@elyracode/ai 0.9.15 → 0.9.17

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.
@@ -1318,6 +1318,57 @@ export const MODELS = {
1318
1318
  contextWindow: 272000,
1319
1319
  maxTokens: 128000,
1320
1320
  },
1321
+ "openai.gpt-5.6-luna": {
1322
+ id: "openai.gpt-5.6-luna",
1323
+ name: "GPT-5.6 Luna",
1324
+ api: "bedrock-converse-stream",
1325
+ provider: "amazon-bedrock",
1326
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1327
+ reasoning: true,
1328
+ input: ["text", "image"],
1329
+ cost: {
1330
+ input: 1,
1331
+ output: 6,
1332
+ cacheRead: 0.1,
1333
+ cacheWrite: 1.25,
1334
+ },
1335
+ contextWindow: 272000,
1336
+ maxTokens: 128000,
1337
+ },
1338
+ "openai.gpt-5.6-sol": {
1339
+ id: "openai.gpt-5.6-sol",
1340
+ name: "GPT-5.6 Sol",
1341
+ api: "bedrock-converse-stream",
1342
+ provider: "amazon-bedrock",
1343
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1344
+ reasoning: true,
1345
+ input: ["text", "image"],
1346
+ cost: {
1347
+ input: 5,
1348
+ output: 30,
1349
+ cacheRead: 0.5,
1350
+ cacheWrite: 6.25,
1351
+ },
1352
+ contextWindow: 272000,
1353
+ maxTokens: 128000,
1354
+ },
1355
+ "openai.gpt-5.6-terra": {
1356
+ id: "openai.gpt-5.6-terra",
1357
+ name: "GPT-5.6 Terra",
1358
+ api: "bedrock-converse-stream",
1359
+ provider: "amazon-bedrock",
1360
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1361
+ reasoning: true,
1362
+ input: ["text", "image"],
1363
+ cost: {
1364
+ input: 2.5,
1365
+ output: 15,
1366
+ cacheRead: 0.25,
1367
+ cacheWrite: 3.125,
1368
+ },
1369
+ contextWindow: 272000,
1370
+ maxTokens: 128000,
1371
+ },
1321
1372
  "openai.gpt-oss-120b": {
1322
1373
  id: "openai.gpt-oss-120b",
1323
1374
  name: "gpt-oss-120b",
@@ -2942,7 +2993,7 @@ export const MODELS = {
2942
2993
  cost: {
2943
2994
  input: 2.25,
2944
2995
  output: 2.75,
2945
- cacheRead: 0,
2996
+ cacheRead: 2.25,
2946
2997
  cacheWrite: 0,
2947
2998
  },
2948
2999
  contextWindow: 131072,
@@ -3403,6 +3454,60 @@ export const MODELS = {
3403
3454
  contextWindow: 1050000,
3404
3455
  maxTokens: 128000,
3405
3456
  },
3457
+ "gpt-5.6-luna": {
3458
+ id: "gpt-5.6-luna",
3459
+ name: "GPT-5.6 Luna",
3460
+ api: "openai-responses",
3461
+ provider: "cloudflare-ai-gateway",
3462
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3463
+ reasoning: true,
3464
+ thinkingLevelMap: { "off": null },
3465
+ input: ["text", "image"],
3466
+ cost: {
3467
+ input: 1,
3468
+ output: 6,
3469
+ cacheRead: 0.1,
3470
+ cacheWrite: 0,
3471
+ },
3472
+ contextWindow: 1050000,
3473
+ maxTokens: 128000,
3474
+ },
3475
+ "gpt-5.6-sol": {
3476
+ id: "gpt-5.6-sol",
3477
+ name: "GPT-5.6 Sol",
3478
+ api: "openai-responses",
3479
+ provider: "cloudflare-ai-gateway",
3480
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3481
+ reasoning: true,
3482
+ thinkingLevelMap: { "off": null },
3483
+ input: ["text", "image"],
3484
+ cost: {
3485
+ input: 5,
3486
+ output: 30,
3487
+ cacheRead: 0.5,
3488
+ cacheWrite: 0,
3489
+ },
3490
+ contextWindow: 1050000,
3491
+ maxTokens: 128000,
3492
+ },
3493
+ "gpt-5.6-terra": {
3494
+ id: "gpt-5.6-terra",
3495
+ name: "GPT-5.6 Terra",
3496
+ api: "openai-responses",
3497
+ provider: "cloudflare-ai-gateway",
3498
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
3499
+ reasoning: true,
3500
+ thinkingLevelMap: { "off": null },
3501
+ input: ["text", "image"],
3502
+ cost: {
3503
+ input: 2.5,
3504
+ output: 15,
3505
+ cacheRead: 0.25,
3506
+ cacheWrite: 0,
3507
+ },
3508
+ contextWindow: 1050000,
3509
+ maxTokens: 128000,
3510
+ },
3406
3511
  "o1": {
3407
3512
  id: "o1",
3408
3513
  name: "o1",
@@ -3560,6 +3665,24 @@ export const MODELS = {
3560
3665
  contextWindow: 131072,
3561
3666
  maxTokens: 131072,
3562
3667
  },
3668
+ "workers-ai/@cf/zai-org/glm-5.2": {
3669
+ id: "workers-ai/@cf/zai-org/glm-5.2",
3670
+ name: "Glm 5.2",
3671
+ api: "openai-completions",
3672
+ provider: "cloudflare-ai-gateway",
3673
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/compat",
3674
+ compat: { "sendSessionAffinityHeaders": true },
3675
+ reasoning: true,
3676
+ input: ["text"],
3677
+ cost: {
3678
+ input: 1.4,
3679
+ output: 4.4,
3680
+ cacheRead: 0.26,
3681
+ cacheWrite: 0,
3682
+ },
3683
+ contextWindow: 262144,
3684
+ maxTokens: 262144,
3685
+ },
3563
3686
  },
3564
3687
  "cloudflare-workers-ai": {
3565
3688
  "@cf/google/gemma-4-26b-a4b-it": {
@@ -4562,7 +4685,7 @@ export const MODELS = {
4562
4685
  input: 1,
4563
4686
  output: 6,
4564
4687
  cacheRead: 0.1,
4565
- cacheWrite: 0,
4688
+ cacheWrite: 1.25,
4566
4689
  },
4567
4690
  contextWindow: 1050000,
4568
4691
  maxTokens: 128000,
@@ -4581,7 +4704,7 @@ export const MODELS = {
4581
4704
  input: 5,
4582
4705
  output: 30,
4583
4706
  cacheRead: 0.5,
4584
- cacheWrite: 0,
4707
+ cacheWrite: 6.25,
4585
4708
  },
4586
4709
  contextWindow: 1050000,
4587
4710
  maxTokens: 128000,
@@ -4600,7 +4723,7 @@ export const MODELS = {
4600
4723
  input: 2.5,
4601
4724
  output: 15,
4602
4725
  cacheRead: 0.25,
4603
- cacheWrite: 0,
4726
+ cacheWrite: 3.125,
4604
4727
  },
4605
4728
  contextWindow: 1050000,
4606
4729
  maxTokens: 128000,
@@ -4865,9 +4988,9 @@ export const MODELS = {
4865
4988
  reasoning: true,
4866
4989
  input: ["text", "image"],
4867
4990
  cost: {
4868
- input: 0.3,
4869
- output: 2.5,
4870
- cacheRead: 0.075,
4991
+ input: 1.5,
4992
+ output: 9,
4993
+ cacheRead: 0.15,
4871
4994
  cacheWrite: 0,
4872
4995
  },
4873
4996
  contextWindow: 1048576,
@@ -4882,8 +5005,8 @@ export const MODELS = {
4882
5005
  reasoning: true,
4883
5006
  input: ["text", "image"],
4884
5007
  cost: {
4885
- input: 0.1,
4886
- output: 0.4,
5008
+ input: 0.25,
5009
+ output: 1.5,
4887
5010
  cacheRead: 0.025,
4888
5011
  cacheWrite: 0,
4889
5012
  },
@@ -6161,9 +6284,9 @@ export const MODELS = {
6161
6284
  contextWindow: 262144,
6162
6285
  maxTokens: 32768,
6163
6286
  },
6164
- "kimi-for-coding": {
6165
- id: "kimi-for-coding",
6166
