMap: { "off": null },
3392
+ input: ["text", "image"],
3393
+ cost: {
3394
+ input: 0.25,
3395
+ output: 2,
3396
+ cacheRead: 0.025,
3397
+ cacheWrite: 0,
3398
+ },
3399
+ contextWindow: 400000,
3400
+ maxTokens: 128000,
3401
+ },
3402
+ "gpt-5-nano": {
3403
+ id: "gpt-5-nano",
3404
+ name: "GPT-5 Nano",
3405
+ api: "openai-responses",
3406
+ provider: "cloudflare-ai-gateway",
3407
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3408
+ reasoning: true,
3409
+ thinkingLevelMap: { "off": null },
3410
+ input: ["text", "image"],
3411
+ cost: {
3412
+ input: 0.05,
3413
+ output: 0.4,
3414
+ cacheRead: 0.005,
3415
+ cacheWrite: 0,
3416
+ },
3417
+ contextWindow: 400000,
3418
+ maxTokens: 128000,
3419
+ },
3420
+ "gpt-5-pro": {
3421
+ id: "gpt-5-pro",
3422
+ name: "GPT-5 Pro",
3423
+ api: "openai-responses",
3424
+ provider: "cloudflare-ai-gateway",
3425
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3426
+ reasoning: true,
3427
+ thinkingLevelMap: { "off": null },
3428
+ input: ["text", "image"],
3429
+ cost: {
3430
+ input: 15,
3431
+ output: 120,
3432
+ cacheRead: 0,
3433
+ cacheWrite: 0,
3434
+ },
3435
+ contextWindow: 400000,
3436
+ maxTokens: 272000,
3437
+ },
3438
+ "gpt-5.1": {
3439
+ id: "gpt-5.1",
3440
+ name: "GPT-5.1",
3431
3441
  api: "openai-responses",
3432
3442
  provider: "cloudflare-ai-gateway",
3433
3443
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
@@ -3461,9 +3471,9 @@ export const MODELS = {
3461
3471
  contextWindow: 400000,
3462
3472
  maxTokens: 128000,
3463
3473
  },
3464
- "gpt-5.2-codex": {
3465
- id: "gpt-5.2-codex",
3466
- name: "GPT-5.2 Codex",
3474
+ "gpt-5.2-chat-latest": {
3475
+ id: "gpt-5.2-chat-latest",
3476
+ name: "GPT-5.2 Chat",
3467
3477
  api: "openai-responses",
3468
3478
  provider: "cloudflare-ai-gateway",
3469
3479
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
@@ -3476,9 +3486,45 @@ export const MODELS = {
3476
3486
  cacheRead: 0.175,
3477
3487
  cacheWrite: 0,
3478
3488
  },
3489
+ contextWindow: 128000,
3490
+ maxTokens: 16384,
3491
+ },
3492
+ "gpt-5.2-pro": {
3493
+ id: "gpt-5.2-pro",
3494
+ name: "GPT-5.2 Pro",
3495
+ api: "openai-responses",
3496
+ provider: "cloudflare-ai-gateway",
3497
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3498
+ reasoning: true,
3499
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3500
+ input: ["text", "image"],
3501
+ cost: {
3502
+ input: 21,
3503
+ output: 168,
3504
+ cacheRead: 0,
3505
+ cacheWrite: 0,
3506
+ },
3479
3507
  contextWindow: 400000,
3480
3508
  maxTokens: 128000,
3481
3509
  },
3510
+ "gpt-5.3-chat-latest": {
3511
+ id: "gpt-5.3-chat-latest",
3512
+ name: "GPT-5.3 Chat (latest)",
3513
+ api: "openai-responses",
3514
+ provider: "cloudflare-ai-gateway",
3515
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3516
+ reasoning: false,
3517
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3518
+ input: ["text", "image"],
3519
+ cost: {
3520
+ input: 1.75,
3521
+ output: 14,
3522
+ cacheRead: 0.175,
3523
+ cacheWrite: 0,
3524
+ },
3525
+ contextWindow: 128000,
3526
+ maxTokens: 16384,
3527
+ },
3482
3528
  "gpt-5.3-codex": {
3483
3529
  id: "gpt-5.3-codex",
3484
3530
  name: "GPT-5.3 Codex",
@@ -3497,6 +3543,24 @@ export const MODELS = {
3497
3543
  contextWindow: 400000,
3498
3544
  maxTokens: 128000,
3499
3545
  },
3546
+ "gpt-5.3-codex-spark": {
3547
+ id: "gpt-5.3-codex-spark",
3548
+ name: "GPT-5.3 Codex Spark",
3549
+ api: "openai-responses",
3550
+ provider: "cloudflare-ai-gateway",
3551
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3552
+ reasoning: true,
3553
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3554
+ input: ["text", "image"],
3555
+ cost: {
3556
+ input: 1.75,
3557
+ output: 14,
3558
+ cacheRead: 0.175,
3559
+ cacheWrite: 0,
3560
+ },
3561
+ contextWindow: 128000,
3562
+ maxTokens: 32000,
3563
+ },
3500
3564
  "gpt-5.4": {
3501
3565
  id: "gpt-5.4",
3502
3566
  name: "GPT-5.4",
@@ -3515,6 +3579,60 @@ export const MODELS = {
3515
3579
  contextWindow: 1050000,
3516
3580
  maxTokens: 128000,
3517
3581
  },
3582
+ "gpt-5.4-mini": {
3583
+ id: "gpt-5.4-mini",
3584
+ name: "GPT-5.4 mini",
3585
+ api: "openai-responses",
3586
+ provider: "cloudflare-ai-gateway",
3587
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3588
+ reasoning: true,
3589
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3590
+ input: ["text", "image"],
3591
+ cost: {
3592
+ input: 0.75,
3593
+ output: 4.5,
3594
+ cacheRead: 0.075,
3595
+ cacheWrite: 0,
3596
+ },
3597
+ contextWindow: 400000,
3598
+ maxTokens: 128000,
3599
+ },
3600
+ "gpt-5.4-nano": {
3601
+ id: "gpt-5.4-nano",
3602
+ name: "GPT-5.4 nano",
3603
+ api: "openai-responses",
3604
+ provider: "cloudflare-ai-gateway",
3605
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3606
+ reasoning: true,
3607
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3608
+ input: ["text", "image"],
3609
+ cost: {
3610
+ input: 0.2,
3611
+ output: 1.25,
3612
+ cacheRead: 0.02,
3613
+ cacheWrite: 0,
3614
+ },
3615
+ contextWindow: 400000,
3616
+ maxTokens: 128000,
3617
+ },
3618
+ "gpt-5.4-pro": {
3619
+ id: "gpt-5.4-pro",
3620
+ name: "GPT-5.4 Pro",
3621
+ api: "openai-responses",
3622
+ provider: "cloudflare-ai-gateway",
3623
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3624
+ reasoning: true,
3625
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
3626
+ input: ["text", "image"],
3627
+ cost: {
3628
+ input: 30,
3629
+ output: 180,
3630
+ cacheRead: 0,
3631
+ cacheWrite: 0,
3632
+ },
3633
+ contextWindow: 1050000,
3634
+ maxTokens: 128000,
3635
+ },
3518
3636
  "gpt-5.5": {
3519
3637
  id: "gpt-5.5",
3520
3638
  name: "GPT-5.5",
@@ -3533,6 +3651,42 @@ export const MODELS = {
3533
3651
  contextWindow: 1050000,
3534
3652
  maxTokens: 128000,
3535
3653
  },
3654
+ "gpt-5.5-pro": {
3655
+ id: "gpt-5.5-pro",
3656
+ name: "GPT-5.5 Pro",
3657
+ api: "openai-responses",
3658
+ provider: "cloudflare-ai-gateway",
3659
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3660
+ reasoning: true,
3661
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "minimal": null, "low": null },
3662
+ input: ["text", "image"],
3663
+ cost: {
3664
+ input: 30,
3665
+ output: 180,
3666
+ cacheRead: 0,
3667
+ cacheWrite: 0,
3668
+ },
3669
+ contextWindow: 1050000,
3670
+ maxTokens: 128000,
3671
+ },
3672
+ "gpt-5.6": {
3673
+ id: "gpt-5.6",
3674
+ name: "GPT-5.6",
3675
+ api: "openai-responses",
3676
+ provider: "cloudflare-ai-gateway",
3677
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3678
+ reasoning: true,
3679
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
3680
+ input: ["text", "image"],
3681
+ cost: {
3682
+ input: 5,
3683
+ output: 30,
3684
+ cacheRead: 0.5,
3685
+ cacheWrite: 6.25,
3686
+ },
3687
+ contextWindow: 1050000,
3688
+ maxTokens: 128000,
3689
+ },
3536
3690
  "gpt-5.6-luna": {
3537
3691
  id: "gpt-5.6-luna",
3538
3692
  name: "GPT-5.6 Luna",
@@ -3543,10 +3697,10 @@ export const MODELS = {
3543
3697
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
3544
3698
  input: ["text", "image"],
3545
3699
  cost: {
3546
- input: 1,
3547
- output: 6,
3548
- cacheRead: 0.1,
3549
- cacheWrite: 0,
3700
+ input: 0.2,
3701
+ output: 1.2,
3702
+ cacheRead: 0.02,
3703
+ cacheWrite: 0.25,
3550
3704
  },
3551
3705
  contextWindow: 1050000,
3552
3706
  maxTokens: 128000,
@@ -3564,7 +3718,7 @@ export const MODELS = {
3564
3718
  input: 5,
3565
3719
  output: 30,
3566
3720
  cacheRead: 0.5,
3567
- cacheWrite: 0,
3721
+ cacheWrite: 6.25,
3568
3722
  },
3569
3723
  contextWindow: 1050000,
3570
3724
  maxTokens: 128000,
@@ -3579,10 +3733,10 @@ export const MODELS = {
3579
3733
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
3580
3734
  input: ["text", "image"],
3581
3735
  cost: {
3582
- input: 2.5,
3583
- output: 15,
3584
- cacheRead: 0.25,
3585
- cacheWrite: 0,
3736
+ input: 2,
3737
+ output: 12,
3738
+ cacheRead: 0.2,
3739
+ cacheWrite: 2.5,
3586
3740
  },
3587
3741
  contextWindow: 1050000,
3588
3742
  maxTokens: 128000,
@@ -3604,6 +3758,23 @@ export const MODELS = {
3604
3758
  contextWindow: 200000,
3605
3759
  maxTokens: 100000,
3606
3760
  },
3761
+ "o1-pro": {
3762
+ id: "o1-pro",
3763
+ name: "o1-pro",
3764
+ api: "openai-responses",
3765
+ provider: "cloudflare-ai-gateway",
3766
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3767
+ reasoning: true,
3768
+ input: ["text", "image"],
3769
+ cost: {
3770
+ input: 150,
3771
+ output: 600,
3772
+ cacheRead: 0,
3773
+ cacheWrite: 0,
3774
+ },
3775
+ contextWindow: 200000,
3776
+ maxTokens: 100000,
3777
+ },
3607
3778
  "o3": {
3608
3779
  id: "o3",
3609
3780
  name: "o3",
@@ -3660,57 +3831,201 @@ export const MODELS = {
3660
3831
  name: "o4-mini",
3661
3832
  api: "openai-responses",
3662
3833
  provider: "cloudflare-ai-gateway",
3663
- baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3834
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3835
+ reasoning: true,
3836
+ input: ["text", "image"],
3837
+ cost: {
3838
+ input: 1.1,
3839
+ output: 4.4,
3840
+ cacheRead: 0.275,
3841
+ cacheWrite: 0,
3842
+ },
3843
+ contextWindow: 200000,
3844
+ maxTokens: 100000,
3845
+ },
3846
+ "workers-ai/@cf/google/gemma-4-26b-a4b-it": {
3847
+ id: "workers-ai/@cf/google/gemma-4-26b-a4b-it",
3848
+ name: "Gemma 4 26B A4B IT",
3849
+ api: "openai-completions",
3850
+ provider: "cloudflare-ai-gateway",
3851
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3852
+ compat: { "sendSessionAffinityHeaders": true },
3853
+ reasoning: true,
3854
+ input: ["text", "image"],
3855
+ cost: {
3856
+ input: 0.1,
3857
+ output: 0.3,
3858
+ cacheRead: 0,
3859
+ cacheWrite: 0,
3860
+ },
3861
+ contextWindow: 256000,
3862
+ maxTokens: 16384,
3863
+ },
3864
+ "workers-ai/@cf/ibm-granite/granite-4.0-h-micro": {
3865
+ id: "workers-ai/@cf/ibm-granite/granite-4.0-h-micro",
3866
+ name: "Granite 4.0 H Micro",
3867
+ api: "openai-completions",
3868
+ provider: "cloudflare-ai-gateway",
3869
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3870
+ compat: { "sendSessionAffinityHeaders": true },
3871
+ reasoning: false,
3872
+ input: ["text"],
3873
+ cost: {
3874
+ input: 0.017,
3875
+ output: 0.112,
3876
+ cacheRead: 0,
3877
+ cacheWrite: 0,
3878
+ },
3879
+ contextWindow: 131000,
3880
+ maxTokens: 131000,
3881
+ },
3882
+ "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast": {
3883
+ id: "workers-ai/@cf/meta/llama-3.3-70b-instruct-fp8-fast",
3884
+ name: "Llama 3.3 70B Instruct fp8 Fast",
3885
+ api: "openai-completions",
3886
+ provider: "cloudflare-ai-gateway",
3887
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3888
+ compat: { "sendSessionAffinityHeaders": true },
3889
+ reasoning: false,
3890
+ input: ["text"],
3891
+ cost: {
3892
+ input: 0.293,
3893
+ output: 2.253,
3894
+ cacheRead: 0,
3895
+ cacheWrite: 0,
3896
+ },
3897
+ contextWindow: 24000,
3898
+ maxTokens: 24000,
3899
+ },
3900
+ "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct": {
3901
+ id: "workers-ai/@cf/meta/llama-4-scout-17b-16e-instruct",
3902
+ name: "Llama 4 Scout 17B 16E Instruct",
3903
+ api: "openai-completions",
3904
+ provider: "cloudflare-ai-gateway",
3905
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3906
+ compat: { "sendSessionAffinityHeaders": true },
3907
+ reasoning: false,
3908
+ input: ["text", "image"],
3909
+ cost: {
3910
+ input: 0.27,
3911
+ output: 0.85,
3912
+ cacheRead: 0,
3913
+ cacheWrite: 0,
3914
+ },
3915
+ contextWindow: 131000,
3916
+ maxTokens: 16384,
3917
+ },
3918
+ "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct": {
3919
+ id: "workers-ai/@cf/mistralai/mistral-small-3.1-24b-instruct",
3920
+ name: "Mistral Small 3.1 24B Instruct",
3921
+ api: "openai-completions",
3922
+ provider: "cloudflare-ai-gateway",
3923
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3924
+ compat: { "sendSessionAffinityHeaders": true },
3925
+ reasoning: false,
3926
+ input: ["text"],
3927
+ cost: {
3928
+ input: 0.351,
3929
+ output: 0.555,
3930
+ cacheRead: 0,
3931
+ cacheWrite: 0,
3932
+ },
3933
+ contextWindow: 128000,
3934
+ maxTokens: 128000,
3935
+ },
3936
+ "workers-ai/@cf/moonshotai/kimi-k2.6": {
3937
+ id: "workers-ai/@cf/moonshotai/kimi-k2.6",
3938
+ name: "Kimi K2.6",
3939
+ api: "openai-completions",
3940
+ provider: "cloudflare-ai-gateway",
3941
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3942
+ compat: { "sendSessionAffinityHeaders": true },
3943
+ reasoning: true,
3944
+ input: ["text", "image"],
3945
+ cost: {
3946
+ input: 0.95,
3947
+ output: 4,
3948
+ cacheRead: 0.16,
3949
+ cacheWrite: 0,
3950
+ },
3951
+ contextWindow: 262144,
3952
+ maxTokens: 256000,
3953
+ },
3954
+ "workers-ai/@cf/moonshotai/kimi-k2.7-code": {
3955
+ id: "workers-ai/@cf/moonshotai/kimi-k2.7-code",
3956
+ name: "Kimi K2.7 Code",
3957
+ api: "openai-completions",
3958
+ provider: "cloudflare-ai-gateway",
3959
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3960
+ compat: { "sendSessionAffinityHeaders": true },