- name: "Kimi For Coding",
6287
+ "k3": {
6288
+ id: "k3",
6289
+ name: "Kimi K3",
6167
6290
  api: "anthropic-messages",
6168
6291
  provider: "kimi-coding",
6169
6292
  baseUrl: "https://api.kimi.com/coding",
@@ -6176,18 +6299,18 @@ export const MODELS = {
6176
6299
  cacheRead: 0,
6177
6300
  cacheWrite: 0,
6178
6301
  },
6179
- contextWindow: 262144,
6180
- maxTokens: 32768,
6302
+ contextWindow: 1048576,
6303
+ maxTokens: 131072,
6181
6304
  },
6182
- "kimi-k2-thinking": {
6183
- id: "kimi-k2-thinking",
6184
- name: "Kimi K2 Thinking",
6305
+ "kimi-for-coding-highspeed": {
6306
+ id: "kimi-for-coding-highspeed",
6307
+ name: "Kimi For Coding HighSpeed",
6185
6308
  api: "anthropic-messages",
6186
6309
  provider: "kimi-coding",
6187
6310
  baseUrl: "https://api.kimi.com/coding",
6188
6311
  headers: { "User-Agent": "KimiCLI/1.5" },
6189
6312
  reasoning: true,
6190
- input: ["text"],
6313
+ input: ["text", "image"],
6191
6314
  cost: {
6192
6315
  input: 0,
6193
6316
  output: 0,
@@ -6979,6 +7102,24 @@ export const MODELS = {
6979
7102
  contextWindow: 262144,
6980
7103
  maxTokens: 262144,
6981
7104
  },
7105
+ "kimi-k3": {
7106
+ id: "kimi-k3",
7107
+ name: "Kimi K3",
7108
+ api: "openai-completions",
7109
+ provider: "moonshotai",
7110
+ baseUrl: "https://api.moonshot.ai/v1",
7111
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7112
+ reasoning: true,
7113
+ input: ["text", "image"],
7114
+ cost: {
7115
+ input: 3,
7116
+ output: 15,
7117
+ cacheRead: 0.3,
7118
+ cacheWrite: 0,
7119
+ },
7120
+ contextWindow: 1048576,
7121
+ maxTokens: 131072,
7122
+ },
6982
7123
  },
6983
7124
  "moonshotai-cn": {
6984
7125
  "kimi-k2-0711-preview": {
@@ -7143,6 +7284,24 @@ export const MODELS = {
7143
7284
  contextWindow: 262144,
7144
7285
  maxTokens: 262144,
7145
7286
  },
7287
+ "kimi-k3": {
7288
+ id: "kimi-k3",
7289
+ name: "Kimi K3",
7290
+ api: "openai-completions",
7291
+ provider: "moonshotai-cn",
7292
+ baseUrl: "https://api.moonshot.cn/v1",
7293
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7294
+ reasoning: true,
7295
+ input: ["text", "image"],
7296
+ cost: {
7297
+ input: 3,
7298
+ output: 15,
7299
+ cacheRead: 0.3,
7300
+ cacheWrite: 0,
7301
+ },
7302
+ contextWindow: 1048576,
7303
+ maxTokens: 131072,
7304
+ },
7146
7305
  },
7147
7306
  "openai": {
7148
7307
  "gpt-4": {
@@ -9068,9 +9227,9 @@ export const MODELS = {
9068
9227
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9069
9228
  input: ["text"],
9070
9229
  cost: {
9071
- input: 1.74,
9072
- output: 3.48,
9073
- cacheRead: 0.0145,
9230
+ input: 0.435,
9231
+ output: 0.87,
9232
+ cacheRead: 0.003625,
9074
9233
  cacheWrite: 0,
9075
9234
  },
9076
9235
  contextWindow: 1000000,
@@ -9110,6 +9269,23 @@ export const MODELS = {
9110
9269
  contextWindow: 1000000,
9111
9270
  maxTokens: 131072,
9112
9271
  },
9272
+ "grok-4.5": {
9273
+ id: "grok-4.5",
9274
+ name: "Grok 4.5",
9275
+ api: "openai-completions",
9276
+ provider: "opencode-go",
9277
+ baseUrl: "https://opencode.ai/zen/go/v1",
9278
+ reasoning: true,
9279
+ input: ["text", "image"],
9280
+ cost: {
9281
+ input: 2,
9282
+ output: 6,
9283
+ cacheRead: 0.5,
9284
+ cacheWrite: 0,
9285
+ },
9286
+ contextWindow: 500000,
9287
+ maxTokens: 500000,
9288
+ },
9113
9289
  "kimi-k2.6": {
9114
9290
  id: "kimi-k2.6",
9115
9291
  name: "Kimi K2.6",
@@ -9144,6 +9320,23 @@ export const MODELS = {
9144
9320
  contextWindow: 262144,
9145
9321
  maxTokens: 262144,
9146
9322
  },
9323
+ "kimi-k3": {
9324
+ id: "kimi-k3",
9325
+ name: "Kimi K3 (2x usage)",
9326
+ api: "openai-completions",
9327
+ provider: "opencode-go",
9328
+ baseUrl: "https://opencode.ai/zen/go/v1",
9329
+ reasoning: true,
9330
+ input: ["text", "image"],
9331
+ cost: {
9332
+ input: 3,
9333
+ output: 15,
9334
+ cacheRead: 0.3,
9335
+ cacheWrite: 0,
9336
+ },
9337
+ contextWindow: 1048576,
9338
+ maxTokens: 131072,
9339
+ },
9147
9340
  "mimo-v2.5": {
9148
9341
  id: "mimo-v2.5",
9149
9342
  name: "MiMo V2.5",
@@ -9170,9 +9363,9 @@ export const MODELS = {
9170
9363
  reasoning: true,
9171
9364
  input: ["text"],
9172
9365
  cost: {
9173
- input: 1.74,
9174
- output: 3.48,
9175
- cacheRead: 0.0145,
9366
+ input: 0.435,
9367
+ output: 0.87,
9368
+ cacheRead: 0.003625,
9176
9369
  cacheWrite: 0,
9177
9370
  },
9178
9371
  contextWindow: 1048576,
@@ -9876,13 +10069,13 @@ export const MODELS = {
9876
10069
  reasoning: false,
9877
10070
  input: ["text"],
9878
10071
  cost: {
9879
- input: 0.24,
9880
- output: 0.8999999999999999,
10072
+ input: 0.27,
10073
+ output: 1.12,
9881
10074
  cacheRead: 0.135,
9882
10075
  cacheWrite: 0,
9883
10076
  },
9884
10077
  contextWindow: 163840,
9885
- maxTokens: 16384,
10078
+ maxTokens: 65536,
9886
10079
  },
9887
10080
  "deepseek/deepseek-chat-v3.1": {
9888
10081
  id: "deepseek/deepseek-chat-v3.1",
@@ -9893,8 +10086,8 @@ export const MODELS = {
9893
10086
  reasoning: true,
9894
10087
  input: ["text"],
9895
10088
  cost: {
9896
- input: 0.21,
9897
- output: 0.7899999999999999,
10089
+ input: 0.25,
10090
+ output: 0.95,
9898
10091
  cacheRead: 0.13,
9899
10092
  cacheWrite: 0,
9900
10093
  },
@@ -9945,11 +10138,11 @@ export const MODELS = {
9945
10138
  input: ["text"],
9946
10139
  cost: {
9947
10140
  input: 0.27,
9948
- output: 0.95,
9949
- cacheRead: 0.13,
10141
+ output: 1,
10142
+ cacheRead: 0.135,
9950
10143
  cacheWrite: 0,
9951
10144
  },
9952
- contextWindow: 163840,
10145
+ contextWindow: 131072,
9953
10146
  maxTokens: 32768,
9954
10147
  },
9955
10148
  "deepseek/deepseek-v3.2": {
@@ -9961,13 +10154,13 @@ export const MODELS = {
9961
10154
  reasoning: true,
9962
10155
  input: ["text"],
9963
10156
  cost: {
9964
- input: 0.2145,
9965
- output: 0.32175,
9966
- cacheRead: 0.02145,
10157
+ input: 0.26899999999999996,
10158
+ output: 0.39999999999999997,
10159
+ cacheRead: 0.13449999999999998,
9967
10160
  cacheWrite: 0,
9968
10161
  },
9969
- contextWindow: 131072,
9970
- maxTokens: 64000,
10162
+ contextWindow: 163840,
10163
+ maxTokens: 65536,
9971
10164
  },
9972
10165
  "deepseek/deepseek-v3.2-exp": {
9973
10166
  id: "deepseek/deepseek-v3.2-exp",
@@ -9997,13 +10190,13 @@ export const MODELS = {
9997
10190
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9998
10191
  input: ["text"],
9999
10192
  cost: {
10000
- input: 0.077,
10001
- output: 0.154,
10002
- cacheRead: 0.015399999999999999,
10193
+ input: 0.098,
10194
+ output: 0.196,
10195
+ cacheRead: 0.0196,
10003
10196
  cacheWrite: 0,
10004
10197
  },
10005
10198
  contextWindow: 1048576,
10006
- maxTokens: 384000,
10199
+ maxTokens: 4096,
10007
10200
  },
10008
10201
  "deepseek/deepseek-v4-pro": {
10009
10202
  id: "deepseek/deepseek-v4-pro",
@@ -10254,13 +10447,13 @@ export const MODELS = {
10254
10447
  reasoning: false,
10255
10448
  input: ["text", "image"],
10256
10449
  cost: {
10257
- input: 0.08,
10258
- output: 0.16,
10450
+ input: 0.09999999999999999,
10451
+ output: 0.3,
10259
10452
  cacheRead: 0,
10260
10453
  cacheWrite: 0,
10261
10454
  },
10262
10455
  contextWindow: 131072,
10263
- maxTokens: 16384,
10456
+ maxTokens: 4096,
10264
10457
  },
10265
10458
  "google/gemma-4-26b-a4b-it": {
10266
10459
  id: "google/gemma-4-26b-a4b-it",
@@ -10271,13 +10464,13 @@ export const MODELS = {
10271
10464
  reasoning: true,
10272
10465
  input: ["text", "image"],
10273
10466
  cost: {
10274
- input: 0.06,
10275
- output: 0.33,
10467
+ input: 0.09999999999999999,
10468
+ output: 0.3,
10276
10469
  cacheRead: 0,
10277
10470
  cacheWrite: 0,
10278
10471
  },
10279
10472
  contextWindow: 262144,
10280
- maxTokens: 4096,
10473
+ maxTokens: 256000,
10281
10474
  },
10282
10475
  "google/gemma-4-26b-a4b-it:free": {
10283
10476
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10305,9 +10498,9 @@ export const MODELS = {
10305
10498
  reasoning: true,
10306
10499
  input: ["text", "image"],
10307
10500
  cost: {
10308
- input: 0.12,
10309
- output: 0.35,
10310
- cacheRead: 0.09,
10501
+ input: 0.22,
10502
+ output: 0.55,
10503
+ cacheRead: 0.12,
10311
10504
  cacheWrite: 0,
10312
10505
  },
10313
10506
  contextWindow: 262144,
@@ -10328,7 +10521,7 @@ export const MODELS = {
10328
10521
  cacheWrite: 0,
10329
10522
  },
10330
10523
  contextWindow: 262144,
10331
- maxTokens: 8192,
10524
+ maxTokens: 32768,
10332
10525
  },
10333
10526
  "ibm-granite/granite-4.1-8b": {
10334
10527
  id: "ibm-granite/granite-4.1-8b",
@@ -10415,9 +10608,26 @@ export const MODELS = {
10415
10608
  contextWindow: 262144,
10416
10609
  maxTokens: 65536,
10417
10610
  },
10418
- "kwaipilot/kat-coder-pro-v2": {
10419
- id: "kwaipilot/kat-coder-pro-v2",
10420
- name: "Kwaipilot: KAT-Coder-Pro V2",
10611
+ "kwaipilot/kat-coder-air-v2.5": {
10612
+ id: "kwaipilot/kat-coder-air-v2.5",
10613
+ name: "Kwaipilot: KAT-Coder-Air V2.5",
10614
+ api: "openai-completions",
10615
+ provider: "openrouter",
10616
+ baseUrl: "https://openrouter.ai/api/v1",
10617
+ reasoning: false,
10618
+ input: ["text"],
10619
+ cost: {
10620
+ input: 0.15,
10621
+ output: 0.6,
10622
+ cacheRead: 0.03,
10623
+ cacheWrite: 0,
10624
+ },
10625
+ contextWindow: 256000,
10626
+ maxTokens: 80000,
10627
+ },
10628
+ "kwaipilot/kat-coder-pro-v2": {
10629
+ id: "kwaipilot/kat-coder-pro-v2",
10630
+ name: "Kwaipilot: KAT-Coder-Pro V2",
10421
10631
  api: "openai-completions",
10422
10632
  provider: "openrouter",
10423
10633
  baseUrl: "https://openrouter.ai/api/v1",
@@ -10432,22 +10642,22 @@ export const MODELS = {
10432
10642
  contextWindow: 256000,
10433
10643
  maxTokens: 80000,
10434
10644
  },
10435
- "liquid/lfm-2.5-1.2b-thinking:free": {
10436
- id: "liquid/lfm-2.5-1.2b-thinking:free",
10437
- name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
10645
+ "kwaipilot/kat-coder-pro-v2.5": {
10646
+ id: "kwaipilot/kat-coder-pro-v2.5",
10647
+ name: "Kwaipilot: KAT-Coder-Pro V2.5",
10438
10648
  api: "openai-completions",
10439
10649
  provider: "openrouter",
10440
10650
  baseUrl: "https://openrouter.ai/api/v1",
10441
- reasoning: true,
10651
+ reasoning: false,
10442
10652
  input: ["text"],
10443
10653
  cost: {
10444
- input: 0,
10445
- output: 0,
10446
- cacheRead: 0,
10654
+ input: 0.74,
10655
+ output: 2.96,
10656
+ cacheRead: 0.15,
10447
10657
  cacheWrite: 0,
10448
10658
  },
10449
- contextWindow: 32768,
10450
- maxTokens: 4096,
10659
+ contextWindow: 256000,
10660
+ maxTokens: 80000,
10451
10661
  },
10452
10662
  "meta-llama/llama-3.1-70b-instruct": {
10453
10663
  id: "meta-llama/llama-3.1-70b-instruct",
@@ -10475,13 +10685,13 @@ export const MODELS = {
10475
10685
  reasoning: false,
10476
10686
  input: ["text"],
10477
10687
  cost: {
10478
- input: 0.02,
10479
- output: 0.03,
10480
- cacheRead: 0,
10688
+ input: 0.049999999999999996,
10689
+ output: 0.08,
10690
+ cacheRead: 0.024999999999999998,
10481
10691
  cacheWrite: 0,
10482
10692
  },
10483
10693
  contextWindow: 131072,
10484
- maxTokens: 16384,
10694
+ maxTokens: 131072,
10485
10695
  },
10486
10696
  "meta-llama/llama-3.3-70b-instruct": {
10487
10697
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -10492,13 +10702,13 @@ export const MODELS = {
10492
10702
  reasoning: false,
10493
10703
  input: ["text"],
10494
10704
  cost: {
10495
- input: 0.09999999999999999,
10496
- output: 0.32,
10705
+ input: 0.13,
10706
+ output: 0.39999999999999997,
10497
10707
  cacheRead: 0,
10498
10708
  cacheWrite: 0,
10499
10709
  },
10500
10710
  contextWindow: 131072,
10501
- maxTokens: 16384,
10711
+ maxTokens: 128000,
10502
10712
  },
10503
10713
  "meta-llama/llama-3.3-70b-instruct:free": {
10504
10714
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -10551,6 +10761,23 @@ export const MODELS = {
10551
10761
  contextWindow: 10000000,
10552
10762
  maxTokens: 16384,
10553
10763
  },
10764
+ "meta/muse-spark-1.1": {
10765
+ id: "meta/muse-spark-1.1",
10766
+ name: "Meta: Muse Spark 1.1",
10767
+ api: "openai-completions",