3961
+ reasoning: true,
3962
+ input: ["text", "image"],
3963
+ cost: {
3964
+ input: 0.95,
3965
+ output: 4,
3966
+ cacheRead: 0.19,
3967
+ cacheWrite: 0,
3968
+ },
3969
+ contextWindow: 262144,
3970
+ maxTokens: 262144,
3971
+ },
3972
+ "workers-ai/@cf/nvidia/nemotron-3-120b-a12b": {
3973
+ id: "workers-ai/@cf/nvidia/nemotron-3-120b-a12b",
3974
+ name: "Nemotron 3 Super 120B",
3975
+ api: "openai-completions",
3976
+ provider: "cloudflare-ai-gateway",
3977
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3978
+ compat: { "sendSessionAffinityHeaders": true },
3664
3979
  reasoning: true,
3665
- input: ["text", "image"],
3980
+ input: ["text"],
3666
3981
  cost: {
3667
- input: 1.1,
3668
- output: 4.4,
3669
- cacheRead: 0.28,
3982
+ input: 0.5,
3983
+ output: 1.5,
3984
+ cacheRead: 0,
3670
3985
  cacheWrite: 0,
3671
3986
  },
3672
- contextWindow: 200000,
3673
- maxTokens: 100000,
3987
+ contextWindow: 256000,
3988
+ maxTokens: 256000,
3674
3989
  },
3675
- "workers-ai/@cf/moonshotai/kimi-k2.5": {
3676
- id: "workers-ai/@cf/moonshotai/kimi-k2.5",
3677
- name: "Kimi K2.5",
3990
+ "workers-ai/@cf/openai/gpt-oss-120b": {
3991
+ id: "workers-ai/@cf/openai/gpt-oss-120b",
3992
+ name: "GPT OSS 120B",
3678
3993
  api: "openai-completions",
3679
3994
  provider: "cloudflare-ai-gateway",
3680
3995
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3681
3996
  compat: { "sendSessionAffinityHeaders": true },
3682
3997
  reasoning: true,
3683
- input: ["text", "image"],
3998
+ input: ["text"],
3684
3999
  cost: {
3685
- input: 0.6,
3686
- output: 3,
3687
- cacheRead: 0.1,
4000
+ input: 0.35,
4001
+ output: 0.75,
4002
+ cacheRead: 0,
3688
4003
  cacheWrite: 0,
3689
4004
  },
3690
- contextWindow: 256000,
3691
- maxTokens: 256000,
4005
+ contextWindow: 128000,
4006
+ maxTokens: 16384,
3692
4007
  },
3693
- "workers-ai/@cf/moonshotai/kimi-k2.6": {
3694
- id: "workers-ai/@cf/moonshotai/kimi-k2.6",
3695
- name: "Kimi K2.6",
4008
+ "workers-ai/@cf/openai/gpt-oss-20b": {
4009
+ id: "workers-ai/@cf/openai/gpt-oss-20b",
4010
+ name: "GPT OSS 20B",
3696
4011
  api: "openai-completions",
3697
4012
  provider: "cloudflare-ai-gateway",
3698
4013
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3699
4014
  compat: { "sendSessionAffinityHeaders": true },
3700
4015
  reasoning: true,
3701
- input: ["text", "image"],
4016
+ input: ["text"],
3702
4017
  cost: {
3703
- input: 0.95,
3704
- output: 4,
3705
- cacheRead: 0.16,
4018
+ input: 0.2,
4019
+ output: 0.3,
4020
+ cacheRead: 0,
3706
4021
  cacheWrite: 0,
3707
4022
  },
3708
- contextWindow: 256000,
3709
- maxTokens: 256000,
4023
+ contextWindow: 128000,
4024
+ maxTokens: 16384,
3710
4025
  },
3711
- "workers-ai/@cf/nvidia/nemotron-3-120b-a12b": {
3712
- id: "workers-ai/@cf/nvidia/nemotron-3-120b-a12b",
3713
- name: "Nemotron 3 Super 120B",
4026
+ "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8": {
4027
+ id: "workers-ai/@cf/qwen/qwen3-30b-a3b-fp8",
4028
+ name: "Qwen3 30B A3b fp8",
3714
4029
  api: "openai-completions",
3715
4030
  provider: "cloudflare-ai-gateway",
3716
4031
  baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
@@ -3718,13 +4033,13 @@ export const MODELS = {
3718
4033
  reasoning: true,
3719
4034
  input: ["text"],
3720
4035
  cost: {
3721
- input: 0.5,
3722
- output: 1.5,
4036
+ input: 0.0509,
4037
+ output: 0.335,
3723
4038
  cacheRead: 0,
3724
4039
  cacheWrite: 0,
3725
4040
  },
3726
- contextWindow: 256000,
3727
- maxTokens: 256000,
4041
+ contextWindow: 32768,
4042
+ maxTokens: 32768,
3728
4043
  },
3729
4044
  "workers-ai/@cf/zai-org/glm-4.7-flash": {
3730
4045
  id: "workers-ai/@cf/zai-org/glm-4.7-flash",
@@ -3736,7 +4051,7 @@ export const MODELS = {
3736
4051
  reasoning: true,
3737
4052
  input: ["text"],
3738
4053
  cost: {
3739
- input: 0.06,
4054
+ input: 0.0605,
3740
4055
  output: 0.4,
3741
4056
  cacheRead: 0,
3742
4057
  cacheWrite: 0,
@@ -3760,10 +4075,48 @@ export const MODELS = {
3760
4075
  cacheWrite: 0,
3761
4076
  },
3762
4077
  contextWindow: 262144,
3763
- maxTokens: 262144,
4078
+ maxTokens: 256000,
3764
4079
  },
3765
4080
  },
3766
4081
  "cloudflare-workers-ai": {
4082
+ "@cf/deepseek-ai/deepseek-v4-flash-0731": {
4083
+ id: "@cf/deepseek-ai/deepseek-v4-flash-0731",
4084
+ name: "DeepSeek V4 Flash 0731",
4085
+ api: "openai-completions",
4086
+ provider: "cloudflare-workers-ai",
4087
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
4088
+ compat: { "sendSessionAffinityHeaders": true },
4089
+ reasoning: true,
4090
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
4091
+ input: ["text"],
4092
+ cost: {
4093
+ input: 0.44,
4094
+ output: 1.32,
4095
+ cacheRead: 0.014,
4096
+ cacheWrite: 0,
4097
+ },
4098
+ contextWindow: 1310720,
4099
+ maxTokens: 1048576,
4100
+ },
4101
+ "@cf/deepseek-ai/deepseek-v4-pro-0813": {
4102
+ id: "@cf/deepseek-ai/deepseek-v4-pro-0813",
4103
+ name: "DeepSeek V4 Pro 0813",
4104
+ api: "openai-completions",
4105
+ provider: "cloudflare-workers-ai",
4106
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
4107
+ compat: { "sendSessionAffinityHeaders": true },
4108
+ reasoning: true,
4109
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
4110
+ input: ["text"],
4111
+ cost: {
4112
+ input: 1.32,
4113
+ output: 3.96,
4114
+ cacheRead: 0.044,
4115
+ cacheWrite: 0,
4116
+ },
4117
+ contextWindow: 1048576,
4118
+ maxTokens: 1048576,
4119
+ },
3767
4120
  "@cf/google/gemma-4-26b-a4b-it": {
3768
4121
  id: "@cf/google/gemma-4-26b-a4b-it",
3769
4122
  name: "Gemma 4 26B A4B IT",
@@ -3962,6 +4315,24 @@ export const MODELS = {
3962
4315
  contextWindow: 32768,
3963
4316
  maxTokens: 32768,
3964
4317
  },
4318
+ "@cf/qwen/qwen3.8-27b": {
4319
+ id: "@cf/qwen/qwen3.8-27b",
4320
+ name: "Qwen3.8 27B",
4321
+ api: "openai-completions",
4322
+ provider: "cloudflare-workers-ai",
4323
+ baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
4324
+ compat: { "sendSessionAffinityHeaders": true },
4325
+ reasoning: true,
4326
+ input: ["text", "image"],
4327
+ cost: {
4328
+ input: 0.45,
4329
+ output: 3.2,
4330
+ cacheRead: 0.05,
4331
+ cacheWrite: 0,
4332
+ },
4333
+ contextWindow: 262144,
4334
+ maxTokens: 262144,
4335
+ },
3965
4336
  "@cf/zai-org/glm-4.7-flash": {
3966
4337
  id: "@cf/zai-org/glm-4.7-flash",
3967
4338
  name: "GLM-4.7-Flash",
@@ -4094,6 +4465,24 @@ export const MODELS = {
4094
4465
  contextWindow: 1000000,
4095
4466
  maxTokens: 384000,
4096
4467
  },
4468
+ "accounts/fireworks/models/deepseek-v4-pro-0813": {
4469
+ id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4470
+ name: "DeepSeek V4 Pro 0813",
4471
+ api: "anthropic-messages",
4472
+ provider: "fireworks",
4473
+ baseUrl: "https://api.fireworks.ai/inference",
4474
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4475
+ reasoning: true,
4476
+ input: ["text"],
4477
+ cost: {
4478
+ input: 1.32,
4479
+ output: 3.96,
4480
+ cacheRead: 0.044,
4481
+ cacheWrite: 0,
4482
+ },
4483
+ contextWindow: 1000000,
4484
+ maxTokens: 384000,
4485
+ },
4097
4486
  "accounts/fireworks/models/glm-5p2": {
4098
4487
  id: "accounts/fireworks/models/glm-5p2",
4099
4488
  name: "GLM 5.2",
@@ -4752,6 +5141,25 @@ export const MODELS = {
4752
5141
  contextWindow: 1000000,
4753
5142
  maxTokens: 64000,
4754
5143
  },
5144
+ "gemini-3.7-flash": {
5145
+ id: "gemini-3.7-flash",
5146
+ name: "Gemini 3.7 Flash",
5147
+ api: "openai-completions",
5148
+ provider: "github-copilot",
5149
+ baseUrl: "https://api.individual.githubcopilot.com",
5150
+ 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" },
5151
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
5152
+ reasoning: true,
5153
+ input: ["text", "image"],
5154
+ cost: {
5155
+ input: 0.75,
5156
+ output: 3.75,
5157
+ cacheRead: 0.075,
5158
+ cacheWrite: 0,
5159
+ },
5160
+ contextWindow: 1000000,
5161
+ maxTokens: 64000,
5162
+ },
4755
5163
  "gpt-4.1": {
4756
5164
  id: "gpt-4.1",
4757
5165
  name: "GPT-4.1",
@@ -4953,10 +5361,10 @@ export const MODELS = {
4953
5361
  thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh", "max": "max" },
4954
5362
  input: ["text", "image"],
4955
5363
  cost: {
4956
- input: 5,
4957
- output: 30,
4958
- cacheRead: 0.5,
4959
- cacheWrite: 6.25,
5364
+ input: 2.5,
5365
+ output: 15,
5366
+ cacheRead: 0.25,
5367
+ cacheWrite: 3.125,
4960
5368
  },
4961
5369
  contextWindow: 1050000,
4962
5370
  maxTokens: 128000,
@@ -4999,6 +5407,25 @@ export const MODELS = {
4999
5407
  contextWindow: 500000,
5000
5408
  maxTokens: 128000,
5001
5409
  },
5410
+ "grok-4.6": {
5411
+ id: "grok-4.6",
5412
+ name: "Grok 4.6",
5413
+ api: "openai-completions",
5414
+ provider: "github-copilot",
5415
+ baseUrl: "https://api.individual.githubcopilot.com",
5416
+ 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" },
5417
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
5418
+ reasoning: true,
5419
+ input: ["text", "image"],
5420
+ cost: {
5421
+ input: 2,
5422
+ output: 6,
5423
+ cacheRead: 0.5,
5424
+ cacheWrite: 0,
5425
+ },
5426
+ contextWindow: 500000,
5427
+ maxTokens: 128000,
5428
+ },
5002
5429
  "kimi-k2.7-code": {
5003
5430
  id: "kimi-k2.7-code",
5004
5431
  name: "Kimi K2.7 Code",
@@ -5366,9 +5793,9 @@ export const MODELS = {
5366
5793
  thinkingLevelMap: { "off": null },
5367
5794
  input: ["text", "image"],
5368
5795
  cost: {
5369
- input: 1.5,
5370
- output: 7.5,
5371
- cacheRead: 0.15,
5796
+ input: 0.75,
5797
+ output: 3.75,
5798
+ cacheRead: 0.075,
5372
5799
  cacheWrite: 0,
5373
5800
  },
5374
5801
  contextWindow: 1048576,
@@ -5401,9 +5828,9 @@ export const MODELS = {
5401
5828
  reasoning: true,
5402
5829
  input: ["text", "image"],
5403
5830
  cost: {
5404
- input: 1.5,
5405
- output: 9,
5406
- cacheRead: 0.15,
5831
+ input: 0.75,
5832
+ output: 3.75,
5833
+ cacheRead: 0.075,
5407
5834
  cacheWrite: 0,
5408
5835
  },
5409
5836
  contextWindow: 1048576,
@@ -5418,9 +5845,9 @@ export const MODELS = {
5418
5845
  reasoning: true,
5419
5846
  input: ["text", "image"],
5420
5847
  cost: {
5421
- input: 0.25,
5422
- output: 1.5,
5423
- cacheRead: 0.025,
5848
+ input: 0.3,
5849
+ output: 2.5,
5850
+ cacheRead: 0.03,
5424
5851
  cacheWrite: 0,
5425
5852
  },
5426
5853
  contextWindow: 1048576,
@@ -5903,6 +6330,24 @@ export const MODELS = {
5903
6330
  contextWindow: 524288,
5904
6331
  maxTokens: 128000,
5905
6332
  },
6333
+ "Qwen/Qwen2.5-Coder-32B-Instruct": {
6334
+ id: "Qwen/Qwen2.5-Coder-32B-Instruct",
6335
+ name: "Qwen2.5-Coder-32B-Instruct",
6336
+ api: "openai-completions",
6337
+ provider: "huggingface",
6338
+ baseUrl: "https://router.huggingface.co/v1",
6339
+ compat: { "supportsDeveloperRole": false },
6340
+ reasoning: false,
6341
+ input: ["text"],
6342
+ cost: {
6343
+ input: 0.06,
6344
+ output: 0.2,
6345
+ cacheRead: 0,
6346
+ cacheWrite: 0,
6347
+ },
6348
+ contextWindow: 131072,
6349
+ maxTokens: 8192,
6350
+ },
5906
6351
  "Qwen/Qwen3-235B-A22B": {
5907
6352
  id: "Qwen/Qwen3-235B-A22B",
5908
6353
  name: "Qwen3 235B-A22B",
@@ -5957,6 +6402,24 @@ export const MODELS = {
5957
6402
  contextWindow: 262144,
5958
6403
  maxTokens: 131072,
5959
6404
  },
6405
+ "Qwen/Qwen3-30B-A3B": {
6406
+ id: "Qwen/Qwen3-30B-A3B",
6407
+ name: "Qwen3 30B A3B",
6408
+ api: "openai-completions",
6409
+ provider: "huggingface",
6410
+ baseUrl: "https://router.huggingface.co/v1",
6411
+ compat: { "supportsDeveloperRole": false },
6412
+ reasoning: true,
6413
+ input: ["text"],
6414
+ cost: {
6415
+ input: 0.12,
6416
+ output: 0.5,
6417
+ cacheRead: 0,
6418
+ cacheWrite: 0,
6419
+ },
6420
+ contextWindow: 40960,
6421
+ maxTokens: 16384,
6422
+ },
5960
6423
  "Qwen/Qwen3-32B": {
5961
6424
  id: "Qwen/Qwen3-32B",
5962
6425
  name: "Qwen3 32B",
@@ -6065,6 +6528,42 @@ export const MODELS = {
6065
6528
  contextWindow: 262144,
6066
6529
  maxTokens: 131072,
6067
6530
  },
6531
+ "Qwen/Qwen3-VL-235B-A22B-Instruct": {
6532
+ id: "Qwen/Qwen3-VL-235B-A22B-Instruct",
6533
+ name: "Qwen3 VL 235B A22B Instruct",
6534
+ api: "openai-completions",
6535
+ provider: "huggingface",
6536
+ baseUrl: "https://router.huggingface.co/v1",
6537
+ compat: { "supportsDeveloperRole": false },
6538
+ reasoning: false,
6539
+ input: ["text", "image"],
6540
+ cost: {
6541
+ input: 0.3,
6542
+ output: 1.5,
6543
+ cacheRead: 0,
6544
+ cacheWrite: 0,
6545
+ },
6546
+ contextWindow: 131072,
6547
+ maxTokens: 32768,
6548
+ },
6549
+ "Qwen/Qwen3-VL-235B-A22B-Thinking": {
6550
+ id: "Qwen/Qwen3-VL-235B-A22B-Thinking",
6551
+ name: "Qwen3 VL 235B A22B Thinking",
6552
+ api: "openai-completions",
6553
+ provider: "huggingface",
6554
+ baseUrl: "https://router.huggingface.co/v1",
6555
+ compat: { "supportsDeveloperRole": false },
6556
+ reasoning: true,
6557
+ input: ["text", "image"],
6558
+ cost: {
6559
+ input: 0.98,
6560
+ output: 3.95,
6561
+ cacheRead: 0,
6562
+ cacheWrite: 0,
6563
+ },
6564
+ contextWindow: 131072,
6565
+ maxTokens: 32768,
6566
+ },
6068
6567
  "Qwen/Qwen3.5-122B-A10B": {
6069
6568
  id: "Qwen/Qwen3.5-122B-A10B",