10768
+ provider: "openrouter",
10769
+ baseUrl: "https://openrouter.ai/api/v1",
10770
+ reasoning: true,
10771
+ input: ["text", "image"],
10772
+ cost: {
10773
+ input: 1.25,
10774
+ output: 4.25,
10775
+ cacheRead: 0.15,
10776
+ cacheWrite: 0,
10777
+ },
10778
+ contextWindow: 1048576,
10779
+ maxTokens: 4096,
10780
+ },
10554
10781
  "minimax/minimax-m1": {
10555
10782
  id: "minimax/minimax-m1",
10556
10783
  name: "MiniMax: MiniMax M1",
@@ -10560,7 +10787,7 @@ export const MODELS = {
10560
10787
  reasoning: true,
10561
10788
  input: ["text"],
10562
10789
  cost: {
10563
- input: 0.39999999999999997,
10790
+ input: 0.55,
10564
10791
  output: 2.2,
10565
10792
  cacheRead: 0,
10566
10793
  cacheWrite: 0,
@@ -10628,13 +10855,13 @@ export const MODELS = {
10628
10855
  reasoning: true,
10629
10856
  input: ["text"],
10630
10857
  cost: {
10631
- input: 0.24,
10632
- output: 0.96,
10633
- cacheRead: 0,
10858
+ input: 0.25,
10859
+ output: 1,
10860
+ cacheRead: 0.049999999999999996,
10634
10861
  cacheWrite: 0,
10635
10862
  },
10636
10863
  contextWindow: 204800,
10637
- maxTokens: 196608,
10864
+ maxTokens: 131072,
10638
10865
  },
10639
10866
  "minimax/minimax-m3": {
10640
10867
  id: "minimax/minimax-m3",
@@ -10651,7 +10878,7 @@ export const MODELS = {
10651
10878
  cacheWrite: 0,
10652
10879
  },
10653
10880
  contextWindow: 1048576,
10654
- maxTokens: 131072,
10881
+ maxTokens: 512000,
10655
10882
  },
10656
10883
  "mistralai/codestral-2508": {
10657
10884
  id: "mistralai/codestral-2508",
@@ -10849,13 +11076,13 @@ export const MODELS = {
10849
11076
  reasoning: false,
10850
11077
  input: ["text"],
10851
11078
  cost: {
10852
- input: 0.02,
11079
+ input: 0.019000000000000003,
10853
11080
  output: 0.03,
10854
11081
  cacheRead: 0,
10855
11082
  cacheWrite: 0,
10856
11083
  },
10857
11084
  contextWindow: 131072,
10858
- maxTokens: 4096,
11085
+ maxTokens: 16384,
10859
11086
  },
10860
11087
  "mistralai/mistral-saba": {
10861
11088
  id: "mistralai/mistral-saba",
@@ -10900,13 +11127,13 @@ export const MODELS = {
10900
11127
  reasoning: false,
10901
11128
  input: ["text", "image"],
10902
11129
  cost: {
10903
- input: 0.075,
10904
- output: 0.19999999999999998,
10905
- cacheRead: 0,
11130
+ input: 0.09999999999999999,
11131
+ output: 0.3,
11132
+ cacheRead: 0.01,
10906
11133
  cacheWrite: 0,
10907
11134
  },
10908
- contextWindow: 128000,
10909
- maxTokens: 16384,
11135
+ contextWindow: 131072,
11136
+ maxTokens: 4096,
10910
11137
  },
10911
11138
  "mistralai/mixtral-8x22b-instruct": {
10912
11139
  id: "mistralai/mixtral-8x22b-instruct",
@@ -11019,13 +11246,13 @@ export const MODELS = {
11019
11246
  reasoning: true,
11020
11247
  input: ["text", "image"],
11021
11248
  cost: {
11022
- input: 0.66,
11023
- output: 3.41,
11024
- cacheRead: 0.15,
11249
+ input: 0.95,
11250
+ output: 4,
11251
+ cacheRead: 0.16,
11025
11252
  cacheWrite: 0,
11026
11253
  },
11027
11254
  contextWindow: 262144,
11028
- maxTokens: 262144,
11255
+ maxTokens: 4096,
11029
11256
  },
11030
11257
  "moonshotai/kimi-k2.7-code": {
11031
11258
  id: "moonshotai/kimi-k2.7-code",
@@ -11036,14 +11263,31 @@ export const MODELS = {
11036
11263
  reasoning: true,
11037
11264
  input: ["text", "image"],
11038
11265
  cost: {
11039
- input: 0.72,
11040
- output: 3.5,
11041
- cacheRead: 0.15,
11266
+ input: 1,
11267
+ output: 4.4,
11268
+ cacheRead: 0.19999999999999998,
11042
11269
  cacheWrite: 0,
11043
11270
  },
11044
11271
  contextWindow: 262144,
11045
11272
  maxTokens: 262144,
11046
11273
  },
11274
+ "moonshotai/kimi-k3": {
11275
+ id: "moonshotai/kimi-k3",
11276
+ name: "MoonshotAI: Kimi K3",
11277
+ api: "openai-completions",
11278
+ provider: "openrouter",
11279
+ baseUrl: "https://openrouter.ai/api/v1",
11280
+ reasoning: true,
11281
+ input: ["text", "image"],
11282
+ cost: {
11283
+ input: 3,
11284
+ output: 15,
11285
+ cacheRead: 0.3,
11286
+ cacheWrite: 0,
11287
+ },
11288
+ contextWindow: 1048576,
11289
+ maxTokens: 4096,
11290
+ },
11047
11291
  "nex-agi/nex-n2-mini": {
11048
11292
  id: "nex-agi/nex-n2-mini",
11049
11293
  name: "Nex AGI: Nex-N2-Mini",
@@ -11078,23 +11322,6 @@ export const MODELS = {
11078
11322
  contextWindow: 262144,
11079
11323
  maxTokens: 262144,
11080
11324
  },
11081
- "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
11082
- id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
11083
- name: "NVIDIA: Llama 3.3 Nemotron Super 49B V1.5",
11084
- api: "openai-completions",
11085
- provider: "openrouter",
11086
- baseUrl: "https://openrouter.ai/api/v1",
11087
- reasoning: true,
11088
- input: ["text"],
11089
- cost: {
11090
- input: 0.39999999999999997,
11091
- output: 0.39999999999999997,
11092
- cacheRead: 0,
11093
- cacheWrite: 0,
11094
- },
11095
- contextWindow: 131072,
11096
- maxTokens: 16384,
11097
- },
11098
11325
  "nvidia/nemotron-3-nano-30b-a3b": {
11099
11326
  id: "nvidia/nemotron-3-nano-30b-a3b",
11100
11327
  name: "NVIDIA: Nemotron 3 Nano 30B A3B",
@@ -11155,9 +11382,9 @@ export const MODELS = {
11155
11382
  reasoning: true,
11156
11383
  input: ["text"],
11157
11384
  cost: {
11158
- input: 0.08,
11159
- output: 0.44999999999999996,
11160
- cacheRead: 0,
11385
+ input: 0.21,
11386
+ output: 0.45499999999999996,
11387
+ cacheRead: 0.06,
11161
11388
  cacheWrite: 0,
11162
11389
  },
11163
11390
  contextWindow: 1000000,
@@ -11189,13 +11416,13 @@ export const MODELS = {
11189
11416
  reasoning: true,
11190
11417
  input: ["text"],
11191
11418
  cost: {
11192
- input: 0.5,
11193
- output: 2.2,
11194
- cacheRead: 0.09999999999999999,
11419
+ input: 0.6,
11420
+ output: 3.5999999999999996,
11421
+ cacheRead: 0.19999999999999998,
11195
11422
  cacheWrite: 0,
11196
11423
  },
11197
11424
  contextWindow: 1000000,
11198
- maxTokens: 16384,
11425
+ maxTokens: 4096,
11199
11426
  },
11200
11427
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
11201
11428
  id: "nvidia/nemotron-3-ultra-550b-a55b:free",
@@ -11331,7 +11558,7 @@ export const MODELS = {
11331
11558
  cacheWrite: 0,
11332
11559
  },
11333
11560
  contextWindow: 1047576,
11334
- maxTokens: 4096,
11561
+ maxTokens: 32768,
11335
11562
  },
11336
11563
  "openai/gpt-4.1-mini": {
11337
11564
  id: "openai/gpt-4.1-mini",
@@ -11378,7 +11605,7 @@ export const MODELS = {
11378
11605
  cost: {
11379
11606
  input: 2.5,
11380
11607
  output: 10,
11381
- cacheRead: 0,
11608
+ cacheRead: 1.25,
11382
11609
  cacheWrite: 0,
11383
11610
  },
11384
11611
  contextWindow: 128000,
@@ -11463,11 +11690,11 @@ export const MODELS = {
11463
11690
  cost: {
11464
11691
  input: 0.049999999999999996,
11465
11692
  output: 0.39999999999999997,
11466
- cacheRead: 0.01,
11693
+ cacheRead: 0.005,
11467
11694
  cacheWrite: 0,
11468
11695
  },
11469
11696
  contextWindow: 400000,
11470
- maxTokens: 4096,
11697
+ maxTokens: 128000,
11471
11698
  },
11472
11699
  "openai/gpt-5-pro": {
11473
11700
  id: "openai/gpt-5-pro",
@@ -11497,7 +11724,7 @@ export const MODELS = {
11497
11724
  cost: {
11498
11725
  input: 1.25,
11499
11726
  output: 10,
11500
- cacheRead: 0.13,
11727
+ cacheRead: 0.125,
11501
11728
  cacheWrite: 0,
11502
11729
  },
11503
11730
  contextWindow: 400000,
@@ -11514,11 +11741,11 @@ export const MODELS = {
11514
11741
  cost: {
11515
11742
  input: 1.25,
11516
11743
  output: 10,
11517
- cacheRead: 0.13,
11744
+ cacheRead: 0.125,
11518
11745
  cacheWrite: 0,
11519
11746
  },
11520
11747
  contextWindow: 128000,
11521
- maxTokens: 32000,
11748
+ maxTokens: 16384,
11522
11749
  },
11523
11750
  "openai/gpt-5.1-codex": {
11524
11751
  id: "openai/gpt-5.1-codex",
@@ -11531,7 +11758,7 @@ export const MODELS = {
11531
11758
  cost: {
11532
11759
  input: 1.25,
11533
11760
  output: 10,
11534
- cacheRead: 0.13,
11761
+ cacheRead: 0.125,
11535
11762
  cacheWrite: 0,
11536
11763
  },
11537
11764
  contextWindow: 400000,
@@ -11949,25 +12176,8 @@ export const MODELS = {
11949
12176
  reasoning: true,
11950
12177
  input: ["text"],
11951
12178
  cost: {
11952
- input: 0.036,
11953
- output: 0.18,
11954
- cacheRead: 0,
11955
- cacheWrite: 0,
11956
- },
11957
- contextWindow: 131072,
11958
- maxTokens: 4096,
11959
- },
11960
- "openai/gpt-oss-120b:free": {
11961
- id: "openai/gpt-oss-120b:free",
11962
- name: "OpenAI: gpt-oss-120b (free)",
11963
- api: "openai-completions",
11964
- provider: "openrouter",
11965
- baseUrl: "https://openrouter.ai/api/v1",
11966
- reasoning: true,
11967
- input: ["text"],
11968
- cost: {
11969
- input: 0,
11970
- output: 0,
12179
+ input: 0.037,
12180
+ output: 0.16999999999999998,
11971
12181
  cacheRead: 0,
11972
12182
  cacheWrite: 0,
11973
12183
  },
@@ -11983,13 +12193,13 @@ export const MODELS = {
11983
12193
  reasoning: true,
11984
12194
  input: ["text"],
11985
12195
  cost: {
11986
- input: 0.029,
11987
- output: 0.14,
11988
- cacheRead: 0,
12196
+ input: 0.03,
12197
+ output: 0.13,
12198
+ cacheRead: 0.03,
11989
12199
  cacheWrite: 0,
11990
12200
  },
11991
12201
  contextWindow: 131072,
11992
- maxTokens: 4096,
12202
+ maxTokens: 131072,
11993
12203
  },
11994
12204
  "openai/gpt-oss-20b:free": {
11995
12205
  id: "openai/gpt-oss-20b:free",
@@ -12195,6 +12405,23 @@ export const MODELS = {
12195
12405
  contextWindow: 2000000,
12196
12406
  maxTokens: 4096,
12197
12407
  },
12408
+ "openrouter/auto-beta": {
12409
+ id: "openrouter/auto-beta",
12410
+ name: "Auto Router (Beta)",
12411
+ api: "openai-completions",
12412
+ provider: "openrouter",
12413
+ baseUrl: "https://openrouter.ai/api/v1",
12414
+ reasoning: true,
12415
+ input: ["text", "image"],
12416
+ cost: {
12417
+ input: -1000000,
12418
+ output: -1000000,
12419
+ cacheRead: 0,
12420
+ cacheWrite: 0,
12421
+ },
12422
+ contextWindow: 2000000,
12423
+ maxTokens: 4096,
12424
+ },
12198
12425
  "openrouter/free": {
12199
12426
  id: "openrouter/free",
12200
12427
  name: "Free Models Router",
@@ -12374,13 +12601,13 @@ export const MODELS = {
12374
12601
  reasoning: true,
12375
12602
  input: ["text"],
12376
12603
  cost: {
12377
- input: 0.09999999999999999,
12604
+ input: 0.12,
12378
12605
  output: 0.24,
12379
12606
  cacheRead: 0,
12380
12607
  cacheWrite: 0,
12381
12608
  },
12382
12609
  contextWindow: 131702,
12383
- maxTokens: 40960,
12610
+ maxTokens: 16384,
12384
12611
  },
12385
12612
  "qwen/qwen3-235b-a22b": {
12386
12613
  id: "qwen/qwen3-235b-a22b",
@@ -12442,13 +12669,13 @@ export const MODELS = {
12442
12669
  reasoning: true,
12443
12670
  input: ["text"],
12444
12671
  cost: {
12445
- input: 0.12,
12446
- output: 0.5,
12672
+ input: 0.13,
12673
+ output: 0.52,
12447
12674
  cacheRead: 0,
12448
12675
  cacheWrite: 0,
12449
12676
  },
12450
12677
  contextWindow: 131072,
12451
- maxTokens: 16384,
12678
+ maxTokens: 8192,
12452
12679
  },
12453
12680
  "qwen/qwen3-30b-a3b-instruct-2507": {
12454
12681
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -12459,13 +12686,13 @@ export const MODELS = {
12459
12686
  reasoning: false,
12460
12687
  input: ["text"],
12461
12688
  cost: {
12462
- input: 0.04815,
12463
- output: 0.19305,
12689
+ input: 0.09999999999999999,
12690
+ output: 0.3,
12464
12691
  cacheRead: 0,
12465
12692
  cacheWrite: 0,
12466
12693
  },
12467
- contextWindow: 131072,
12468
- maxTokens: 32000,
12694
+ contextWindow: 262144,
12695
+ maxTokens: 4096,
12469
12696
  },
12470
12697
  "qwen/qwen3-30b-a3b-thinking-2507": {
12471
12698
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -12527,9 +12754,9 @@ export const MODELS = {
12527
12754
  reasoning: false,
12528
12755
  input: ["text"],
12529
12756
  cost: {
12530
- input: 0.22,
12531
- output: 1.7999999999999998,
12532
- cacheRead: 0,
12757
+ input: 0.3,
12758
+ output: 1,
12759
+ cacheRead: 0.09999999999999999,
12533
12760
  cacheWrite: 0,
12534
12761
  },
12535
12762
  contextWindow: 1048576,
@@ -12663,13 +12890,13 @@ export const MODELS = {
12663