6070
6569
  name: "Qwen3.5 122B-A10B",
@@ -6183,13 +6682,31 @@ export const MODELS = {
6183
6682
  reasoning: true,
6184
6683
  input: ["text", "image"],
6185
6684
  cost: {
6186
- input: 0.15,
6187
- output: 0.95,
6685
+ input: 0.15,
6686
+ output: 0.95,
6687
+ cacheRead: 0,
6688
+ cacheWrite: 0,
6689
+ },
6690
+ contextWindow: 262144,
6691
+ maxTokens: 65536,
6692
+ },
6693
+ "Qwen/Qwen3.8-2.4T-A95B": {
6694
+ id: "Qwen/Qwen3.8-2.4T-A95B",
6695
+ name: "Qwen3.8 2.4T A95B",
6696
+ api: "openai-completions",
6697
+ provider: "huggingface",
6698
+ baseUrl: "https://router.huggingface.co/v1",
6699
+ compat: { "supportsDeveloperRole": false },
6700
+ reasoning: true,
6701
+ input: ["text"],
6702
+ cost: {
6703
+ input: 2.5,
6704
+ output: 6.25,
6188
6705
  cacheRead: 0,
6189
6706
  cacheWrite: 0,
6190
6707
  },
6191
6708
  contextWindow: 262144,
6192
- maxTokens: 65536,
6709
+ maxTokens: 131072,
6193
6710
  },
6194
6711
  "XiaomiMiMo/MiMo-V2-Flash": {
6195
6712
  id: "XiaomiMiMo/MiMo-V2-Flash",
@@ -6407,6 +6924,24 @@ export const MODELS = {
6407
6924
  contextWindow: 1048576,
6408
6925
  maxTokens: 393216,
6409
6926
  },
6927
+ "deepseek-ai/DeepSeek-V4-Pro-0813": {
6928
+ id: "deepseek-ai/DeepSeek-V4-Pro-0813",
6929
+ name: "DeepSeek V4 Pro 0813",
6930
+ api: "openai-completions",
6931
+ provider: "huggingface",
6932
+ baseUrl: "https://router.huggingface.co/v1",
6933
+ compat: { "supportsDeveloperRole": false },
6934
+ reasoning: true,
6935
+ input: ["text"],
6936
+ cost: {
6937
+ input: 1.32,
6938
+ output: 3.96,
6939
+ cacheRead: 0,
6940
+ cacheWrite: 0,
6941
+ },
6942
+ contextWindow: 1000000,
6943
+ maxTokens: 384000,
6944
+ },
6410
6945
  "google/gemma-4-26B-A4B-it": {
6411
6946
  id: "google/gemma-4-26B-A4B-it",
6412
6947
  name: "Gemma 4 26B A4B IT",
@@ -6443,6 +6978,24 @@ export const MODELS = {
6443
6978
  contextWindow: 262144,
6444
6979
  maxTokens: 32768,
6445
6980
  },
6981
+ "meta-llama/Llama-3.1-8B-Instruct": {
6982
+ id: "meta-llama/Llama-3.1-8B-Instruct",
6983
+ name: "Llama-3.1-8B-Instruct",
6984
+ api: "openai-completions",
6985
+ provider: "huggingface",
6986
+ baseUrl: "https://router.huggingface.co/v1",
6987
+ compat: { "supportsDeveloperRole": false },
6988
+ reasoning: false,
6989
+ input: ["text"],
6990
+ cost: {
6991
+ input: 0.06,
6992
+ output: 0.06,
6993
+ cacheRead: 0,
6994
+ cacheWrite: 0,
6995
+ },
6996
+ contextWindow: 131072,
6997
+ maxTokens: 4096,
6998
+ },
6446
6999
  "meta-llama/Llama-3.3-70B-Instruct": {
6447
7000
  id: "meta-llama/Llama-3.3-70B-Instruct",
6448
7001
  name: "Llama-3.3-70B-Instruct",
@@ -6785,6 +7338,24 @@ export const MODELS = {
6785
7338
  contextWindow: 204800,
6786
7339
  maxTokens: 131072,
6787
7340
  },
7341
+ "zai-org/GLM-4.6V-Flash": {
7342
+ id: "zai-org/GLM-4.6V-Flash",
7343
+ name: "GLM-4.6V-Flash",
7344
+ api: "openai-completions",
7345
+ provider: "huggingface",
7346
+ baseUrl: "https://router.huggingface.co/v1",
7347
+ compat: { "supportsDeveloperRole": false },
7348
+ reasoning: true,
7349
+ input: ["text", "image"],
7350
+ cost: {
7351
+ input: 0.3,
7352
+ output: 0.9,
7353
+ cacheRead: 0,
7354
+ cacheWrite: 0,
7355
+ },
7356
+ contextWindow: 131072,
7357
+ maxTokens: 32768,
7358
+ },
6788
7359
  "zai-org/GLM-4.7": {
6789
7360
  id: "zai-org/GLM-4.7",
6790
7361
  name: "GLM-4.7",
@@ -9077,24 +9648,6 @@ export const MODELS = {
9077
9648
  contextWindow: 1000000,
9078
9649
  maxTokens: 384000,
9079
9650
  },
9080
- "deepseek-v4-flash-free": {
9081
- id: "deepseek-v4-flash-free",
9082
- name: "DeepSeek V4 Flash Free",
9083
- api: "openai-completions",
9084
- provider: "opencode",
9085
- baseUrl: "https://opencode.ai/zen/v1",
9086
- reasoning: true,
9087
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9088
- input: ["text"],
9089
- cost: {
9090
- input: 0,
9091
- output: 0,
9092
- cacheRead: 0,
9093
- cacheWrite: 0,
9094
- },
9095
- contextWindow: 200000,
9096
- maxTokens: 128000,
9097
- },
9098
9651
  "deepseek-v4-pro": {
9099
9652
  id: "deepseek-v4-pro",
9100
9653
  name: "DeepSeek V4 Pro",
@@ -9597,7 +10150,7 @@ export const MODELS = {
9597
10150
  },
9598
10151
  "gpt-5.6-sol": {
9599
10152
  id: "gpt-5.6-sol",
9600
- name: "GPT-5.6 Sol",
10153
+ name: "GPT-5.6 Sol (50% Off)",
9601
10154
  api: "openai-responses",
9602
10155
  provider: "opencode",
9603
10156
  baseUrl: "https://opencode.ai/zen/v1",
@@ -9606,10 +10159,10 @@ export const MODELS = {
9606
10159
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
9607
10160
  input: ["text", "image"],
9608
10161
  cost: {
9609
- input: 5,
9610
- output: 30,
9611
- cacheRead: 0.5,
9612
- cacheWrite: 6.25,
10162
+ input: 2,
10163
+ output: 10,
10164
+ cacheRead: 0.2,
10165
+ cacheWrite: 2.5,
9613
10166
  },
9614
10167
  contextWindow: 1050000,
9615
10168
  maxTokens: 128000,
@@ -9774,23 +10327,6 @@ export const MODELS = {
9774
10327
  contextWindow: 1048576,
9775
10328
  maxTokens: 131072,
9776
10329
  },
9777
- "laguna-s-2.1-free": {
9778
- id: "laguna-s-2.1-free",
9779
- name: "Laguna S 2.1 Free",
9780
- api: "openai-completions",
9781
- provider: "opencode",
9782
- baseUrl: "https://opencode.ai/zen/v1",
9783
- reasoning: true,
9784
- input: ["text"],
9785
- cost: {
9786
- input: 0,
9787
- output: 0,
9788
- cacheRead: 0,
9789
- cacheWrite: 0,
9790
- },
9791
- contextWindow: 256000,
9792
- maxTokens: 32000,
9793
- },
9794
10330
  "mimo-v2.5-free": {
9795
10331
  id: "mimo-v2.5-free",
9796
10332
  name: "MiMo V2.5 Free",
@@ -9859,6 +10395,42 @@ export const MODELS = {
9859
10395
  contextWindow: 512000,
9860
10396
  maxTokens: 128000,
9861
10397
  },
10398
+ "muse-spark-1.2": {
10399
+ id: "muse-spark-1.2",
10400
+ name: "Muse Spark 1.2",
10401
+ api: "openai-responses",
10402
+ provider: "opencode",
10403
+ baseUrl: "https://opencode.ai/zen/v1",
10404
+ compat: { "sessionAffinityFormat": "openai-nosession" },
10405
+ reasoning: true,
10406
+ input: ["text", "image"],
10407
+ cost: {
10408
+ input: 1.25,
10409
+ output: 4.25,
10410
+ cacheRead: 0.15,
10411
+ cacheWrite: 0,
10412
+ },
10413
+ contextWindow: 1048576,
10414
+ maxTokens: 131072,
10415
+ },
10416
+ "muse-spark-1.2-contributor-free": {
10417
+ id: "muse-spark-1.2-contributor-free",
10418
+ name: "Muse Spark 1.2 Free",
10419
+ api: "openai-responses",
10420
+ provider: "opencode",
10421
+ baseUrl: "https://opencode.ai/zen/v1",
10422
+ compat: { "sessionAffinityFormat": "openai-nosession" },
10423
+ reasoning: true,
10424
+ input: ["text", "image"],
10425
+ cost: {
10426
+ input: 0,
10427
+ output: 0,
10428
+ cacheRead: 0,
10429
+ cacheWrite: 0,
10430
+ },
10431
+ contextWindow: 1048576,
10432
+ maxTokens: 131072,
10433
+ },
9862
10434
  "nemotron-3-ultra-free": {
9863
10435
  id: "nemotron-3-ultra-free",
9864
10436
  name: "Nemotron 3 Ultra Free",
@@ -9927,11 +10499,28 @@ export const MODELS = {
9927
10499
  contextWindow: 262144,
9928
10500
  maxTokens: 65536,
9929
10501
  },
10502
+ "x-preview-f-free": {
10503
+ id: "x-preview-f-free",
10504
+ name: "Ox Alpha Free (Unlimited)",
10505
+ api: "openai-completions",
10506
+ provider: "opencode",
10507
+ baseUrl: "https://opencode.ai/zen/v1",
10508
+ reasoning: true,
10509
+ input: ["text", "image"],
10510
+ cost: {
10511
+ input: 0,
10512
+ output: 0,
10513
+ cacheRead: 0,
10514
+ cacheWrite: 0,
10515
+ },
10516
+ contextWindow: 1000000,
10517
+ maxTokens: 131072,
10518
+ },
9930
10519
  },
9931
10520
  "opencode-go": {
9932
10521
  "deepseek-v4-flash": {
9933
10522
  id: "deepseek-v4-flash",
9934
- name: "DeepSeek V4 Flash (2x usage)",
10523
+ name: "DeepSeek V4 Flash",
9935
10524
  api: "openai-completions",
9936
10525
  provider: "opencode-go",
9937
10526
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -9939,9 +10528,27 @@ export const MODELS = {
9939
10528
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9940
10529
  input: ["text"],
9941
10530
  cost: {
9942
- input: 0.07,
9943
- output: 0.14,
9944
- cacheRead: 0.0014,
10531
+ input: 0.22,
10532
+ output: 0.66,
10533
+ cacheRead: 0.007,
10534
+ cacheWrite: 0,
10535
+ },
10536
+ contextWindow: 1000000,
10537
+ maxTokens: 384000,
10538
+ },
10539
+ "deepseek-v4-flash-vision-exp": {
10540
+ id: "deepseek-v4-flash-vision-exp",
10541
+ name: "DeepSeek V4 Flash Vision Exp",
10542
+ api: "openai-completions",
10543
+ provider: "opencode-go",
10544
+ baseUrl: "https://opencode.ai/zen/go/v1",
10545
+ reasoning: true,
10546
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10547
+ input: ["text", "image"],
10548
+ cost: {
10549
+ input: 0.22,
10550
+ output: 0.66,
10551
+ cacheRead: 0.007,
9945
10552
  cacheWrite: 0,
9946
10553
  },
9947
10554
  contextWindow: 1000000,
@@ -9957,9 +10564,9 @@ export const MODELS = {
9957
10564
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9958
10565
  input: ["text"],
9959
10566
  cost: {
9960
- input: 0.435,
9961
- output: 0.87,
9962
- cacheRead: 0.003625,
10567
+ input: 0.66,
10568
+ output: 1.98,
10569
+ cacheRead: 0.022,
9963
10570
  cacheWrite: 0,
9964
10571
  },
9965
10572
  contextWindow: 1000000,
@@ -9999,9 +10606,26 @@ export const MODELS = {
9999
10606
  contextWindow: 1000000,
10000
10607
  maxTokens: 131072,
10001
10608
  },
10609
+ "glm-5.3": {
10610
+ id: "glm-5.3",
10611
+ name: "GLM-5.3",
10612
+ api: "openai-completions",
10613
+ provider: "opencode-go",
10614
+ baseUrl: "https://opencode.ai/zen/go/v1",
10615
+ reasoning: true,
10616
+ input: ["text"],
10617
+ cost: {
10618
+ input: 1.4,
10619
+ output: 4.4,
10620
+ cacheRead: 0.26,
10621
+ cacheWrite: 0,
10622
+ },
10623
+ contextWindow: 1000000,
10624
+ maxTokens: 131072,
10625
+ },
10002
10626
  "gpt-5.6-luna": {
10003
10627
  id: "gpt-5.6-luna",
10004
- name: "GPT-5.6 Luna (2x usage)",
10628
+ name: "GPT-5.6 Luna",
10005
10629
  api: "openai-responses",
10006
10630
  provider: "opencode-go",
10007
10631
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -10010,10 +10634,10 @@ export const MODELS = {
10010
10634
  thinkingLevelMap: { "off": null, "xhigh": "xhigh", "max": "max" },
10011
10635
  input: ["text", "image"],
10012
10636
  cost: {
10013
- input: 0.1,
10014
- output: 0.6,
10015
- cacheRead: 0.01,
10016
- cacheWrite: 0.125,
10637
+ input: 0.2,
10638
+ output: 1.2,
10639
+ cacheRead: 0.02,
10640
+ cacheWrite: 0.25,
10017
10641
  },
10018
10642
  contextWindow: 1050000,
10019
10643
  maxTokens: 128000,
@@ -10038,16 +10662,16 @@ export const MODELS = {
10038
10662
  },
10039
10663
  "hy3": {
10040
10664
  id: "hy3",
10041
- name: "Hy3",
10665
+ name: "Hy3 (8x usage)",
10042
10666
  api: "openai-completions",
10043
10667
  provider: "opencode-go",
10044
10668
  baseUrl: "https://opencode.ai/zen/go/v1",
10045
10669
  reasoning: true,
10046
10670
  input: ["text"],
10047
10671
  cost: {
10048
- input: 0.14,
10049
- output: 0.58,
10050
- cacheRead: 0.035,
10672
+ input: 0.0175,
10673
+ output: 0.0725,
10674
+ cacheRead: 0.004375,
10051
10675
  cacheWrite: 0,
10052
10676
  },
10053
10677
  contextWindow: 256000,
@@ -10174,6 +10798,41 @@ export const MODELS = {
10174
10798
  contextWindow: 1000000,
10175
10799
  maxTokens: 131072,
10176
10800
  },
10801
+ "muse-spark-1.2-contributor": {
10802
+ id: "muse-spark-1.2-contributor",
10803
+ name: "Muse Spark 1.2 Contributor",
10804
+ api: "openai-responses",
10805
+ provider: "opencode-go",
10806
+ baseUrl: "https://opencode.ai/zen/go/v1",
10807
+ compat: { "sessionAffinityFormat": "openai-nosession" },
10808
+ reasoning: true,
10809
+ input: ["text", "image"],
10810
+ cost: {
10811
+ input: 0.1,
10812
+ output: 0.2,
10813
+ cacheRead: 0.002,
10814
+ cacheWrite: 0,
10815
+ },
10816
+ contextWindow: 1048576,
10817
+ maxTokens: 131072,
10818
+ },
10819
+ "ox-alpha-free": {
10820
+ id: "ox-alpha-free",
10821
+ name: "Ox Alpha Free (Unlimited)",
10822
+ api: "openai-completions",
10823
+ provider: "opencode-go",
10824
+ baseUrl: "https://opencode.ai/zen/go/v1",
10825
+ reasoning: true,
10826
+ input: ["text", "image"],
10827
+ cost: {
10828
+ input: 0,
10829
+ output: 0,
10830
+ cacheRead: 0,
10831
+ cacheWrite: 0,
10832
+ },
10833
+ contextWindow: 1000000,
10834
+ maxTokens: 131072,
10835
+ },
10177
10836
  "qwen3.6-plus": {
10178
10837
  id: "qwen3.6-plus",
10179
10838
  name: "Qwen3.6 Plus",
@@ -10195,9 +10854,9 @@ export const MODELS = {
10195
10854
  "qwen3.7-max": {
10196
10855
  id: "qwen3.7-max",
10197
10856
  name: "Qwen3.7 Max",
10198
- api: "anthropic-messages",
10857
+ api: "openai-completions",
10199
10858
  provider: "opencode-go",
10200
- baseUrl: "https://opencode.ai/zen/go",
10859
+ baseUrl: "https://opencode.ai/zen/go/v1",
10201
10860
  reasoning: true,
10202
10861
  input: ["text"],
10203
10862
  cost: {
@@ -10212,9 +10871,9 @@ export const MODELS = {
10212
10871
  "qwen3.7-plus": {
10213
10872
  id: "qwen3.7-plus",
10214
10873
  name: "Qwen3.7 Plus",
10215
- api: "anthropic-messages",
10874
+ api: "openai-completions",
10216
10875
  provider: "opencode-go",
10217
- baseUrl: "https://opencode.ai/zen/go",