12890
  reasoning: false,
12664
12891
  input: ["text"],
12665
12892
  cost: {
12666
- input: 0.09,
12893
+ input: 0.09999999999999999,
12667
12894
  output: 1.1,
12668
- cacheRead: 0,
12895
+ cacheRead: 0.07,
12669
12896
  cacheWrite: 0,
12670
12897
  },
12671
12898
  contextWindow: 262144,
12672
- maxTokens: 16384,
12899
+ maxTokens: 262144,
12673
12900
  },
12674
12901
  "qwen/qwen3-next-80b-a3b-instruct:free": {
12675
12902
  id: "qwen/qwen3-next-80b-a3b-instruct:free",
@@ -12714,13 +12941,13 @@ export const MODELS = {
12714
12941
  reasoning: false,
12715
12942
  input: ["text", "image"],
12716
12943
  cost: {
12717
- input: 0.19999999999999998,
12718
- output: 0.88,
12719
- cacheRead: 0.11,
12944
+ input: 0.21,
12945
+ output: 1.9,
12946
+ cacheRead: 0.09999999999999999,
12720
12947
  cacheWrite: 0,
12721
12948
  },
12722
- contextWindow: 262144,
12723
- maxTokens: 16384,
12949
+ contextWindow: 131072,
12950
+ maxTokens: 32768,
12724
12951
  },
12725
12952
  "qwen/qwen3-vl-235b-a22b-thinking": {
12726
12953
  id: "qwen/qwen3-vl-235b-a22b-thinking",
@@ -12839,7 +13066,7 @@ export const MODELS = {
12839
13066
  cacheWrite: 0,
12840
13067
  },
12841
13068
  contextWindow: 262144,
12842
- maxTokens: 262144,
13069
+ maxTokens: 65536,
12843
13070
  },
12844
13071
  "qwen/qwen3.5-27b": {
12845
13072
  id: "qwen/qwen3.5-27b",
@@ -12850,13 +13077,13 @@ export const MODELS = {
12850
13077
  reasoning: true,
12851
13078
  input: ["text", "image"],
12852
13079
  cost: {
12853
- input: 0.195,
12854
- output: 1.56,
13080
+ input: 0.26,
13081
+ output: 2.6,
12855
13082
  cacheRead: 0,
12856
13083
  cacheWrite: 0,
12857
13084
  },
12858
13085
  contextWindow: 262144,
12859
- maxTokens: 65536,
13086
+ maxTokens: 81920,
12860
13087
  },
12861
13088
  "qwen/qwen3.5-35b-a3b": {
12862
13089
  id: "qwen/qwen3.5-35b-a3b",
@@ -12869,11 +13096,11 @@ export const MODELS = {
12869
13096
  cost: {
12870
13097
  input: 0.14,
12871
13098
  output: 1,
12872
- cacheRead: 0.049999999999999996,
13099
+ cacheRead: 0,
12873
13100
  cacheWrite: 0,
12874
13101
  },
12875
13102
  contextWindow: 262144,
12876
- maxTokens: 81920,
13103
+ maxTokens: 262144,
12877
13104
  },
12878
13105
  "qwen/qwen3.5-397b-a17b": {
12879
13106
  id: "qwen/qwen3.5-397b-a17b",
@@ -12884,13 +13111,13 @@ export const MODELS = {
12884
13111
  reasoning: true,
12885
13112
  input: ["text", "image"],
12886
13113
  cost: {
12887
- input: 0.385,
12888
- output: 2.4499999999999997,
12889
- cacheRead: 0.111,
13114
+ input: 0.39,
13115
+ output: 2.34,
13116
+ cacheRead: 0,
12890
13117
  cacheWrite: 0,
12891
13118
  },
12892
- contextWindow: 256000,
12893
- maxTokens: 4096,
13119
+ contextWindow: 262144,
13120
+ maxTokens: 65536,
12894
13121
  },
12895
13122
  "qwen/qwen3.5-9b": {
12896
13123
  id: "qwen/qwen3.5-9b",
@@ -12969,13 +13196,13 @@ export const MODELS = {
12969
13196
  reasoning: true,
12970
13197
  input: ["text", "image"],
12971
13198
  cost: {
12972
- input: 0.28500000000000003,
12973
- output: 2.4,
12974
- cacheRead: 0.15,
13199
+ input: 0.44999999999999996,
13200
+ output: 2.7,
13201
+ cacheRead: 0,
12975
13202
  cacheWrite: 0,
12976
13203
  },
12977
13204
  contextWindow: 262144,
12978
- maxTokens: 262140,
13205
+ maxTokens: 65536,
12979
13206
  },
12980
13207
  "qwen/qwen3.6-35b-a3b": {
12981
13208
  id: "qwen/qwen3.6-35b-a3b",
@@ -13054,10 +13281,10 @@ export const MODELS = {
13054
13281
  reasoning: true,
13055
13282
  input: ["text"],
13056
13283
  cost: {
13057
- input: 1.25,
13058
- output: 3.75,
13059
- cacheRead: 0.25,
13060
- cacheWrite: 1.5625,
13284
+ input: 1.475,
13285
+ output: 4.425,
13286
+ cacheRead: 0.295,
13287
+ cacheWrite: 1.84375,
13061
13288
  },
13062
13289
  contextWindow: 1000000,
13063
13290
  maxTokens: 65536,
@@ -13190,13 +13417,13 @@ export const MODELS = {
13190
13417
  reasoning: true,
13191
13418
  input: ["text"],
13192
13419
  cost: {
13193
- input: 0.14,
13194
- output: 0.58,
13195
- cacheRead: 0.035,
13420
+ input: 0.19999999999999998,
13421
+ output: 0.7999999999999999,
13422
+ cacheRead: 0.049999999999999996,
13196
13423
  cacheWrite: 0,
13197
13424
  },
13198
13425
  contextWindow: 262144,
13199
- maxTokens: 4096,
13426
+ maxTokens: 131072,
13200
13427
  },
13201
13428
  "tencent/hy3:free": {
13202
13429
  id: "tencent/hy3:free",
@@ -13232,6 +13459,23 @@ export const MODELS = {
13232
13459
  contextWindow: 32768,
13233
13460
  maxTokens: 32768,
13234
13461
  },
13462
+ "thinkingmachines/inkling": {
13463
+ id: "thinkingmachines/inkling",
13464
+ name: "Thinking Machines: Inkling",
13465
+ api: "openai-completions",
13466
+ provider: "openrouter",
13467
+ baseUrl: "https://openrouter.ai/api/v1",
13468
+ reasoning: true,
13469
+ input: ["text", "image"],
13470
+ cost: {
13471
+ input: 1,
13472
+ output: 4.05,
13473
+ cacheRead: 0.16999999999999998,
13474
+ cacheWrite: 0,
13475
+ },
13476
+ contextWindow: 1048576,
13477
+ maxTokens: 4096,
13478
+ },
13235
13479
  "upstage/solar-pro-3": {
13236
13480
  id: "upstage/solar-pro-3",
13237
13481
  name: "Upstage: Solar Pro 3",
@@ -13326,13 +13570,13 @@ export const MODELS = {
13326
13570
  reasoning: true,
13327
13571
  input: ["text", "image"],
13328
13572
  cost: {
13329
- input: 0.105,
13573
+ input: 0.14,
13330
13574
  output: 0.28,
13331
- cacheRead: 0.028,
13575
+ cacheRead: 0.0028,
13332
13576
  cacheWrite: 0,
13333
13577
  },
13334
13578
  contextWindow: 1048576,
13335
- maxTokens: 4096,
13579
+ maxTokens: 131072,
13336
13580
  },
13337
13581
  "xiaomi/mimo-v2.5-pro": {
13338
13582
  id: "xiaomi/mimo-v2.5-pro",
@@ -13411,13 +13655,13 @@ export const MODELS = {
13411
13655
  reasoning: true,
13412
13656
  input: ["text"],
13413
13657
  cost: {
13414
- input: 0.43,
13415
- output: 1.75,
13416
- cacheRead: 0.08,
13658
+ input: 0.5,
13659
+ output: 2,
13660
+ cacheRead: 0.09999999999999999,
13417
13661
  cacheWrite: 0,
13418
13662
  },
13419
- contextWindow: 200000,
13420
- maxTokens: 16384,
13663
+ contextWindow: 202752,
13664
+ maxTokens: 131072,
13421
13665
  },
13422
13666
  "z-ai/glm-4.6v": {
13423
13667
  id: "z-ai/glm-4.6v",
@@ -13462,13 +13706,13 @@ export const MODELS = {
13462
13706
  reasoning: true,
13463
13707
  input: ["text"],
13464
13708
  cost: {
13465
- input: 0.06,
13709
+ input: 0.060500000000000005,
13466
13710
  output: 0.39999999999999997,
13467
- cacheRead: 0.01,
13711
+ cacheRead: 0,
13468
13712
  cacheWrite: 0,
13469
13713
  },
13470
- contextWindow: 202752,
13471
- maxTokens: 16384,
13714
+ contextWindow: 200000,
13715
+ maxTokens: 131072,
13472
13716
  },
13473
13717
  "z-ai/glm-5": {
13474
13718
  id: "z-ai/glm-5",
@@ -13485,7 +13729,7 @@ export const MODELS = {
13485
13729
  cacheWrite: 0,
13486
13730
  },
13487
13731
  contextWindow: 202752,
13488
- maxTokens: 4096,
13732
+ maxTokens: 202752,
13489
13733
  },
13490
13734
  "z-ai/glm-5-turbo": {
13491
13735
  id: "z-ai/glm-5-turbo",
@@ -13501,7 +13745,7 @@ export const MODELS = {
13501
13745
  cacheRead: 0.24,
13502
13746
  cacheWrite: 0,
13503
13747
  },
13504
- contextWindow: 262144,
13748
+ contextWindow: 202752,
13505
13749
  maxTokens: 131072,
13506
13750
  },
13507
13751
  "z-ai/glm-5.1": {
@@ -13530,9 +13774,9 @@ export const MODELS = {
13530
13774
  reasoning: true,
13531
13775
  input: ["text"],
13532
13776
  cost: {
13533
- input: 0.42,
13534
- output: 1.32,
13535
- cacheRead: 0.078,
13777
+ input: 0.34299999999999997,
13778
+ output: 1.078,
13779
+ cacheRead: 0.06369999999999999,
13536
13780
  cacheWrite: 0,
13537
13781
  },
13538
13782
  contextWindow: 1048576,
@@ -13667,13 +13911,13 @@ export const MODELS = {
13667
13911
  reasoning: true,
13668
13912
  input: ["text", "image"],
13669
13913
  cost: {
13670
- input: 0.66,
13671
- output: 3.41,
13672
- cacheRead: 0.15,
13914
+ input: 3,
13915
+ output: 15,
13916
+ cacheRead: 0.3,
13673
13917
  cacheWrite: 0,
13674
13918
  },
13675
- contextWindow: 262144,
13676
- maxTokens: 262144,
13919
+ contextWindow: 1048576,
13920
+ maxTokens: 4096,
13677
13921
  },
13678
13922
  "~openai/gpt-latest": {
13679
13923
  id: "~openai/gpt-latest",
@@ -14046,6 +14290,25 @@ export const MODELS = {
14046
14290
  contextWindow: 131072,
14047
14291
  maxTokens: 131072,
14048
14292
  },
14293
+ "thinkingmachines/Inkling": {
14294
+ id: "thinkingmachines/Inkling",
14295
+ name: "Inkling",
14296
+ api: "openai-completions",
14297
+ provider: "together",
14298
+ baseUrl: "https://api.together.ai/v1",
14299
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
14300
+ reasoning: true,
14301
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
14302
+ input: ["text", "image"],
14303
+ cost: {
14304
+ input: 1,
14305
+ output: 4.05,
14306
+ cacheRead: 0.17,
14307
+ cacheWrite: 0,
14308
+ },
14309
+ contextWindow: 524288,
14310
+ maxTokens: 131072,
14311
+ },
14049
14312
  "zai-org/GLM-5": {
14050
14313
  id: "zai-org/GLM-5",
14051
14314
  name: "GLM-5",
@@ -14648,7 +14911,7 @@ export const MODELS = {
14648
14911
  cacheWrite: 18.75,
14649
14912
  },
14650
14913
  contextWindow: 200000,
14651
- maxTokens: 32000,
14914
+ maxTokens: 8192,
14652
14915
  },
14653
14916
  "anthropic/claude-opus-4.1": {
14654
14917
  id: "anthropic/claude-opus-4.1",
@@ -14720,6 +14983,24 @@ export const MODELS = {
14720
14983
  contextWindow: 1000000,
14721
14984
  maxTokens: 128000,
14722
14985
  },
14986
+ "anthropic/claude-opus-4.7-fast": {
14987
+ id: "anthropic/claude-opus-4.7-fast",
14988
+ name: "Claude Opus 4.7 (Fast)",
14989
+ api: "anthropic-messages",
14990
+ provider: "vercel-ai-gateway",
14991
+ baseUrl: "https://ai-gateway.vercel.sh",
14992
+ reasoning: true,
14993
+ thinkingLevelMap: { "xhigh": "xhigh" },
14994
+ input: ["text", "image"],
14995
+ cost: {
14996
+ input: 30,
14997
+ output: 150,
14998
+ cacheRead: 3,
14999
+ cacheWrite: 37.5,
15000
+ },
15001
+ contextWindow: 1000000,
15002
+ maxTokens: 128000,
15003
+ },
14723
15004
  "anthropic/claude-opus-4.8": {
14724
15005
  id: "anthropic/claude-opus-4.8",
14725
15006
  name: "Claude Opus 4.8",
@@ -14738,6 +15019,24 @@ export const MODELS = {
14738
15019
  contextWindow: 1000000,
14739
15020
  maxTokens: 128000,
14740
15021
  },
15022
+ "anthropic/claude-opus-4.8-fast": {
15023
+ id: "anthropic/claude-opus-4.8-fast",
15024
+ name: "Claude Opus 4.8 (Fast)",
15025
+ api: "anthropic-messages",
15026
+ provider: "vercel-ai-gateway",
15027
+ baseUrl: "https://ai-gateway.vercel.sh",
15028
+ reasoning: true,
15029
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
15030
+ input: ["text", "image"],
15031
+ cost: {
15032
+ input: 10,
15033
+ output: 50,
15034
+ cacheRead: 1,
15035
+ cacheWrite: 12.5,
15036
+ },
15037
+ contextWindow: 1000000,
15038
+ maxTokens: 128000,
15039
+ },
14741
15040
  "anthropic/claude-sonnet-4": {
14742
15041
  id: "anthropic/claude-sonnet-4",
14743
15042
  name: "Claude Sonnet 4",
@@ -14753,7 +15052,7 @@ export const MODELS = {
14753
15052
  cacheWrite: 3.75,
14754
15053
  },
14755
15054
  contextWindow: 1000000,
14756
- maxTokens: 64000,
15055
+ maxTokens: 8192,
14757
15056
  },
14758
15057
  "anthropic/claude-sonnet-4.5": {
14759
15058
  id: "anthropic/claude-sonnet-4.5",
@@ -14806,23 +15105,6 @@ export const MODELS = {
14806
15105
  contextWindow: 1000000,
14807
15106
  maxTokens: 128000,
14808
15107
  },
14809
- "arcee-ai/trinity-large-preview": {
14810
- id: "arcee-ai/trinity-large-preview",
14811
- name: "Trinity Large Preview",
14812
- api: "anthropic-messages",
14813
- provider: "vercel-ai-gateway",
14814
- baseUrl: "https://ai-gateway.vercel.sh",
14815
- reasoning: false,
14816
- input: ["text"],
14817
- cost: {
14818
- input: 0.25,
14819
- output: 1,
14820
- cacheRead: 0,
14821
- cacheWrite: 0,
14822
- },
14823
- contextWindow: 131000,
14824
- maxTokens: 131000,
14825
- },
14826
15108
  "arcee-ai/trinity-large-thinking": {
14827
15109
  id: "arcee-ai/trinity-large-thinking",
14828
15110