10876
+ baseUrl: "https://opencode.ai/zen/go/v1",
10218
10877
  reasoning: true,
10219
10878
  input: ["text", "image"],
10220
10879
  cost: {
@@ -10229,9 +10888,9 @@ export const MODELS = {
10229
10888
  "qwen3.8-max": {
10230
10889
  id: "qwen3.8-max",
10231
10890
  name: "Qwen3.8 Max",
10232
- api: "anthropic-messages",
10891
+ api: "openai-completions",
10233
10892
  provider: "opencode-go",
10234
- baseUrl: "https://opencode.ai/zen/go",
10893
+ baseUrl: "https://opencode.ai/zen/go/v1",
10235
10894
  reasoning: true,
10236
10895
  input: ["text", "image"],
10237
10896
  cost: {
@@ -10245,23 +10904,6 @@ export const MODELS = {
10245
10904
  },
10246
10905
  },
10247
10906
  "openrouter": {
10248
- "ai21/jamba-large-1.7": {
10249
- id: "ai21/jamba-large-1.7",
10250
- name: "AI21: Jamba Large 1.7",
10251
- api: "openai-completions",
10252
- provider: "openrouter",
10253
- baseUrl: "https://openrouter.ai/api/v1",
10254
- reasoning: false,
10255
- input: ["text"],
10256
- cost: {
10257
- input: 2,
10258
- output: 8,
10259
- cacheRead: 0,
10260
- cacheWrite: 0,
10261
- },
10262
- contextWindow: 256000,
10263
- maxTokens: 4096,
10264
- },
10265
10907
  "aion-labs/aion-2.0": {
10266
10908
  id: "aion-labs/aion-2.0",
10267
10909
  name: "AionLabs: Aion-2.0",
@@ -11149,13 +11791,13 @@ export const MODELS = {
11149
11791
  reasoning: false,
11150
11792
  input: ["text"],
11151
11793
  cost: {
11152
- input: 0.27,
11153
- output: 1.12,
11154
- cacheRead: 0.135,
11794
+ input: 0.25,
11795
+ output: 1,
11796
+ cacheRead: 0,
11155
11797
  cacheWrite: 0,
11156
11798
  },
11157
11799
  contextWindow: 163840,
11158
- maxTokens: 65536,
11800
+ maxTokens: 163840,
11159
11801
  },
11160
11802
  "deepseek/deepseek-chat-v3.1": {
11161
11803
  id: "deepseek/deepseek-chat-v3.1",
@@ -11166,13 +11808,13 @@ export const MODELS = {
11166
11808
  reasoning: true,
11167
11809
  input: ["text"],
11168
11810
  cost: {
11169
- input: 0.25,
11170
- output: 0.95,
11171
- cacheRead: 0.13,
11811
+ input: 0.55,
11812
+ output: 1.6500000000000001,
11813
+ cacheRead: 0.55,
11172
11814
  cacheWrite: 0,
11173
11815
  },
11174
11816
  contextWindow: 163840,
11175
- maxTokens: 32768,
11817
+ maxTokens: 161000,
11176
11818
  },
11177
11819
  "deepseek/deepseek-r1": {
11178
11820
  id: "deepseek/deepseek-r1",
@@ -11218,8 +11860,8 @@ export const MODELS = {
11218
11860
  input: ["text"],
11219
11861
  cost: {
11220
11862
  input: 0.27,
11221
- output: 0.95,
11222
- cacheRead: 0.13,
11863
+ output: 1,
11864
+ cacheRead: 0.135,
11223
11865
  cacheWrite: 0,
11224
11866
  },
11225
11867
  contextWindow: 163840,
@@ -11234,13 +11876,13 @@ export const MODELS = {
11234
11876
  reasoning: true,
11235
11877
  input: ["text"],
11236
11878
  cost: {
11237
- input: 0.26899999999999996,
11238
- output: 0.39999999999999997,
11239
- cacheRead: 0.13449999999999998,
11879
+ input: 0.26,
11880
+ output: 0.38,
11881
+ cacheRead: 0.13,
11240
11882
  cacheWrite: 0,
11241
11883
  },
11242
11884
  contextWindow: 163840,
11243
- maxTokens: 65536,
11885
+ maxTokens: 163840,
11244
11886
  },
11245
11887
  "deepseek/deepseek-v3.2-exp": {
11246
11888
  id: "deepseek/deepseek-v3.2-exp",
@@ -11270,13 +11912,13 @@ export const MODELS = {
11270
11912
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11271
11913
  input: ["text"],
11272
11914
  cost: {
11273
- input: 0.14,
11274
- output: 0.28,
11275
- cacheRead: 0.028,
11915
+ input: 0.05866,
11916
+ output: 0.11732,
11917
+ cacheRead: 0.011732000000000001,
11276
11918
  cacheWrite: 0,
11277
11919
  },
11278
11920
  contextWindow: 1048576,
11279
- maxTokens: 393216,
11921
+ maxTokens: 384000,
11280
11922
  },
11281
11923
  "deepseek/deepseek-v4-flash-0731": {
11282
11924
  id: "deepseek/deepseek-v4-flash-0731",
@@ -11289,17 +11931,36 @@ export const MODELS = {
11289
11931
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11290
11932
  input: ["text"],
11291
11933
  cost: {
11292
- input: 0.14,
11293
- output: 0.28,
11294
- cacheRead: 0.028,
11934
+ input: 0.08,
11935
+ output: 0.18,
11936
+ cacheRead: 0.016,
11937
+ cacheWrite: 0,
11938
+ },
11939
+ contextWindow: 1310720,
11940
+ maxTokens: 384000,
11941
+ },
11942
+ "deepseek/deepseek-v4-flash-vision-exp": {
11943
+ id: "deepseek/deepseek-v4-flash-vision-exp",
11944
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp",
11945
+ api: "openai-completions",
11946
+ provider: "openrouter",
11947
+ baseUrl: "https://openrouter.ai/api/v1",
11948
+ compat: { "requiresReasoningContentOnAssistantMessages": true },
11949
+ reasoning: true,
11950
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11951
+ input: ["text", "image"],
11952
+ cost: {
11953
+ input: 0.22,
11954
+ output: 0.66,
11955
+ cacheRead: 0.007,
11295
11956
  cacheWrite: 0,
11296
11957
  },
11297
11958
  contextWindow: 1048576,
11298
- maxTokens: 393216,
11959
+ maxTokens: 384000,
11299
11960
  },
11300
11961
  "deepseek/deepseek-v4-pro": {
11301
11962
  id: "deepseek/deepseek-v4-pro",
11302
- name: "DeepSeek: DeepSeek V4 Pro",
11963
+ name: "DeepSeek: DeepSeek V4 Pro 0423",
11303
11964
  api: "openai-completions",
11304
11965
  provider: "openrouter",
11305
11966
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11308,13 +11969,13 @@ export const MODELS = {
11308
11969
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11309
11970
  input: ["text"],
11310
11971
  cost: {
11311
- input: 1.1680000000000001,
11312
- output: 2.3360000000000003,
11313
- cacheRead: 0.09855000000000001,
11972
+ input: 0.413772,
11973
+ output: 0.827544,
11974
+ cacheRead: 0.034481000000000005,
11314
11975
  cacheWrite: 0,
11315
11976
  },
11316
11977
  contextWindow: 1048576,
11317
- maxTokens: 393216,
11978
+ maxTokens: 384000,
11318
11979
  },
11319
11980
  "deepseek/deepseek-v4-pro-0813": {
11320
11981
  id: "deepseek/deepseek-v4-pro-0813",
@@ -11327,13 +11988,30 @@ export const MODELS = {
11327
11988
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11328
11989
  input: ["text"],
11329
11990
  cost: {
11330
- input: 0.435,
11331
- output: 0.87,
11332
- cacheRead: 0.003625,
11991
+ input: 1.122,
11992
+ output: 3.366,
11993
+ cacheRead: 0.037399999999999996,
11333
11994
  cacheWrite: 0,
11334
11995
  },
11335
11996
  contextWindow: 1048576,
11336
- maxTokens: 384000,
11997
+ maxTokens: 4096,
11998
+ },
11999
+ "dots-studio/dots-3-note-preview:free": {
12000
+ id: "dots-studio/dots-3-note-preview:free",
12001
+ name: "Dots Studio: Dots3-Note Preview (free)",
12002
+ api: "openai-completions",
12003
+ provider: "openrouter",
12004
+ baseUrl: "https://openrouter.ai/api/v1",
12005
+ reasoning: true,
12006
+ input: ["text", "image"],
12007
+ cost: {
12008
+ input: 0,
12009
+ output: 0,
12010
+ cacheRead: 0,
12011
+ cacheWrite: 0,
12012
+ },
12013
+ contextWindow: 512000,
12014
+ maxTokens: 512000,
11337
12015
  },
11338
12016
  "google/gemini-2.5-flash": {
11339
12017
  id: "google/gemini-2.5-flash",
@@ -11803,13 +12481,13 @@ export const MODELS = {
11803
12481
  reasoning: true,
11804
12482
  input: ["text", "image"],
11805
12483
  cost: {
11806
- input: 0.12,
11807
- output: 0.39999999999999997,
11808
- cacheRead: 0.049999999999999996,
12484
+ input: 0.07,
12485
+ output: 0.33999999999999997,
12486
+ cacheRead: 0,
11809
12487
  cacheWrite: 0,
11810
12488
  },
11811
12489
  contextWindow: 262144,
11812
- maxTokens: 262144,
12490
+ maxTokens: 16384,
11813
12491
  },
11814
12492
  "google/gemma-4-26b-a4b-it:free": {
11815
12493
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12030,7 +12708,7 @@ export const MODELS = {
12030
12708
  cacheRead: 0,
12031
12709
  cacheWrite: 0,
12032
12710
  },
12033
- contextWindow: 128000,
12711
+ contextWindow: 65536,
12034
12712
  maxTokens: 8192,
12035
12713
  },
12036
12714
  "meituan/longcat-2.0": {
@@ -12111,12 +12789,12 @@ export const MODELS = {
12111
12789
  input: ["text", "image"],
12112
12790
  cost: {
12113
12791
  input: 0.19999999999999998,
12114
- output: 0.696,
12792
+ output: 0.7999999999999999,
12115
12793
  cacheRead: 0,
12116
12794
  cacheWrite: 0,
12117
12795
  },
12118
12796
  contextWindow: 1048576,
12119
- maxTokens: 4096,
12797
+ maxTokens: 16384,
12120
12798
  },
12121
12799
  "meta-llama/llama-4-scout": {
12122
12800
  id: "meta-llama/llama-4-scout",
@@ -12186,6 +12864,23 @@ export const MODELS = {
12186
12864
  contextWindow: 1048576,
12187
12865
  maxTokens: 4096,
12188
12866
  },
12867
+ "meta/muse-spark-1.2-contributor": {
12868
+ id: "meta/muse-spark-1.2-contributor",
12869
+ name: "Meta: Muse Spark 1.2 Contributor",
12870
+ api: "openai-completions",
12871
+ provider: "openrouter",
12872
+ baseUrl: "https://openrouter.ai/api/v1",
12873
+ reasoning: true,
12874
+ input: ["text", "image"],
12875
+ cost: {
12876
+ input: 0.09999999999999999,
12877
+ output: 0.19999999999999998,
12878
+ cacheRead: 0.002,
12879
+ cacheWrite: 0,
12880
+ },
12881
+ contextWindow: 1048576,
12882
+ maxTokens: 4096,
12883
+ },
12189
12884
  "minimax/minimax-m1": {
12190
12885
  id: "minimax/minimax-m1",
12191
12886
  name: "MiniMax: MiniMax M1",
@@ -12246,13 +12941,13 @@ export const MODELS = {
12246
12941
  reasoning: true,
12247
12942
  input: ["text"],
12248
12943
  cost: {
12249
- input: 0.22,
12250
- output: 0.8999999999999999,
12251
- cacheRead: 0.049999999999999996,
12944
+ input: 0.27,
12945
+ output: 1.08,
12946
+ cacheRead: 0.027,
12252
12947
  cacheWrite: 0,
12253
12948
  },
12254
12949
  contextWindow: 204800,
12255
- maxTokens: 196608,
12950
+ maxTokens: 128000,
12256
12951
  },
12257
12952
  "minimax/minimax-m2.7": {
12258
12953
  id: "minimax/minimax-m2.7",
@@ -12297,9 +12992,9 @@ export const MODELS = {
12297
12992
  reasoning: true,
12298
12993
  input: ["text", "image"],
12299
12994
  cost: {
12300
- input: 0.15,
12301
- output: 0.6,
12302
- cacheRead: 0.03,
12995
+ input: 0.3,
12996
+ output: 1.2,
12997
+ cacheRead: 0.06,
12303
12998
  cacheWrite: 0,
12304
12999
  },
12305
13000
  contextWindow: 524288,
@@ -12535,12 +13230,12 @@ export const MODELS = {
12535
13230
  reasoning: false,
12536
13231
  input: ["text", "image"],
12537
13232
  cost: {
12538
- input: 0.09375,
12539
- output: 0.25,
13233
+ input: 0.075,
13234
+ output: 0.19999999999999998,
12540
13235
  cacheRead: 0,
12541
13236
  cacheWrite: 0,
12542
13237
  },
12543
- contextWindow: 256000,
13238
+ contextWindow: 131072,
12544
13239
  maxTokens: 16384,
12545
13240
  },
12546
13241
  "mistralai/mixtral-8x22b-instruct": {
@@ -12655,9 +13350,9 @@ export const MODELS = {
12655
13350
  reasoning: true,
12656
13351
  input: ["text", "image"],
12657
13352
  cost: {
12658
- input: 0.95,
12659
- output: 4,
12660
- cacheRead: 0.16,
13353
+ input: 0.5415,
13354
+ output: 2.2800000000000002,
13355
+ cacheRead: 0.09119999999999999,
12661
13356
  cacheWrite: 0,
12662
13357
  },
12663
13358
  contextWindow: 262144,
@@ -12674,7 +13369,7 @@ export const MODELS = {
12674
13369
  cost: {
12675
13370
  input: 0.67,
12676
13371
  output: 3.4,
12677
- cacheRead: 0.15,
13372
+ cacheRead: 0.16999999999999998,
12678
13373
  cacheWrite: 0,
12679
13374
  },
12680
13375
  contextWindow: 262144,
@@ -12689,9 +13384,9 @@ export const MODELS = {
12689
13384
  reasoning: true,
12690
13385
  input: ["text", "image"],
12691
13386
  cost: {
12692
- input: 0.475,
12693
- output: 2,
12694
- cacheRead: 0.095,
13387
+ input: 0.95,
13388
+ output: 4,
13389
+ cacheRead: 0.19,
12695
13390
  cacheWrite: 0,
12696
13391
  },
12697
13392
  contextWindow: 262144,
@@ -12759,11 +13454,11 @@ export const MODELS = {
12759
13454
  cost: {
12760
13455
  input: 0.049999999999999996,
12761
13456
  output: 0.19999999999999998,
12762
- cacheRead: 0.024999999999999998,
13457
+ cacheRead: 0.03,
12763
13458
  cacheWrite: 0,
12764
13459
  },
12765
13460
  contextWindow: 262144,
12766
- maxTokens: 228000,
13461
+ maxTokens: 262144,
12767
13462
  },
12768
13463
  "nvidia/nemotron-3-nano-30b-a3b:free": {
12769
13464
  id: "nvidia/nemotron-3-nano-30b-a3b:free",
@@ -12859,9 +13554,9 @@ export const MODELS = {
12859
13554
  reasoning: true,
12860
13555
  input: ["text"],
12861
13556
  cost: {
12862
- input: 0.3,
12863
- output: 1.7999999999999998,
12864
- cacheRead: 0.09999999999999999,
13557
+ input: 0.6,
13558
+ output: 3.5999999999999996,
13559
+ cacheRead: 0.19999999999999998,
12865
13560
  cacheWrite: 0,
12866
13561
  },
12867
13562
  contextWindow: 512288,
@@ -12893,13 +13588,13 @@ export const MODELS = {
12893
13588
  reasoning: true,
12894
13589
  input: ["text"],
12895
13590
  cost: {
12896
- input: 0.09999999999999999,
12897
- output: 0.25,
12898
- cacheRead: 0.049999999999999996,
13591
+ input: 0.08,
13592
+ output: 0.19999999999999998,
13593
+ cacheRead: 0.04,
12899
13594
  cacheWrite: 0,
12900
13595
  },
12901
- contextWindow: 1000000,
12902
- maxTokens: 262144,
13596
+ contextWindow: 262144,
13597
+ maxTokens: 131072,
12903
13598
  },
12904
13599
  "nvidia/nemotron-3.5-lightning:free": {
12905
13600