  name: "Trinity Large Thinking",
@@ -14951,12 +15233,12 @@ export const MODELS = {
14951
15233
  reasoning: true,
14952
15234
  input: ["text"],
14953
15235
  cost: {
14954
- input: 0.6,
14955
- output: 1.7,
14956
- cacheRead: 0,
15236
+ input: 0.25,
15237
+ output: 0.95,
15238
+ cacheRead: 0.13,
14957
15239
  cacheWrite: 0,
14958
15240
  },
14959
- contextWindow: 128000,
15241
+ contextWindow: 163840,
14960
15242
  maxTokens: 128000,
14961
15243
  },
14962
15244
  "deepseek/deepseek-v3.1-terminus": {
@@ -15350,40 +15632,6 @@ export const MODELS = {
15350
15632
  contextWindow: 256000,
15351
15633
  maxTokens: 80000,
15352
15634
  },
15353
- "meituan/longcat-flash-chat": {
15354
- id: "meituan/longcat-flash-chat",
15355
- name: "LongCat Flash Chat",
15356
- api: "anthropic-messages",
15357
- provider: "vercel-ai-gateway",
15358
- baseUrl: "https://ai-gateway.vercel.sh",
15359
- reasoning: false,
15360
- input: ["text"],
15361
- cost: {
15362
- input: 0,
15363
- output: 0,
15364
- cacheRead: 0,
15365
- cacheWrite: 0,
15366
- },
15367
- contextWindow: 128000,
15368
- maxTokens: 100000,
15369
- },
15370
- "meituan/longcat-flash-thinking-2601": {
15371
- id: "meituan/longcat-flash-thinking-2601",
15372
- name: "LongCat Flash Thinking 2601",
15373
- api: "anthropic-messages",
15374
- provider: "vercel-ai-gateway",
15375
- baseUrl: "https://ai-gateway.vercel.sh",
15376
- reasoning: true,
15377
- input: ["text"],
15378
- cost: {
15379
- input: 0,
15380
- output: 0,
15381
- cacheRead: 0,
15382
- cacheWrite: 0,
15383
- },
15384
- contextWindow: 32768,
15385
- maxTokens: 32768,
15386
- },
15387
15635
  "meta/llama-3.1-70b": {
15388
15636
  id: "meta/llama-3.1-70b",
15389
15637
  name: "Llama 3.1 70B Instruct",
@@ -15418,40 +15666,6 @@ export const MODELS = {
15418
15666
  contextWindow: 128000,
15419
15667
  maxTokens: 8192,
15420
15668
  },
15421
- "meta/llama-3.2-11b": {
15422
- id: "meta/llama-3.2-11b",
15423
- name: "Llama 3.2 11B Vision Instruct",
15424
- api: "anthropic-messages",
15425
- provider: "vercel-ai-gateway",
15426
- baseUrl: "https://ai-gateway.vercel.sh",
15427
- reasoning: false,
15428
- input: ["text", "image"],
15429
- cost: {
15430
- input: 0.16,
15431
- output: 0.16,
15432
- cacheRead: 0,
15433
- cacheWrite: 0,
15434
- },
15435
- contextWindow: 128000,
15436
- maxTokens: 8192,
15437
- },
15438
- "meta/llama-3.2-90b": {
15439
- id: "meta/llama-3.2-90b",
15440
- name: "Llama 3.2 90B Vision Instruct",
15441
- api: "anthropic-messages",
15442
- provider: "vercel-ai-gateway",
15443
- baseUrl: "https://ai-gateway.vercel.sh",
15444
- reasoning: false,
15445
- input: ["text", "image"],
15446
- cost: {
15447
- input: 0.72,
15448
- output: 0.72,
15449
- cacheRead: 0,
15450
- cacheWrite: 0,
15451
- },
15452
- contextWindow: 128000,
15453
- maxTokens: 8192,
15454
- },
15455
15669
  "meta/llama-3.3-70b": {
15456
15670
  id: "meta/llama-3.3-70b",
15457
15671
  name: "Llama 3.3 70B Instruct",
@@ -15690,23 +15904,6 @@ export const MODELS = {
15690
15904
  contextWindow: 256000,
15691
15905
  maxTokens: 256000,
15692
15906
  },
15693
- "mistral/devstral-small": {
15694
- id: "mistral/devstral-small",
15695
- name: "Devstral Small 1.1",
15696
- api: "anthropic-messages",
15697
- provider: "vercel-ai-gateway",
15698
- baseUrl: "https://ai-gateway.vercel.sh",
15699
- reasoning: false,
15700
- input: ["text"],
15701
- cost: {
15702
- input: 0.09999999999999999,
15703
- output: 0.3,
15704
- cacheRead: 0,
15705
- cacheWrite: 0,
15706
- },
15707
- contextWindow: 128000,
15708
- maxTokens: 64000,
15709
- },
15710
15907
  "mistral/devstral-small-2": {
15711
15908
  id: "mistral/devstral-small-2",
15712
15909
  name: "Devstral Small 2",
@@ -15911,23 +16108,6 @@ export const MODELS = {
15911
16108
  contextWindow: 128000,
15912
16109
  maxTokens: 4000,
15913
16110
  },
15914
- "mistral/pixtral-large": {
15915
- id: "mistral/pixtral-large",
15916
- name: "Pixtral Large",
15917
- api: "anthropic-messages",
15918
- provider: "vercel-ai-gateway",
15919
- baseUrl: "https://ai-gateway.vercel.sh",
15920
- reasoning: false,
15921
- input: ["text", "image"],
15922
- cost: {
15923
- input: 2,
15924
- output: 6,
15925
- cacheRead: 0,
15926
- cacheWrite: 0,
15927
- },
15928
- contextWindow: 128000,
15929
- maxTokens: 4000,
15930
- },
15931
16111
  "moonshotai/kimi-k2": {
15932
16112
  id: "moonshotai/kimi-k2",
15933
16113
  name: "Kimi K2 Instruct",
@@ -16030,6 +16210,23 @@ export const MODELS = {
16030
16210
  contextWindow: 262144,
16031
16211
  maxTokens: 32768,
16032
16212
  },
16213
+ "moonshotai/kimi-k3": {
16214
+ id: "moonshotai/kimi-k3",
16215
+ name: "Kimi K3",
16216
+ api: "anthropic-messages",
16217
+ provider: "vercel-ai-gateway",
16218
+ baseUrl: "https://ai-gateway.vercel.sh",
16219
+ reasoning: true,
16220
+ input: ["text", "image"],
16221
+ cost: {
16222
+ input: 3,
16223
+ output: 15,
16224
+ cacheRead: 0.3,
16225
+ cacheWrite: 0,
16226
+ },
16227
+ contextWindow: 1000000,
16228
+ maxTokens: 131072,
16229
+ },
16033
16230
  "nvidia/nemotron-3-nano-30b-a3b": {
16034
16231
  id: "nvidia/nemotron-3-nano-30b-a3b",
16035
16232
  name: "Nemotron 3 Nano 30B A3B",
@@ -16875,6 +17072,23 @@ export const MODELS = {
16875
17072
  contextWindow: 256000,
16876
17073
  maxTokens: 256000,
16877
17074
  },
17075
+ "thinkingmachines/inkling": {
17076
+ id: "thinkingmachines/inkling",
17077
+ name: "Inkling",
17078
+ api: "anthropic-messages",
17079
+ provider: "vercel-ai-gateway",
17080
+ baseUrl: "https://ai-gateway.vercel.sh",
17081
+ reasoning: true,
17082
+ input: ["text", "image"],
17083
+ cost: {
17084
+ input: 1,
17085
+ output: 4.05,
17086
+ cacheRead: 0.16999999999999998,
17087
+ cacheWrite: 0,
17088
+ },
17089
+ contextWindow: 256000,
17090
+ maxTokens: 256000,
17091
+ },
16878
17092
  "xai/grok-4.1-fast-non-reasoning": {
16879
17093
  id: "xai/grok-4.1-fast-non-reasoning",
16880
17094
  name: "Grok 4.1 Fast Non-Reasoning",