  id: "nvidia/nemotron-3.5-lightning:free",
@@ -13899,10 +14594,10 @@ export const MODELS = {
13899
14594
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
13900
14595
  input: ["text", "image"],
13901
14596
  cost: {
13902
- input: 0.09999999999999999,
13903
- output: 0.6,
13904
- cacheRead: 0.01,
13905
- cacheWrite: 0.125,
14597
+ input: 0.19999999999999998,
14598
+ output: 1.2,
14599
+ cacheRead: 0.02,
14600
+ cacheWrite: 0.25,
13906
14601
  },
13907
14602
  contextWindow: 1050000,
13908
14603
  maxTokens: 128000,
@@ -13917,10 +14612,10 @@ export const MODELS = {
13917
14612
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
13918
14613
  input: ["text", "image"],
13919
14614
  cost: {
13920
- input: 0.09999999999999999,
13921
- output: 0.6,
13922
- cacheRead: 0.01,
13923
- cacheWrite: 0.125,
14615
+ input: 0.19999999999999998,
14616
+ output: 1.2,
14617
+ cacheRead: 0.02,
14618
+ cacheWrite: 0.25,
13924
14619
  },
13925
14620
  contextWindow: 1050000,
13926
14621
  maxTokens: 128000,
@@ -13971,10 +14666,10 @@ export const MODELS = {
13971
14666
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
13972
14667
  input: ["text", "image"],
13973
14668
  cost: {
13974
- input: 5,
13975
- output: 30,
13976
- cacheRead: 0.5,
13977
- cacheWrite: 6.25,
14669
+ input: 2,
14670
+ output: 10,
14671
+ cacheRead: 0.19999999999999998,
14672
+ cacheWrite: 2.5,
13978
14673
  },
13979
14674
  contextWindow: 1050000,
13980
14675
  maxTokens: 128000,
@@ -13989,10 +14684,10 @@ export const MODELS = {
13989
14684
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
13990
14685
  input: ["text", "image"],
13991
14686
  cost: {
13992
- input: 5,
13993
- output: 30,
13994
- cacheRead: 0.5,
13995
- cacheWrite: 6.25,
14687
+ input: 2,
14688
+ output: 10,
14689
+ cacheRead: 0.19999999999999998,
14690
+ cacheWrite: 2.5,
13996
14691
  },
13997
14692
  contextWindow: 1050000,
13998
14693
  maxTokens: 128000,
@@ -14007,10 +14702,10 @@ export const MODELS = {
14007
14702
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
14008
14703
  input: ["text", "image"],
14009
14704
  cost: {
14010
- input: 2.5,
14011
- output: 15,
14012
- cacheRead: 0.25,
14013
- cacheWrite: 0,
14705
+ input: 1,
14706
+ output: 5,
14707
+ cacheRead: 0.09999999999999999,
14708
+ cacheWrite: 1.25,
14014
14709
  },
14015
14710
  contextWindow: 1050000,
14016
14711
  maxTokens: 128000,
@@ -14025,10 +14720,10 @@ export const MODELS = {
14025
14720
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
14026
14721
  input: ["text", "image"],
14027
14722
  cost: {
14028
- input: 2.5,
14029
- output: 15,
14030
- cacheRead: 0.25,
14031
- cacheWrite: 0,
14723
+ input: 1,
14724
+ output: 5,
14725
+ cacheRead: 0.09999999999999999,
14726
+ cacheWrite: 1.25,
14032
14727
  },
14033
14728
  contextWindow: 1050000,
14034
14729
  maxTokens: 128000,
@@ -14043,10 +14738,10 @@ export const MODELS = {
14043
14738
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
14044
14739
  input: ["text", "image"],
14045
14740
  cost: {
14046
- input: 1,
14047
- output: 6,
14048
- cacheRead: 0.09999999999999999,
14049
- cacheWrite: 1.25,
14741
+ input: 2,
14742
+ output: 12,
14743
+ cacheRead: 0.19999999999999998,
14744
+ cacheWrite: 2.5,
14050
14745
  },
14051
14746
  contextWindow: 1050000,
14052
14747
  maxTokens: 128000,
@@ -14061,10 +14756,10 @@ export const MODELS = {
14061
14756
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
14062
14757
  input: ["text", "image"],
14063
14758
  cost: {
14064
- input: 1,
14065
- output: 6,
14066
- cacheRead: 0.09999999999999999,
14067
- cacheWrite: 1.25,
14759
+ input: 2,
14760
+ output: 12,
14761
+ cacheRead: 0.19999999999999998,
14762
+ cacheWrite: 2.5,
14068
14763
  },
14069
14764
  contextWindow: 1050000,
14070
14765
  maxTokens: 128000,
@@ -14182,9 +14877,9 @@ export const MODELS = {
14182
14877
  reasoning: true,
14183
14878
  input: ["text"],
14184
14879
  cost: {
14185
- input: 0.03,
14880
+ input: 0.037,
14186
14881
  output: 0.16999999999999998,
14187
- cacheRead: 0.03,
14882
+ cacheRead: 0,
14188
14883
  cacheWrite: 0,
14189
14884
  },
14190
14885
  contextWindow: 131072,
@@ -14207,23 +14902,6 @@ export const MODELS = {
14207
14902
  contextWindow: 131072,
14208
14903
  maxTokens: 131072,
14209
14904
  },
14210
- "openai/gpt-oss-20b:free": {
14211
- id: "openai/gpt-oss-20b:free",
14212
- name: "OpenAI: gpt-oss-20b (free)",
14213
- api: "openai-completions",
14214
- provider: "openrouter",
14215
- baseUrl: "https://openrouter.ai/api/v1",
14216
- reasoning: true,
14217
- input: ["text"],
14218
- cost: {
14219
- input: 0,
14220
- output: 0,
14221
- cacheRead: 0,
14222
- cacheWrite: 0,
14223
- },
14224
- contextWindow: 131072,
14225
- maxTokens: 32768,
14226
- },
14227
14905
  "openai/gpt-oss-safeguard-20b": {
14228
14906
  id: "openai/gpt-oss-safeguard-20b",
14229
14907
  name: "OpenAI: gpt-oss-safeguard-20b",
@@ -14692,10 +15370,10 @@ export const MODELS = {
14692
15370
  reasoning: true,
14693
15371
  input: ["text"],
14694
15372
  cost: {
14695
- input: 0.39999999999999997,
14696
- output: 1.2,
15373
+ input: 0.26,
15374
+ output: 0.78,
14697
15375
  cacheRead: 0,
14698
- cacheWrite: 0.5,
15376
+ cacheWrite: 0,
14699
15377
  },
14700
15378
  contextWindow: 1000000,
14701
15379
  maxTokens: 32768,
@@ -14777,13 +15455,13 @@ export const MODELS = {
14777
15455
  reasoning: true,
14778
15456
  input: ["text"],
14779
15457
  cost: {
14780
- input: 0.12,
14781
- output: 0.5,
15458
+ input: 0.13,
15459
+ output: 0.52,
14782
15460
  cacheRead: 0,
14783
15461
  cacheWrite: 0,
14784
15462
  },
14785
15463
  contextWindow: 131072,
14786
- maxTokens: 16384,
15464
+ maxTokens: 8192,
14787
15465
  },
14788
15466
  "qwen/qwen3-30b-a3b-instruct-2507": {
14789
15467
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -15015,9 +15693,9 @@ export const MODELS = {
15015
15693
  reasoning: false,
15016
15694
  input: ["text", "image"],
15017
15695
  cost: {
15018
- input: 0.26,
15019
- output: 1.04,
15020
- cacheRead: 0,
15696
+ input: 0.21,
15697
+ output: 1.9,
15698
+ cacheRead: 0.09999999999999999,
15021
15699
  cacheWrite: 0,
15022
15700
  },
15023
15701
  contextWindow: 262144,
@@ -15134,13 +15812,13 @@ export const MODELS = {
15134
15812
  reasoning: true,
15135
15813
  input: ["text", "image"],
15136
15814
  cost: {
15137
- input: 0.29,
15138
- output: 2.4,
15815
+ input: 0.26,
15816
+ output: 2.08,
15139
15817
  cacheRead: 0,
15140
15818
  cacheWrite: 0,
15141
15819
  },
15142
15820
  contextWindow: 262144,
15143
- maxTokens: 81920,
15821
+ maxTokens: 65536,
15144
15822
  },
15145
15823
  "qwen/qwen3.5-27b": {
15146
15824
  id: "qwen/qwen3.5-27b",
@@ -15185,13 +15863,13 @@ export const MODELS = {
15185
15863
  reasoning: true,
15186
15864
  input: ["text", "image"],
15187
15865
  cost: {
15188
- input: 0.5,
15189
- output: 3.5999999999999996,
15190
- cacheRead: 0.3,
15866
+ input: 0.39,
15867
+ output: 2.34,
15868
+ cacheRead: 0,
15191
15869
  cacheWrite: 0,
15192
15870
  },
15193
15871
  contextWindow: 262144,
15194
- maxTokens: 262144,
15872
+ maxTokens: 65536,
15195
15873
  },
15196
15874
  "qwen/qwen3.5-9b": {
15197
15875
  id: "qwen/qwen3.5-9b",
@@ -15287,7 +15965,7 @@ export const MODELS = {
15287
15965
  reasoning: true,
15288
15966
  input: ["text", "image"],
15289
15967
  cost: {
15290
- input: 0.15,
15968
+ input: 0.14,
15291
15969
  output: 1,
15292
15970
  cacheRead: 0.049999999999999996,
15293
15971
  cacheWrite: 0,
@@ -15411,8 +16089,25 @@ export const MODELS = {
15411
16089
  cacheRead: 0.25,
15412
16090
  cacheWrite: 0,
15413
16091
  },
15414
- contextWindow: 1010000,
15415
- maxTokens: 262144,
16092
+ contextWindow: 1048576,
16093
+ maxTokens: 131072,
16094
+ },
16095
+ "qwen/qwen3.8-27b": {
16096
+ id: "qwen/qwen3.8-27b",
16097
+ name: "Qwen: Qwen3.8 27B",
16098
+ api: "openai-completions",
16099
+ provider: "openrouter",
16100
+ baseUrl: "https://openrouter.ai/api/v1",
16101
+ reasoning: true,
16102
+ input: ["text", "image"],
16103
+ cost: {
16104
+ input: 0.39999999999999997,
16105
+ output: 3,
16106
+ cacheRead: 0.049999999999999996,
16107
+ cacheWrite: 0,
16108
+ },
16109
+ contextWindow: 1000000,
16110
+ maxTokens: 131072,
15416
16111
  },
15417
16112
  "qwen/qwen3.8-max": {
15418
16113
  id: "qwen/qwen3.8-max",
@@ -15516,6 +16211,24 @@ export const MODELS = {
15516
16211
  contextWindow: 131072,
15517
16212
  maxTokens: 16384,
15518
16213
  },
16214
+ "stealth/ox-alpha": {
16215
+ id: "stealth/ox-alpha",
16216
+ name: "Ox Alpha",
16217
+ api: "openai-completions",
16218
+ provider: "openrouter",
16219
+ baseUrl: "https://openrouter.ai/api/v1",
16220
+ reasoning: true,
16221
+ thinkingLevelMap: { max: "max" },
16222
+ input: ["text", "image"],
16223
+ cost: {
16224
+ input: 0,
16225
+ output: 0,
16226
+ cacheRead: 0,
16227
+ cacheWrite: 0,
16228
+ },
16229
+ contextWindow: 1048576,
16230
+ maxTokens: 131072,
16231
+ },
15519
16232
  "stepfun/step-3.5-flash": {
15520
16233
  id: "stepfun/step-3.5-flash",
15521
16234
  name: "StepFun: Step 3.5 Flash",
@@ -15576,9 +16289,9 @@ export const MODELS = {
15576
16289
  reasoning: true,
15577
16290
  input: ["text"],
15578
16291
  cost: {
15579
- input: 0.063,
15580
- output: 0.21,
15581
- cacheRead: 0.020999999999999998,
16292
+ input: 0.18,
16293
+ output: 0.6,
16294
+ cacheRead: 0.06,
15582
16295
  cacheWrite: 0,
15583
16296
  },
15584
16297
  contextWindow: 262144,
@@ -15632,7 +16345,24 @@ export const MODELS = {
15632
16345
  cacheRead: 0.09999999999999999,
15633
16346
  cacheWrite: 0,
15634
16347
  },
15635
- contextWindow: 524288,
16348
+ contextWindow: 1048576,
16349
+ maxTokens: 262144,
16350
+ },
16351
+ "thinkingmachines/inkling-small:free": {
16352
+ id: "thinkingmachines/inkling-small:free",
16353
+ name: "Thinking Machines: Inkling Small (free)",
16354
+ api: "openai-completions",
16355
+ provider: "openrouter",
16356
+ baseUrl: "https://openrouter.ai/api/v1",
16357
+ reasoning: true,
16358
+ input: ["text", "image"],
16359
+ cost: {
16360
+ input: 0,
16361
+ output: 0,
16362
+ cacheRead: 0,
16363
+ cacheWrite: 0,
16364
+ },
16365
+ contextWindow: 262144,
15636
16366
  maxTokens: 262144,
15637
16367
  },
15638
16368
  "thinkingmachines/inkling:batch": {
@@ -15644,14 +16374,31 @@ export const MODELS = {
15644
16374
  reasoning: true,
15645
16375
  input: ["text", "image"],
15646
16376
  cost: {
15647
- input: 0.5,
15648
- output: 2.025,
15649
- cacheRead: 0.08499999999999999,
16377
+ input: 1,
16378
+ output: 4.05,
16379
+ cacheRead: 0.16999999999999998,
15650
16380
  cacheWrite: 0,
15651
16381
  },
15652
16382
  contextWindow: 524288,
15653
16383
  maxTokens: 4096,
15654
16384
  },
16385
+ "thinkingmachines/inkling:free": {
16386
+ id: "thinkingmachines/inkling:free",
16387
+ name: "Thinking Machines: Inkling (free)",
16388
+ api: "openai-completions",
16389
+ provider: "openrouter",
16390
+ baseUrl: "https://openrouter.ai/api/v1",
16391
+ reasoning: true,
16392
+ input: ["text", "image"],
16393
+ cost: {
16394
+ input: 0,
16395
+ output: 0,
16396
+ cacheRead: 0,
16397
+ cacheWrite: 0,
16398
+ },
16399
+ contextWindow: 262144,
16400
+ maxTokens: 262144,
16401
+ },
15655
16402
  "upstage/solar-pro-3": {
15656
16403
  id: "upstage/solar-pro-3",
15657
16404
  name: "Upstage: Solar Pro 3",
@@ -15939,7 +16686,7 @@ export const MODELS = {
15939
16686
  cacheWrite: 0,
15940
16687
  },
15941
16688
  contextWindow: 204800,
15942
- maxTokens: 131072,
16689
+ maxTokens: 128000,
15943
16690
  },
15944
16691
  "z-ai/glm-5-turbo": {
15945
16692
  id: "z-ai/glm-5-turbo",
@@ -15967,13 +16714,13 @@ export const MODELS = {
15967
16714
  reasoning: true,
15968
16715
  input: ["text"],
15969
16716
  cost: {
15970
- input: 1.4,
15971
- output: 4.4,
15972
- cacheRead: 0.26,
16717
+ input: 0.966,
16718
+ output: 3.036,
16719
+ cacheRead: 0.1794,
15973
16720
  cacheWrite: 0,
15974
16721
  },
15975
16722
  contextWindow: 204800,
15976
- maxTokens: 131072,
16723
+ maxTokens: 128000,
15977
16724
  },
15978
16725
  "z-ai/glm-5.2": {
15979
16726
  id: "z-ai/glm-5.2",
@@ -15984,13 +16731,13 @@ export const MODELS = {
15984
16731
  reasoning: true,
15985
16732
  input: ["text"],
15986
16733
  cost: {
15987
- input: 0.63,
15988
- output: 1.9800000000000002,
15989
- cacheRead: 0.0945,
16734
+ input: 0.966,
16735
+ output: 3.036,
16736
+ cacheRead: 0.1932,
15990
16737
  cacheWrite: 0,
15991
16738
  },
15992
16739
  contextWindow: 1048576,
15993
- maxTokens: 4096,
16740
+ maxTokens: 131072,
15994
16741
  },
15995
16742
  "z-ai/glm-5.2:batch": {
15996
16743
  id: "z-ai/glm-5.2:batch",
@@ -16001,14 +16748,48 @@ export const MODELS = {
16001
16748
  reasoning: true,
16002
16749
  input: ["text"],
16003
16750
  cost: {
16004
- input: 0.7,
16005
- output: 2.2,
16006
- cacheRead: 0.13,
16751
+ input: 1.4,
16752
+ output: 4.4,
16753
+ cacheRead: 0.26,
16007
16754
  cacheWrite: 0,
16008
16755
  },
16009
- contextWindow: 512000,
16756
+ contextWindow: 1048575,
16010
16757
  maxTokens: 4096,
16011
16758
  },
16759
+ "z-ai/glm-5.2:free": {
16760
+ id: "z-ai/glm-5.2:free",
16761
+ name: "Z.ai: GLM 5.2 (free)",
16762
+ api: "openai-completions",
16763
+ provider: "openrouter",
16764
+ baseUrl: "https://openrouter.ai/api/v1",
16765
+ reasoning: true,
16766
+ input: ["text"],
16767
+ cost: {
16768
+ input: 0,
16769
+ output: 0,
16770
+ cacheRead: 0,
16771
+ cacheWrite: 0,
16772
+ },
16773
+ contextWindow: 256000,
16774
+ maxTokens: 256000,
16775
+ },
16776
+ "z-ai/glm-5.3": {
16777
+ id: "z-ai/glm-5.3",
16778
+ name: "Z.ai: GLM 5.3",
16779
+ api: "openai-completions",
16780
+ provider: "openrouter",
16781
+ baseUrl: "https://openrouter.ai/api/v1",
16782
+ reasoning: true,
16783
+ input: ["text"],
16784
+ cost: {
16785
+ input: 1.4,
16786
+ output: 4.4,
16787
+ cacheRead: 0.26,
16788
+ cacheWrite: 0,
16789
+ },
16790
+ contextWindow: 1048576,
16791
+ maxTokens: 131072,
16792
+ },
16012
16793
  "z-ai/glm-5v-turbo": {
16013
16794
  id: "z-ai/glm-5v-turbo",
16014
16795
  name: "Z.ai: GLM 5V Turbo",
@@ -16109,13 +16890,13 @@ export const MODELS = {
16109
16890
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
16110
16891
  input: ["text"],
16111
16892
  cost: {
16112
- input: 0.079996,
16113
- output: 0.252,
16114
- cacheRead: 0.0252,
16893
+ input: 0.065,
16894
+ output: 0.18,
16895
+ cacheRead: 0.02,
16115
16896
  cacheWrite: 0,
16116
16897
  },
16117
- contextWindow: 1048576,
16118
- maxTokens: 4096,
16898
+ contextWindow: 1310720,
16899
+ maxTokens: 1048576,
16119
16900
  },
16120
16901
  "~google/gemini-flash-latest": {
16121
16902
  id: "~google/gemini-flash-latest",
@@ -16160,13 +16941,13 @@ export const MODELS = {
16160
16941
  reasoning: true,
16161
16942
  input: ["text", "image"],
16162
16943
  cost: {
16163
- input: 2.8,
16164
- output: 14,
16944
+ input: 2.6,
16945
+ output: 13,
16165
16946
  cacheRead: 0.29,
16166
16947
  cacheWrite: 0,
16167
16948
  },
16168
16949
  contextWindow: 1048576,
16169
- maxTokens: 1048576,
16950
+ maxTokens: 974842,
16170
16951
  },
16171
16952
  "~openai/gpt-latest": {
16172
16953
  id: "~openai/gpt-latest",
@@ -16177,10 +16958,10 @@ export const MODELS = {
16177
16958
  reasoning: true,
16178
16959
  input: ["text", "image"],
16179
16960
  cost: {
16180
- input: 5,
16181
- output: 30,
16182
- cacheRead: 0.5,
16183
- cacheWrite: 6.25,
16961
+ input: 2,
16962
+ output: 10,
16963
+ cacheRead: 0.19999999999999998,
16964
+ cacheWrite: 2.5,
16184
16965
  },
16185
16966
  contextWindow: 1050000,
16186
16967
  maxTokens: 128000,
@@ -16219,6 +17000,23 @@ export const MODELS = {
16219
17000
  contextWindow: 500000,
16220
17001
  maxTokens: 4096,
16221
17002
  },
17003
+ "~z-ai/glm-latest": {
17004
+ id: "~z-ai/glm-latest",
17005
+ name: "Z.ai: GLM Latest",
17006
+ api: "openai-completions",
17007
+ provider: "openrouter",
17008
+ baseUrl: "https://openrouter.ai/api/v1",
17009
+ reasoning: true,
17010
+ input: ["text"],
17011
+ cost: {
17012
+ input: 1.4,
17013
+ output: 4.4,
17014
+ cacheRead: 0.26,
17015
+ cacheWrite: 0,
17016
+ },
17017
+ contextWindow: 1048576,
17018
+ maxTokens: 131072,
17019
+ },
16222
17020
  },
16223
17021
  "qwen-token-plan": {
16224
17022
  "MiniMax-M2.5": {
@@ -16314,6 +17112,25 @@ export const MODELS = {
16314
17112
  contextWindow: 1000000,
16315
17113
  maxTokens: 384000,
16316
17114
  },
17115
+ "deepseek-v4-pro-0813": {
17116
+ id: "deepseek-v4-pro-0813",
17117
+ name: "DeepSeek V4 Pro 0813",
17118
+ api: "openai-completions",
17119
+ provider: "qwen-token-plan",
17120
+ baseUrl: "https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1",
17121
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
17122
+ reasoning: true,
17123
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
17124
+ input: ["text"],
17125
+ cost: {
17126
+ input: 0,
17127
+ output: 0,
17128
+ cacheRead: 0,
17129
+ cacheWrite: 0,
17130
+ },
17131
+ contextWindow: 1000000,
17132
+ maxTokens: 384000,
17133
+ },
16317
17134
  "glm-5": {
16318
17135
  id: "glm-5",
16319
17136
  name: "GLM-5",
@@ -16625,6 +17442,25 @@ export const MODELS = {
16625
17442
  contextWindow: 1000000,
16626
17443
  maxTokens: 384000,
16627
17444
  },
17445
+ "deepseek-v4-pro-0813": {
17446
+ id: "deepseek-v4-pro-0813",
17447
+ name: "DeepSeek V4 Pro 0813",
17448
+ api: "openai-completions",
17449
+ provider: "qwen-token-plan-cn",
17450
+ baseUrl: "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
17451
+ compat: { "thinkingFormat": "qwen", "supportsDeveloperRole": false, "supportsStore": false },
17452
+ reasoning: true,
17453
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
17454
+ input: ["text"],
17455
+ cost: {
17456
+ input: 0,
17457
+ output: 0,
17458
+ cacheRead: 0,
17459
+ cacheWrite: 0,
17460
+ },
17461
+ contextWindow: 1000000,
17462
+ maxTokens: 384000,
17463
+ },
16628
17464
  "glm-5": {
16629
17465
  id: "glm-5",
16630
17466
  name: "GLM-5",
@@ -16993,6 +17829,25 @@ export const MODELS = {
16993
17829
  contextWindow: 512000,
16994
17830
  maxTokens: 384000,
16995
17831
  },
17832
+ "deepseek-ai/DeepSeek-V4-Pro-0813": {
17833
+ id: "deepseek-ai/DeepSeek-V4-Pro-0813",
17834
+ name: "DeepSeek V4 Pro 0813",
17835
+ api: "openai-completions",
17836
+ provider: "together",
17837
+ baseUrl: "https://api.together.ai/v1",
17838
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
17839
+ reasoning: true,
17840
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
17841
+ input: ["text"],
17842
+ cost: {
17843
+ input: 1.32,
17844
+ output: 3.96,
17845
+ cacheRead: 0.13,
17846
+ cacheWrite: 0,
17847
+ },
17848
+ contextWindow: 1048576,
17849
+ maxTokens: 384000,
17850
+ },
16996
17851
  "google/gemma-4-31B-it": {
16997
17852
  id: "google/gemma-4-31B-it",
16998
17853
  name: "Gemma 4 31B Instruct",
@@ -17616,16 +18471,33 @@ export const MODELS = {
17616
18471
  provider: "vercel-ai-gateway",
17617
18472
  baseUrl: "https://ai-gateway.vercel.sh",
17618
18473
  reasoning: true,
17619
- input: ["text"],
18474
+ input: ["text", "image"],
17620
18475
  cost: {
17621
18476
  input: 2,
17622
18477
  output: 6,
17623
- cacheRead: 0.25,
18478
+ cacheRead: 0.19999999999999998,
17624
18479
  cacheWrite: 0,
17625
18480
  },
17626
18481
  contextWindow: 262144,
17627
18482
  maxTokens: 131072,
17628
18483
  },
18484
+ "alibaba/qwen3.8-27b": {
18485
+ id: "alibaba/qwen3.8-27b",
18486
+ name: "Qwen3.8 27B",
18487
+ api: "anthropic-messages",
18488
+ provider: "vercel-ai-gateway",
18489
+ baseUrl: "https://ai-gateway.vercel.sh",
18490
+ reasoning: true,
18491
+ input: ["text", "image"],
18492
+ cost: {
18493
+ input: 0.55,
18494
+ output: 3.3000000000000003,
18495
+ cacheRead: 0.11,
18496
+ cacheWrite: 0,
18497
+ },
18498
+ contextWindow: 1000000,
18499
+ maxTokens: 131072,
18500
+ },
17629
18501
  "alibaba/qwen3.8-max": {
17630
18502
  id: "alibaba/qwen3.8-max",
17631
18503
  name: "Qwen 3.8 Max",
@@ -18180,9 +19052,9 @@ export const MODELS = {
18180
19052
  reasoning: true,
18181
19053
  input: ["text"],
18182
19054
  cost: {
18183
- input: 0.19999999999999998,
18184
- output: 0.39999999999999997,
18185
- cacheRead: 0.04,
19055
+ input: 0.13,
19056
+ output: 0.26,
19057
+ cacheRead: 0.028,
18186
19058
  cacheWrite: 0,
18187
19059
  },
18188
19060
  contextWindow: 1000000,
@@ -18197,9 +19069,26 @@ export const MODELS = {
18197
19069
  reasoning: true,
18198
19070
  input: ["text"],
18199
19071
  cost: {
18200
- input: 0.19999999999999998,
18201
- output: 0.39999999999999997,
18202
- cacheRead: 0.04,
19072
+ input: 0.07600000000000001,
19073
+ output: 0.153,
19074
+ cacheRead: 0.014,
19075
+ cacheWrite: 0,
19076
+ },
19077
+ contextWindow: 1000000,
19078
+ maxTokens: 384000,
19079
+ },
19080
+ "deepseek/deepseek-v4-flash-vision-exp": {
19081
+ id: "deepseek/deepseek-v4-flash-vision-exp",
19082
+ name: "DeepSeek V4 Flash Vision Exp",
19083
+ api: "anthropic-messages",
19084
+ provider: "vercel-ai-gateway",
19085
+ baseUrl: "https://ai-gateway.vercel.sh",
19086
+ reasoning: true,
19087
+ input: ["text", "image"],
19088
+ cost: {
19089
+ input: 0.22,
19090
+ output: 0.66,
19091
+ cacheRead: 0.007,
18203
19092
  cacheWrite: 0,
18204
19093
  },
18205
19094
  contextWindow: 1000000,
@@ -18231,9 +19120,9 @@ export const MODELS = {
18231
19120
  reasoning: true,
18232
19121
  input: ["text"],
18233
19122
  cost: {
18234
- input: 0.435,
18235
- output: 0.87,
18236
- cacheRead: 0.0036,
19123
+ input: 1.32,
19124
+ output: 3.9600000000000004,
19125
+ cacheRead: 0.13199999999999998,
18237
19126
  cacheWrite: 0,
18238
19127
  },
18239
19128
  contextWindow: 1000000,
@@ -18384,9 +19273,9 @@ export const MODELS = {
18384
19273
  reasoning: true,
18385
19274
  input: ["text", "image"],
18386
19275
  cost: {
18387
- input: 1.5,
18388
- output: 7.5,
18389
- cacheRead: 0.15,
19276
+ input: 0.75,
19277
+ output: 3.75,
19278
+ cacheRead: 0.075,
18390
19279
  cacheWrite: 0,
18391
19280
  },
18392
19281
  contextWindow: 1000000,
@@ -19273,25 +20162,59 @@ export const MODELS = {
19273
20162
  cacheRead: 0,
19274
20163
  cacheWrite: 0,
19275
20164
  },
19276
- contextWindow: 256000,
19277
- maxTokens: 32000,
20165
+ contextWindow: 256000,
20166
+ maxTokens: 32000,
20167
+ },
20168
+ "nvidia/nemotron-3-ultra-550b-a55b": {
20169
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
20170
+ name: "Nemotron 3 Ultra",
20171
+ api: "anthropic-messages",
20172
+ provider: "vercel-ai-gateway",
20173
+ baseUrl: "https://ai-gateway.vercel.sh",
20174
+ reasoning: true,
20175
+ input: ["text"],
20176
+ cost: {
20177
+ input: 0.6,
20178
+ output: 2.4,
20179
+ cacheRead: 0.12,
20180
+ cacheWrite: 0,
20181
+ },
20182
+ contextWindow: 1000000,
20183
+ maxTokens: 65000,
20184
+ },
20185
+ "nvidia/nemotron-3.5-lightning": {
20186
+ id: "nvidia/nemotron-3.5-lightning",
20187
+ name: "Nemotron 3.5 Lightning 30B",
20188
+ api: "anthropic-messages",
20189
+ provider: "vercel-ai-gateway",
20190
+ baseUrl: "https://ai-gateway.vercel.sh",
20191
+ reasoning: true,
20192
+ input: ["text"],
20193
+ cost: {
20194
+ input: 0,
20195
+ output: 0,
20196
+ cacheRead: 0,
20197
+ cacheWrite: 0,
20198
+ },
20199
+ contextWindow: 1000000,
20200
+ maxTokens: 32768,
19278
20201
  },
19279
- "nvidia/nemotron-3-ultra-550b-a55b": {
19280
- id: "nvidia/nemotron-3-ultra-550b-a55b",
19281
- name: "Nemotron 3 Ultra",
20202
+ "nvidia/nemotron-3.5-lightning-free": {
20203
+ id: "nvidia/nemotron-3.5-lightning-free",
20204
+ name: "Nemotron 3.5 Lightning 30B (Free)",
19282
20205
  api: "anthropic-messages",
19283
20206
  provider: "vercel-ai-gateway",
19284
20207
  baseUrl: "https://ai-gateway.vercel.sh",
19285
20208
  reasoning: true,
19286
20209
  input: ["text"],
19287
20210
  cost: {
19288
- input: 0.6,
19289
- output: 2.4,
19290
- cacheRead: 0.12,
20211
+ input: 0,
20212
+ output: 0,
20213
+ cacheRead: 0,
19291
20214
  cacheWrite: 0,
19292
20215
  },
19293
20216
  contextWindow: 1000000,
19294
- maxTokens: 65000,
20217
+ maxTokens: 32768,
19295
20218
  },
19296
20219
  "nvidia/nemotron-nano-12b-v2-vl": {
19297
20220
  id: "nvidia/nemotron-nano-12b-v2-vl",
@@ -19378,6 +20301,23 @@ export const MODELS = {
19378
20301
  contextWindow: 1047576,
19379
20302
  maxTokens: 32768,
19380
20303
  },
20304
+ "openai/gpt-4.1-fast": {
20305
+ id: "openai/gpt-4.1-fast",
20306
+ name: "GPT-4.1 (Fast)",
20307
+ api: "anthropic-messages",
20308
+ provider: "vercel-ai-gateway",
20309
+ baseUrl: "https://ai-gateway.vercel.sh",
20310
+ reasoning: false,
20311
+ input: ["text", "image"],
20312
+ cost: {
20313
+ input: 3.5,
20314
+ output: 14,
20315
+ cacheRead: 0.875,
20316
+ cacheWrite: 0,
20317
+ },
20318
+ contextWindow: 1047576,
20319
+ maxTokens: 32768,
20320
+ },
19381
20321
  "openai/gpt-4.1-mini": {
19382
20322
  id: "openai/gpt-4.1-mini",
19383
20323
  name: "GPT-4.1 mini",
@@ -19395,6 +20335,23 @@ export const MODELS = {
19395
20335
  contextWindow: 1047576,
19396
20336
  maxTokens: 32768,
19397
20337
  },
20338
+ "openai/gpt-4.1-mini-fast": {
20339
+ id: "openai/gpt-4.1-mini-fast",
20340
+ name: "GPT-4.1 mini (Fast)",
20341
+ api: "anthropic-messages",
20342
+ provider: "vercel-ai-gateway",
20343
+ baseUrl: "https://ai-gateway.vercel.sh",
20344
+ reasoning: false,
20345
+ input: ["text", "image"],
20346
+ cost: {
20347
+ input: 0.7,
20348
+ output: 2.8,
20349
+ cacheRead: 0.175,
20350
+ cacheWrite: 0,
20351
+ },
20352
+ contextWindow: 1047576,
20353
+ maxTokens: 32768,
20354
+ },
19398
20355
  "openai/gpt-4.1-nano": {
19399
20356
  id: "openai/gpt-4.1-nano",
19400
20357
  name: "GPT-4.1 nano",
@@ -19412,6 +20369,23 @@ export const MODELS = {
19412
20369
  contextWindow: 1047576,
19413
20370
  maxTokens: 32768,
19414
20371
  },
20372
+ "openai/gpt-4.1-nano-fast": {
20373
+ id: "openai/gpt-4.1-nano-fast",
20374
+ name: "GPT-4.1 nano (Fast)",
20375
+ api: "anthropic-messages",
20376
+ provider: "vercel-ai-gateway",
20377
+ baseUrl: "https://ai-gateway.vercel.sh",
20378
+ reasoning: false,
20379
+ input: ["text", "image"],
20380
+ cost: {
20381
+ input: 0.19999999999999998,
20382
+ output: 0.7999999999999999,
20383
+ cacheRead: 0.049999999999999996,
20384
+ cacheWrite: 0,
20385
+ },
20386
+ contextWindow: 1047576,
20387
+ maxTokens: 32768,
20388
+ },
19415
20389
  "openai/gpt-4o": {
19416
20390
  id: "openai/gpt-4o",
19417
20391
  name: "GPT-4o",
@@ -19429,6 +20403,23 @@ export const MODELS = {
19429
20403
  contextWindow: 128000,
19430
20404
  maxTokens: 16384,
19431
20405
  },
20406
+ "openai/gpt-4o-fast": {
20407
+ id: "openai/gpt-4o-fast",
20408
+ name: "GPT-4o (Fast)",
20409
+ api: "anthropic-messages",
20410
+ provider: "vercel-ai-gateway",
20411
+ baseUrl: "https://ai-gateway.vercel.sh",
20412
+ reasoning: false,
20413
+ input: ["text", "image"],
20414
+ cost: {
20415
+ input: 4.25,
20416
+ output: 17,
20417
+ cacheRead: 2.125,
20418
+ cacheWrite: 0,
20419
+ },
20420
+ contextWindow: 128000,
20421
+ maxTokens: 16384,
20422
+ },
19432
20423
  "openai/gpt-4o-mini": {
19433
20424
  id: "openai/gpt-4o-mini",
19434
20425
  name: "GPT-4o mini",
@@ -19446,6 +20437,23 @@ export const MODELS = {
19446
20437
  contextWindow: 128000,
19447
20438
  maxTokens: 16384,
19448
20439
  },
20440
+ "openai/gpt-4o-mini-fast": {
20441
+ id: "openai/gpt-4o-mini-fast",
20442
+ name: "GPT-4o mini (Fast)",
20443
+ api: "anthropic-messages",
20444
+ provider: "vercel-ai-gateway",
20445
+ baseUrl: "https://ai-gateway.vercel.sh",
20446
+ reasoning: false,
20447
+ input: ["text", "image"],
20448
+ cost: {
20449
+ input: 0.25,
20450
+ output: 1,
20451
+ cacheRead: 0.125,
20452
+ cacheWrite: 0,
20453
+ },
20454
+ contextWindow: 128000,
20455
+ maxTokens: 16384,
20456
+ },
19449
20457
  "openai/gpt-5": {
19450
20458
  id: "openai/gpt-5",
19451
20459
  name: "GPT-5",
@@ -19480,6 +20488,23 @@ export const MODELS = {
19480
20488
  contextWindow: 400000,
19481
20489
  maxTokens: 128000,
19482
20490
  },
20491
+ "openai/gpt-5-fast": {
20492
+ id: "openai/gpt-5-fast",
20493
+ name: "GPT-5 (Fast)",
20494
+ api: "anthropic-messages",
20495
+ provider: "vercel-ai-gateway",
20496
+ baseUrl: "https://ai-gateway.vercel.sh",
20497
+ reasoning: true,
20498
+ input: ["text", "image"],
20499
+ cost: {
20500
+ input: 2.5,
20501
+ output: 20,
20502
+ cacheRead: 0.25,
20503
+ cacheWrite: 0,
20504
+ },
20505
+ contextWindow: 400000,
20506
+ maxTokens: 128000,
20507
+ },
19483
20508
  "openai/gpt-5-mini": {
19484
20509
  id: "openai/gpt-5-mini",
19485
20510
  name: "GPT-5 mini",
@@ -19497,6 +20522,23 @@ export const MODELS = {
19497
20522
  contextWindow: 400000,
19498
20523
  maxTokens: 128000,
19499
20524
  },
20525
+ "openai/gpt-5-mini-fast": {
20526
+ id: "openai/gpt-5-mini-fast",
20527
+ name: "GPT-5 mini (Fast)",
20528
+ api: "anthropic-messages",
20529
+ provider: "vercel-ai-gateway",
20530
+ baseUrl: "https://ai-gateway.vercel.sh",
20531
+ reasoning: true,
20532
+ input: ["text", "image"],
20533
+ cost: {
20534
+ input: 0.44999999999999996,
20535
+ output: 3.5999999999999996,
20536
+ cacheRead: 0.045,
20537
+ cacheWrite: 0,
20538
+ },
20539
+ contextWindow: 400000,
20540
+ maxTokens: 128000,
20541
+ },
19500
20542
  "openai/gpt-5-nano": {
19501
20543
  id: "openai/gpt-5-nano",
19502
20544
  name: "GPT-5 nano",
@@ -19599,6 +20641,23 @@ export const MODELS = {
19599
20641
  contextWindow: 400000,
19600
20642
  maxTokens: 128000,
19601
20643
  },
20644
+ "openai/gpt-5.1-thinking-fast": {
20645
+ id: "openai/gpt-5.1-thinking-fast",
20646
+ name: "GPT 5.1 Thinking (Fast)",
20647
+ api: "anthropic-messages",
20648
+ provider: "vercel-ai-gateway",
20649
+ baseUrl: "https://ai-gateway.vercel.sh",
20650
+ reasoning: true,
20651
+ input: ["text", "image"],
20652
+ cost: {
20653
+ input: 2.5,
20654
+ output: 20,
20655
+ cacheRead: 0.25,
20656
+ cacheWrite: 0,
20657
+ },
20658
+ contextWindow: 400000,
20659
+ maxTokens: 128000,
20660
+ },
19602
20661
  "openai/gpt-5.2": {
19603
20662
  id: "openai/gpt-5.2",
19604
20663
  name: "GPT 5.2",
@@ -19635,6 +20694,24 @@ export const MODELS = {
19635
20694
  contextWindow: 400000,
19636
20695
  maxTokens: 128000,
19637
20696
  },
20697
+ "openai/gpt-5.2-fast": {
20698
+ id: "openai/gpt-5.2-fast",
20699
+ name: "GPT 5.2 (Fast)",
20700
+ api: "anthropic-messages",
20701
+ provider: "vercel-ai-gateway",
20702
+ baseUrl: "https://ai-gateway.vercel.sh",
20703
+ reasoning: true,
20704
+ thinkingLevelMap: { "xhigh": "xhigh" },
20705
+ input: ["text", "image"],
20706
+ cost: {
20707
+ input: 3.5,
20708
+ output: 28,
20709
+ cacheRead: 0.35,
20710
+ cacheWrite: 0,
20711
+ },
20712
+ contextWindow: 400000,
20713
+ maxTokens: 128000,
20714
+ },
19638
20715
  "openai/gpt-5.2-pro": {
19639
20716
  id: "openai/gpt-5.2-pro",
19640
20717
  name: "GPT 5.2 ",
@@ -19671,6 +20748,24 @@ export const MODELS = {
19671
20748
  contextWindow: 400000,
19672
20749
  maxTokens: 128000,
19673
20750
  },
20751
+ "openai/gpt-5.3-codex-fast": {
20752
+ id: "openai/gpt-5.3-codex-fast",
20753
+ name: "GPT 5.3 Codex (Fast)",
20754
+ api: "anthropic-messages",
20755
+ provider: "vercel-ai-gateway",
20756
+ baseUrl: "https://ai-gateway.vercel.sh",
20757
+ reasoning: true,
20758
+ thinkingLevelMap: { "xhigh": "xhigh" },
20759
+ input: ["text", "image"],
20760
+ cost: {
20761
+ input: 3.5,
20762
+ output: 28,
20763
+ cacheRead: 0.35,
20764
+ cacheWrite: 0,
20765
+ },
20766
+ contextWindow: 400000,
20767
+ maxTokens: 128000,
20768
+ },
19674
20769
  "openai/gpt-5.4": {
19675
20770
  id: "openai/gpt-5.4",
19676
20771
  name: "GPT 5.4",
@@ -19689,6 +20784,24 @@ export const MODELS = {
19689
20784
  contextWindow: 1050000,
19690
20785
  maxTokens: 128000,
19691
20786
  },
20787
+ "openai/gpt-5.4-fast": {
20788
+ id: "openai/gpt-5.4-fast",
20789
+ name: "GPT 5.4 (Fast)",
20790
+ api: "anthropic-messages",
20791
+ provider: "vercel-ai-gateway",
20792
+ baseUrl: "https://ai-gateway.vercel.sh",
20793
+ reasoning: true,
20794
+ thinkingLevelMap: { "xhigh": "xhigh" },
20795
+ input: ["text", "image"],
20796
+ cost: {
20797
+ input: 5,
20798
+ output: 30,
20799
+ cacheRead: 0.5,
20800
+ cacheWrite: 0,
20801
+ },
20802
+ contextWindow: 1050000,
20803
+ maxTokens: 128000,
20804
+ },
19692
20805
  "openai/gpt-5.4-mini": {
19693
20806
  id: "openai/gpt-5.4-mini",
19694
20807
  name: "GPT 5.4 Mini",
@@ -19707,6 +20820,24 @@ export const MODELS = {
19707
20820
  contextWindow: 400000,
19708
20821
  maxTokens: 128000,
19709
20822
  },
20823
+ "openai/gpt-5.4-mini-fast": {
20824
+ id: "openai/gpt-5.4-mini-fast",
20825
+ name: "GPT 5.4 Mini (Fast)",
20826
+ api: "anthropic-messages",
20827
+ provider: "vercel-ai-gateway",
20828
+ baseUrl: "https://ai-gateway.vercel.sh",
20829
+ reasoning: true,
20830
+ thinkingLevelMap: { "xhigh": "xhigh" },
20831
+ input: ["text", "image"],
20832
+ cost: {
20833
+ input: 1.5,
20834
+ output: 9,
20835
+ cacheRead: 0.15,
20836
+ cacheWrite: 0,
20837
+ },
20838
+ contextWindow: 400000,
20839
+ maxTokens: 128000,
20840
+ },
19710
20841
  "openai/gpt-5.4-nano": {
19711
20842
  id: "openai/gpt-5.4-nano",
19712
20843
  name: "GPT 5.4 Nano",
@@ -19761,6 +20892,24 @@ export const MODELS = {
19761
20892
  contextWindow: 1000000,
19762
20893
  maxTokens: 128000,
19763
20894
  },
20895
+ "openai/gpt-5.5-fast": {
20896
+ id: "openai/gpt-5.5-fast",
20897
+ name: "GPT 5.5 (Fast)",
20898
+ api: "anthropic-messages",
20899
+ provider: "vercel-ai-gateway",
20900
+ baseUrl: "https://ai-gateway.vercel.sh",
20901
+ reasoning: true,
20902
+ thinkingLevelMap: { "xhigh": "xhigh" },
20903
+ input: ["text", "image"],
20904
+ cost: {
20905
+ input: 12.5,
20906
+ output: 75,
20907
+ cacheRead: 1.25,
20908
+ cacheWrite: 0,
20909
+ },
20910
+ contextWindow: 1000000,
20911
+ maxTokens: 128000,
20912
+ },
19764
20913
  "openai/gpt-5.5-pro": {
19765
20914
  id: "openai/gpt-5.5-pro",
19766
20915
  name: "GPT 5.5 Pro",
@@ -19797,6 +20946,24 @@ export const MODELS = {
19797
20946
  contextWindow: 1050000,
19798
20947
  maxTokens: 128000,
19799
20948
  },
20949
+ "openai/gpt-5.6-luna-fast": {
20950
+ id: "openai/gpt-5.6-luna-fast",
20951
+ name: "GPT 5.6 Luna (Fast)",
20952
+ api: "anthropic-messages",
20953
+ provider: "vercel-ai-gateway",
20954
+ baseUrl: "https://ai-gateway.vercel.sh",
20955
+ reasoning: true,
20956
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
20957
+ input: ["text", "image"],
20958
+ cost: {
20959
+ input: 0.39999999999999997,
20960
+ output: 2.4,
20961
+ cacheRead: 0.04,
20962
+ cacheWrite: 0.25,
20963
+ },
20964
+ contextWindow: 1050000,
20965
+ maxTokens: 128000,
20966
+ },
19800
20967
  "openai/gpt-5.6-sol": {
19801
20968
  id: "openai/gpt-5.6-sol",
19802
20969
  name: "GPT 5.6 Sol",
@@ -19807,10 +20974,28 @@ export const MODELS = {
19807
20974
  thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
19808
20975
  input: ["text", "image"],
19809
20976
  cost: {
19810
- input: 5,
19811
- output: 30,
19812
- cacheRead: 0.5,
19813
- cacheWrite: 6.25,
20977
+ input: 2,
20978
+ output: 10,
20979
+ cacheRead: 0.19999999999999998,
20980
+ cacheWrite: 2.5,
20981
+ },
20982
+ contextWindow: 1050000,
20983
+ maxTokens: 128000,
20984
+ },
20985
+ "openai/gpt-5.6-sol-fast": {
20986
+ id: "openai/gpt-5.6-sol-fast",
20987
+ name: "GPT 5.6 Sol (Fast)",
20988
+ api: "anthropic-messages",
20989
+ provider: "vercel-ai-gateway",
20990
+ baseUrl: "https://ai-gateway.vercel.sh",
20991
+ reasoning: true,
20992
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
20993
+ input: ["text", "image"],
20994
+ cost: {
20995
+ input: 4,
20996
+ output: 20,
20997
+ cacheRead: 0.39999999999999997,
20998
+ cacheWrite: 2.5,
19814
20999
  },
19815
21000
  contextWindow: 1050000,
19816
21001
  maxTokens: 128000,
@@ -19833,6 +21018,24 @@ export const MODELS = {
19833
21018
  contextWindow: 1050000,
19834
21019
  maxTokens: 128000,
19835
21020
  },
21021
+ "openai/gpt-5.6-terra-fast": {
21022
+ id: "openai/gpt-5.6-terra-fast",
21023
+ name: "GPT 5.6 Terra (Fast)",
21024
+ api: "anthropic-messages",
21025
+ provider: "vercel-ai-gateway",
21026
+ baseUrl: "https://ai-gateway.vercel.sh",
21027
+ reasoning: true,
21028
+ thinkingLevelMap: { "xhigh": "xhigh", "max": "max" },
21029
+ input: ["text", "image"],
21030
+ cost: {
21031
+ input: 4,
21032
+ output: 24,
21033
+ cacheRead: 0.39999999999999997,
21034
+ cacheWrite: 2.5,
21035
+ },
21036
+ contextWindow: 1050000,
21037
+ maxTokens: 128000,
21038
+ },
19836
21039
  "openai/gpt-oss-120b": {
19837
21040
  id: "openai/gpt-oss-120b",
19838
21041
  name: "GPT OSS 120B",
@@ -19918,18 +21121,35 @@ export const MODELS = {
19918
21121
  contextWindow: 200000,
19919
21122
  maxTokens: 100000,
19920
21123
  },
19921
- "openai/o3-deep-research": {
19922
- id: "openai/o3-deep-research",
19923
- name: "o3-deep-research",
21124
+ "openai/o3-deep-research": {
21125
+ id: "openai/o3-deep-research",
21126
+ name: "o3-deep-research",
21127
+ api: "anthropic-messages",
21128
+ provider: "vercel-ai-gateway",
21129
+ baseUrl: "https://ai-gateway.vercel.sh",
21130
+ reasoning: true,
21131
+ input: ["text", "image"],
21132
+ cost: {
21133
+ input: 10,
21134
+ output: 40,
21135
+ cacheRead: 2.5,
21136
+ cacheWrite: 0,
21137
+ },
21138
+ contextWindow: 200000,
21139
+ maxTokens: 100000,
21140
+ },
21141
+ "openai/o3-fast": {
21142
+ id: "openai/o3-fast",
21143
+ name: "o3 (Fast)",
19924
21144
  api: "anthropic-messages",
19925
21145
  provider: "vercel-ai-gateway",
19926
21146
  baseUrl: "https://ai-gateway.vercel.sh",
19927
21147
  reasoning: true,
19928
21148
  input: ["text", "image"],
19929
21149
  cost: {
19930
- input: 10,
19931
- output: 40,
19932
- cacheRead: 2.5,
21150
+ input: 3.5,
21151
+ output: 14,
21152
+ cacheRead: 0.875,
19933
21153
  cacheWrite: 0,
19934
21154
  },
19935
21155
  contextWindow: 200000,
@@ -19986,6 +21206,23 @@ export const MODELS = {
19986
21206
  contextWindow: 200000,
19987
21207
  maxTokens: 100000,
19988
21208
  },
21209
+ "openai/o4-mini-fast": {
21210
+ id: "openai/o4-mini-fast",
21211
+ name: "o4-mini (Fast)",
21212
+ api: "anthropic-messages",
21213
+ provider: "vercel-ai-gateway",
21214
+ baseUrl: "https://ai-gateway.vercel.sh",
21215
+ reasoning: true,
21216
+ input: ["text", "image"],
21217
+ cost: {
21218
+ input: 2,
21219
+ output: 8,
21220
+ cacheRead: 0.5,
21221
+ cacheWrite: 0,
21222
+ },
21223
+ contextWindow: 200000,
21224
+ maxTokens: 100000,
21225
+ },
19989
21226
  "poolside/laguna-s-2.1": {
19990
21227
  id: "poolside/laguna-s-2.1",
19991
21228
  name: "Laguna S 2.1",
@@ -20054,93 +21291,8 @@ export const MODELS = {
20054
21291
  contextWindow: 256000,
20055
21292
  maxTokens: 256000,
20056
21293
  },
20057
- "stepfun/step-3.5-flash": {
20058
- id: "stepfun/step-3.5-flash",
20059
- name: "StepFun 3.5 Flash",
20060
- api: "anthropic-messages",
20061
- provider: "vercel-ai-gateway",
20062
- baseUrl: "https://ai-gateway.vercel.sh",
20063
- reasoning: true,
20064
- input: ["text", "image"],
20065
- cost: {
20066
- input: 0.09,
20067
- output: 0.3,
20068
- cacheRead: 0.02,
20069
- cacheWrite: 0,
20070
- },
20071
- contextWindow: 262114,
20072
- maxTokens: 262114,
20073
- },
20074
- "stepfun/step-3.7-flash": {
20075
- id: "stepfun/step-3.7-flash",
20076
- name: "Step 3.7 Flash",
20077
- api: "anthropic-messages",
20078
- provider: "vercel-ai-gateway",
20079
- baseUrl: "https://ai-gateway.vercel.sh",
20080
- reasoning: true,
20081
- input: ["text", "image"],
20082
- cost: {
20083
- input: 0.19999999999999998,
20084
- output: 1.15,
20085
- cacheRead: 0.04,
20086
- cacheWrite: 0,
20087
- },
20088
- contextWindow: 256000,
20089
- maxTokens: 256000,
20090
- },
20091
- "tencent/hy3": {
20092
- id: "tencent/hy3",
20093
- name: "Hy3",
20094
- api: "anthropic-messages",
20095
- provider: "vercel-ai-gateway",
20096
- baseUrl: "https://ai-gateway.vercel.sh",
20097
- reasoning: true,
20098
- input: ["text"],
20099
- cost: {
20100
- input: 0.14,
20101
- output: 0.58,
20102
- cacheRead: 0.035,
20103
- cacheWrite: 0,
20104
- },
20105
- contextWindow: 262144,
20106
- maxTokens: 262144,
20107
- },
20108
- "thinkingmachines/inkling": {
20109
- id: "thinkingmachines/inkling",
20110
- name: "Inkling",
20111
- api: "anthropic-messages",
20112
- provider: "vercel-ai-gateway",
20113
- baseUrl: "https://ai-gateway.vercel.sh",
20114
- reasoning: true,
20115
- input: ["text", "image"],
20116
- cost: {
20117
- input: 1,
20118
- output: 4.05,
20119
- cacheRead: 0.16999999999999998,
20120
- cacheWrite: 0,
20121
- },
20122
- contextWindow: 256000,
20123
- maxTokens: 256000,
20124
- },
20125
- "thinkingmachines/inkling-small": {
20126
- id: "thinkingmachines/inkling-small",
20127
- name: "Inkling Small",
20128
- api: "anthropic-messages",
20129
- provider: "vercel-ai-gateway",
20130
- baseUrl: "https://ai-gateway.vercel.sh",
20131
- reasoning: true,
20132
- input: ["text", "image"],
20133
- cost: {
20134
- input: 0.5,
20135
- output: 1.2,
20136
- cacheRead: 0.09999999999999999,
20137
- cacheWrite: 0,
20138
- },
20139
- contextWindow: 1000000,
20140
- maxTokens: 1000000,
20141
- },
20142
- "xai/grok-4.1-fast-non-reasoning": {
20143
- id: "xai/grok-4.1-fast-non-reasoning",
21294
+ "spacexai/grok-4.1-fast-non-reasoning": {
21295
+ id: "spacexai/grok-4.1-fast-non-reasoning",
20144
21296
  name: "Grok 4.1 Fast Non-Reasoning",
20145
21297
  api: "anthropic-messages",
20146
21298
  provider: "vercel-ai-gateway",
@@ -20156,8 +21308,8 @@ export const MODELS = {
20156
21308
  contextWindow: 1000000,
20157
21309
  maxTokens: 1000000,
20158
21310
  },
20159
- "xai/grok-4.1-fast-reasoning": {
20160
- id: "xai/grok-4.1-fast-reasoning",
21311
+ "spacexai/grok-4.1-fast-reasoning": {
21312
+ id: "spacexai/grok-4.1-fast-reasoning",
20161
21313
  name: "Grok 4.1 Fast Reasoning",
20162
21314
  api: "anthropic-messages",
20163
21315
  provider: "vercel-ai-gateway",
@@ -20173,8 +21325,8 @@ export const MODELS = {
20173
21325
  contextWindow: 1000000,
20174
21326
  maxTokens: 1000000,
20175
21327
  },
20176
- "xai/grok-4.20-multi-agent": {
20177
- id: "xai/grok-4.20-multi-agent",
21328
+ "spacexai/grok-4.20-multi-agent": {
21329
+ id: "spacexai/grok-4.20-multi-agent",
20178
21330
  name: "Grok 4.20 Multi-Agent",
20179
21331
  api: "anthropic-messages",
20180
21332
  provider: "vercel-ai-gateway",
@@ -20190,8 +21342,8 @@ export const MODELS = {
20190
21342
  contextWindow: 2000000,
20191
21343
  maxTokens: 2000000,
20192
21344
  },
20193
- "xai/grok-4.20-multi-agent-beta": {
20194
- id: "xai/grok-4.20-multi-agent-beta",
21345
+ "spacexai/grok-4.20-multi-agent-beta": {
21346
+ id: "spacexai/grok-4.20-multi-agent-beta",
20195
21347
  name: "Grok 4.20 Multi Agent Beta",
20196
21348
  api: "anthropic-messages",
20197
21349
  provider: "vercel-ai-gateway",
@@ -20207,8 +21359,8 @@ export const MODELS = {
20207
21359
  contextWindow: 2000000,
20208
21360
  maxTokens: 2000000,
20209
21361
  },
20210
- "xai/grok-4.20-non-reasoning": {
20211
- id: "xai/grok-4.20-non-reasoning",
21362
+ "spacexai/grok-4.20-non-reasoning": {
21363
+ id: "spacexai/grok-4.20-non-reasoning",
20212
21364
  name: "Grok 4.20 Non-Reasoning",
20213
21365
  api: "anthropic-messages",
20214
21366
  provider: "vercel-ai-gateway",
@@ -20224,8 +21376,8 @@ export const MODELS = {
20224
21376
  contextWindow: 2000000,
20225
21377
  maxTokens: 2000000,
20226
21378
  },
20227
- "xai/grok-4.20-non-reasoning-beta": {
20228
- id: "xai/grok-4.20-non-reasoning-beta",
21379
+ "spacexai/grok-4.20-non-reasoning-beta": {
21380
+ id: "spacexai/grok-4.20-non-reasoning-beta",
20229
21381
  name: "Grok 4.20 Beta Non-Reasoning",
20230
21382
  api: "anthropic-messages",
20231
21383
  provider: "vercel-ai-gateway",
@@ -20241,8 +21393,8 @@ export const MODELS = {
20241
21393
  contextWindow: 2000000,
20242
21394
  maxTokens: 2000000,
20243
21395
  },
20244
- "xai/grok-4.20-reasoning": {
20245
- id: "xai/grok-4.20-reasoning",
21396
+ "spacexai/grok-4.20-reasoning": {
21397
+ id: "spacexai/grok-4.20-reasoning",
20246
21398
  name: "Grok 4.20 Reasoning",
20247
21399
  api: "anthropic-messages",
20248
21400
  provider: "vercel-ai-gateway",
@@ -20258,8 +21410,8 @@ export const MODELS = {
20258
21410
  contextWindow: 2000000,
20259
21411
  maxTokens: 2000000,
20260
21412
  },
20261
- "xai/grok-4.20-reasoning-beta": {
20262
- id: "xai/grok-4.20-reasoning-beta",
21413
+ "spacexai/grok-4.20-reasoning-beta": {
21414
+ id: "spacexai/grok-4.20-reasoning-beta",
20263
21415
  name: "Grok 4.20 Beta Reasoning",
20264
21416
  api: "anthropic-messages",
20265
21417
  provider: "vercel-ai-gateway",
@@ -20275,8 +21427,8 @@ export const MODELS = {
20275
21427
  contextWindow: 2000000,
20276
21428
  maxTokens: 2000000,
20277
21429
  },
20278
- "xai/grok-4.3": {
20279
- id: "xai/grok-4.3",
21430
+ "spacexai/grok-4.3": {
21431
+ id: "spacexai/grok-4.3",
20280
21432
  name: "Grok 4.3",
20281
21433
  api: "anthropic-messages",
20282
21434
  provider: "vercel-ai-gateway",
@@ -20292,8 +21444,8 @@ export const MODELS = {
20292
21444
  contextWindow: 1000000,
20293
21445
  maxTokens: 1000000,
20294
21446
  },
20295
- "xai/grok-4.5": {
20296
- id: "xai/grok-4.5",
21447
+ "spacexai/grok-4.5": {
21448
+ id: "spacexai/grok-4.5",
20297
21449
  name: "Grok 4.5",
20298
21450
  api: "anthropic-messages",
20299
21451
  provider: "vercel-ai-gateway",
@@ -20309,14 +21461,14 @@ export const MODELS = {
20309
21461
  contextWindow: 500000,
20310
21462
  maxTokens: 500000,
20311
21463
  },
20312
- "xai/grok-4.6": {
20313
- id: "xai/grok-4.6",
21464
+ "spacexai/grok-4.6": {
21465
+ id: "spacexai/grok-4.6",
20314
21466
  name: "Grok 4.6",
20315
21467
  api: "anthropic-messages",
20316
21468
  provider: "vercel-ai-gateway",
20317
21469
  baseUrl: "https://ai-gateway.vercel.sh",
20318
21470
  reasoning: true,
20319
- input: ["text"],
21471
+ input: ["text", "image"],
20320
21472
  cost: {
20321
21473
  input: 2,
20322
21474
  output: 6,
@@ -20326,8 +21478,8 @@ export const MODELS = {
20326
21478
  contextWindow: 500000,
20327
21479
  maxTokens: 500000,
20328
21480
  },
20329
- "xai/grok-build-0.1": {
20330
- id: "xai/grok-build-0.1",
21481
+ "spacexai/grok-build-0.1": {
21482
+ id: "spacexai/grok-build-0.1",
20331
21483
  name: "Grok Build 0.1",
20332
21484
  api: "anthropic-messages",
20333
21485
  provider: "vercel-ai-gateway",
@@ -20343,6 +21495,91 @@ export const MODELS = {
20343
21495
  contextWindow: 256000,
20344
21496
  maxTokens: 256000,
20345
21497
  },
21498
+ "stepfun/step-3.5-flash": {
21499
+ id: "stepfun/step-3.5-flash",
21500
+ name: "StepFun 3.5 Flash",
21501
+ api: "anthropic-messages",
21502
+ provider: "vercel-ai-gateway",
21503
+ baseUrl: "https://ai-gateway.vercel.sh",
21504
+ reasoning: true,
21505
+ input: ["text", "image"],
21506
+ cost: {
21507
+ input: 0.09,
21508
+ output: 0.3,
21509
+ cacheRead: 0.02,
21510
+ cacheWrite: 0,
21511
+ },
21512
+ contextWindow: 262114,
21513
+ maxTokens: 262114,
21514
+ },
21515
+ "stepfun/step-3.7-flash": {
21516
+ id: "stepfun/step-3.7-flash",
21517
+ name: "Step 3.7 Flash",
21518
+ api: "anthropic-messages",
21519
+ provider: "vercel-ai-gateway",
21520
+ baseUrl: "https://ai-gateway.vercel.sh",
21521
+ reasoning: true,
21522
+ input: ["text", "image"],
21523
+ cost: {
21524
+ input: 0.19999999999999998,
21525
+ output: 1.15,
21526
+ cacheRead: 0.04,
21527
+ cacheWrite: 0,
21528
+ },
21529
+ contextWindow: 256000,
21530
+ maxTokens: 256000,
21531
+ },
21532
+ "tencent/hy3": {
21533
+ id: "tencent/hy3",
21534
+ name: "Hy3",
21535
+ api: "anthropic-messages",
21536
+ provider: "vercel-ai-gateway",
21537
+ baseUrl: "https://ai-gateway.vercel.sh",
21538
+ reasoning: true,
21539
+ input: ["text"],
21540
+ cost: {
21541
+ input: 0.13199999999999998,
21542
+ output: 0.5279999999999999,
21543
+ cacheRead: 0.032999999999999995,
21544
+ cacheWrite: 0,
21545
+ },
21546
+ contextWindow: 256000,
21547
+ maxTokens: 128000,
21548
+ },
21549
+ "thinkingmachines/inkling": {
21550
+ id: "thinkingmachines/inkling",
21551
+ name: "Inkling",
21552
+ api: "anthropic-messages",
21553
+ provider: "vercel-ai-gateway",
21554
+ baseUrl: "https://ai-gateway.vercel.sh",
21555
+ reasoning: true,
21556
+ input: ["text", "image"],
21557
+ cost: {
21558
+ input: 1,
21559
+ output: 4.05,
21560
+ cacheRead: 0.16999999999999998,
21561
+ cacheWrite: 0,
21562
+ },
21563
+ contextWindow: 256000,
21564
+ maxTokens: 256000,
21565
+ },
21566
+ "thinkingmachines/inkling-small": {
21567
+ id: "thinkingmachines/inkling-small",
21568
+ name: "Inkling Small",
21569
+ api: "anthropic-messages",
21570
+ provider: "vercel-ai-gateway",
21571
+ baseUrl: "https://ai-gateway.vercel.sh",
21572
+ reasoning: true,
21573
+ input: ["text", "image"],
21574
+ cost: {
21575
+ input: 0.5,
21576
+ output: 1.2,
21577
+ cacheRead: 0.09999999999999999,
21578
+ cacheWrite: 0,
21579
+ },
21580
+ contextWindow: 1000000,
21581
+ maxTokens: 1000000,
21582
+ },
20346
21583
  "xiaomi/mimo-v2.5": {
20347
21584
  id: "xiaomi/mimo-v2.5",
20348
21585
  name: "MiMo M2.5",
@@ -20445,40 +21682,6 @@ export const MODELS = {
20445
21682
  contextWindow: 200000,
20446
21683
  maxTokens: 96000,
20447
21684
  },
20448
- "zai/glm-4.6v": {
20449
- id: "zai/glm-4.6v",
20450
- name: "GLM-4.6V",
20451
- api: "anthropic-messages",
20452
- provider: "vercel-ai-gateway",
20453
- baseUrl: "https://ai-gateway.vercel.sh",
20454
- reasoning: true,
20455
- input: ["text", "image"],
20456
- cost: {
20457
- input: 0.3,
20458
- output: 0.8999999999999999,
20459
- cacheRead: 0.049999999999999996,
20460
- cacheWrite: 0,
20461
- },
20462
- contextWindow: 128000,
20463
- maxTokens: 24000,
20464
- },
20465
- "zai/glm-4.6v-flash": {
20466
- id: "zai/glm-4.6v-flash",
20467
- name: "GLM-4.6V-Flash",
20468
- api: "anthropic-messages",
20469
- provider: "vercel-ai-gateway",
20470
- baseUrl: "https://ai-gateway.vercel.sh",
20471
- reasoning: true,
20472
- input: ["text", "image"],
20473
- cost: {
20474
- input: 0,
20475
- output: 0,
20476
- cacheRead: 0,
20477
- cacheWrite: 0,
20478
- },
20479
- contextWindow: 128000,
20480
- maxTokens: 24000,
20481
- },
20482
21685
  "zai/glm-4.7": {
20483
21686
  id: "zai/glm-4.7",
20484
21687
  name: "GLM 4.7",
@@ -20590,9 +21793,9 @@ export const MODELS = {
20590
21793
  reasoning: true,
20591
21794
  input: ["text"],
20592
21795
  cost: {
20593
- input: 1.1,
20594
- output: 3.851,
20595
- cacheRead: 0.275,
21796
+ input: 0.7999999999999999,
21797
+ output: 2.5500000000000003,
21798
+ cacheRead: 0.16,
20596
21799
  cacheWrite: 0,
20597
21800
  },
20598
21801
  contextWindow: 1000000,
@@ -20615,6 +21818,23 @@ export const MODELS = {
20615
21818
  contextWindow: 1000000,
20616
21819
  maxTokens: 128000,
20617
21820
  },
21821
+ "zai/glm-5.3": {
21822
+ id: "zai/glm-5.3",
21823
+ name: "GLM 5.3",
21824
+ api: "anthropic-messages",
21825
+ provider: "vercel-ai-gateway",
21826
+ baseUrl: "https://ai-gateway.vercel.sh",
21827
+ reasoning: true,
21828
+ input: ["text"],
21829
+ cost: {
21830
+ input: 1.4,
21831
+ output: 4.4,
21832
+ cacheRead: 0.26,
21833
+ cacheWrite: 0,
21834
+ },
21835
+ contextWindow: 1000000,
21836
+ maxTokens: 12800,
21837
+ },
20618
21838
  "zai/glm-5v-turbo": {
20619
21839
  id: "zai/glm-5v-turbo",
20620
21840
  name: "GLM 5V Turbo",
@@ -21134,6 +22354,24 @@ export const MODELS = {
21134
22354
  contextWindow: 1000000,
21135
22355
  maxTokens: 131072,
21136
22356
  },
22357
+ "glm-5.3": {
22358
+ id: "glm-5.3",
22359
+ name: "GLM-5.3",
22360
+ api: "openai-completions",
22361
+ provider: "zai",
22362
+ baseUrl: "https://api.z.ai/api/coding/paas/v4",
22363
+ compat: { "supportsDeveloperRole": false, "thinkingFormat": "zai", "zaiToolStream": true },
22364
+ reasoning: true,
22365
+ input: ["text"],
22366
+ cost: {
22367
+ input: 0,
22368
+ output: 0,
22369
+ cacheRead: 0,
22370
+ cacheWrite: 0,
22371
+ },
22372
+ contextWindow: 1000000,
22373
+ maxTokens: 131072,
22374
+ },
21137
22375
  },
21138
22376
  };
21139
22377
  //# sourceMappingURL=models.generated.js.map