@elyracode/ai 0.9.14 → 0.9.16

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",
@@ -2742,6 +2793,23 @@ export const MODELS = {
2742
2793
  contextWindow: 1050000,
2743
2794
  maxTokens: 128000,
2744
2795
  },
2796
+ "gpt-realtime-2.1": {
2797
+ id: "gpt-realtime-2.1",
2798
+ name: "GPT-Realtime-2.1",
2799
+ api: "azure-openai-responses",
2800
+ provider: "azure-openai-responses",
2801
+ baseUrl: "",
2802
+ reasoning: true,
2803
+ input: ["text", "image"],
2804
+ cost: {
2805
+ input: 4,
2806
+ output: 24,
2807
+ cacheRead: 0.4,
2808
+ cacheWrite: 0,
2809
+ },
2810
+ contextWindow: 128000,
2811
+ maxTokens: 32000,
2812
+ },
2745
2813
  "o1": {
2746
2814
  id: "o1",
2747
2815
  name: "o1",
@@ -2925,7 +2993,7 @@ export const MODELS = {
2925
2993
  cost: {
2926
2994
  input: 2.25,
2927
2995
  output: 2.75,
2928
- cacheRead: 0,
2996
+ cacheRead: 2.25,
2929
2997
  cacheWrite: 0,
2930
2998
  },
2931
2999
  contextWindow: 131072,
@@ -3386,6 +3454,60 @@ export const MODELS = {
3386
3454
  contextWindow: 1050000,
3387
3455
  maxTokens: 128000,
3388
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
+ },
3389
3511
  "o1": {
3390
3512
  id: "o1",
3391
3513
  name: "o1",
@@ -3543,6 +3665,24 @@ export const MODELS = {
3543
3665
  contextWindow: 131072,
3544
3666
  maxTokens: 131072,
3545
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
+ },
3546
3686
  },
3547
3687
  "cloudflare-workers-ai": {
3548
3688
  "@cf/google/gemma-4-26b-a4b-it": {
@@ -4338,7 +4478,7 @@ export const MODELS = {
4338
4478
  cacheRead: 0.2,
4339
4479
  cacheWrite: 0,
4340
4480
  },
4341
- contextWindow: 200000,
4481
+ contextWindow: 1000000,
4342
4482
  maxTokens: 64000,
4343
4483
  },
4344
4484
  "gemini-3.5-flash": {
@@ -4471,7 +4611,7 @@ export const MODELS = {
4471
4611
  cacheRead: 0.25,
4472
4612
  cacheWrite: 0,
4473
4613
  },
4474
- contextWindow: 400000,
4614
+ contextWindow: 1050000,
4475
4615
  maxTokens: 128000,
4476
4616
  },
4477
4617
  "gpt-5.4-mini": {
@@ -4528,7 +4668,64 @@ export const MODELS = {
4528
4668
  cacheRead: 0.5,
4529
4669
  cacheWrite: 0,
4530
4670
  },
4531
- contextWindow: 400000,
4671
+ contextWindow: 1050000,
4672
+ maxTokens: 128000,
4673
+ },
4674
+ "gpt-5.6-luna": {
4675
+ id: "gpt-5.6-luna",
4676
+ name: "GPT-5.6 Luna",
4677
+ api: "openai-responses",
4678
+ provider: "github-copilot",
4679
+ baseUrl: "https://api.individual.githubcopilot.com",
4680
+ 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" },
4681
+ reasoning: true,
4682
+ thinkingLevelMap: { "off": null },
4683
+ input: ["text", "image"],
4684
+ cost: {
4685
+ input: 1,
4686
+ output: 6,
4687
+ cacheRead: 0.1,
4688
+ cacheWrite: 0,
4689
+ },
4690
+ contextWindow: 1050000,
4691
+ maxTokens: 128000,
4692
+ },
4693
+ "gpt-5.6-sol": {
4694
+ id: "gpt-5.6-sol",
4695
+ name: "GPT-5.6 Sol",
4696
+ api: "openai-responses",
4697
+ provider: "github-copilot",
4698
+ baseUrl: "https://api.individual.githubcopilot.com",
4699
+ 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" },
4700
+ reasoning: true,
4701
+ thinkingLevelMap: { "off": null },
4702
+ input: ["text", "image"],
4703
+ cost: {
4704
+ input: 5,
4705
+ output: 30,
4706
+ cacheRead: 0.5,
4707
+ cacheWrite: 0,
4708
+ },
4709
+ contextWindow: 1050000,
4710
+ maxTokens: 128000,
4711
+ },
4712
+ "gpt-5.6-terra": {
4713
+ id: "gpt-5.6-terra",
4714
+ name: "GPT-5.6 Terra",
4715
+ api: "openai-responses",
4716
+ provider: "github-copilot",
4717
+ baseUrl: "https://api.individual.githubcopilot.com",
4718
+ 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" },
4719
+ reasoning: true,
4720
+ thinkingLevelMap: { "off": null },
4721
+ input: ["text", "image"],
4722
+ cost: {
4723
+ input: 2.5,
4724
+ output: 15,
4725
+ cacheRead: 0.25,
4726
+ cacheWrite: 0,
4727
+ },
4728
+ contextWindow: 1050000,
4532
4729
  maxTokens: 128000,
4533
4730
  },
4534
4731
  "kimi-k2.7-code": {
@@ -7677,6 +7874,23 @@ export const MODELS = {
7677
7874
  contextWindow: 1050000,
7678
7875
  maxTokens: 128000,
7679
7876
  },
7877
+ "gpt-realtime-2.1": {
7878
+ id: "gpt-realtime-2.1",
7879
+ name: "GPT-Realtime-2.1",
7880
+ api: "openai-responses",
7881
+ provider: "openai",
7882
+ baseUrl: "https://api.openai.com/v1",
7883
+ reasoning: true,
7884
+ input: ["text", "image"],
7885
+ cost: {
7886
+ input: 4,
7887
+ output: 24,
7888
+ cacheRead: 0.4,
7889
+ cacheWrite: 0,
7890
+ },
7891
+ contextWindow: 128000,
7892
+ maxTokens: 32000,
7893
+ },
7680
7894
  "o1": {
7681
7895
  id: "o1",
7682
7896
  name: "o1",
@@ -8653,6 +8867,60 @@ export const MODELS = {
8653
8867
  contextWindow: 1050000,
8654
8868
  maxTokens: 128000,
8655
8869
  },
8870
+ "gpt-5.6-luna": {
8871
+ id: "gpt-5.6-luna",
8872
+ name: "GPT-5.6 Luna",
8873
+ api: "openai-responses",
8874
+ provider: "opencode",
8875
+ baseUrl: "https://opencode.ai/zen/v1",
8876
+ reasoning: true,
8877
+ thinkingLevelMap: { "off": null },
8878
+ input: ["text", "image"],
8879
+ cost: {
8880
+ input: 1,
8881
+ output: 6,
8882
+ cacheRead: 0.1,
8883
+ cacheWrite: 1.25,
8884
+ },
8885
+ contextWindow: 1050000,
8886
+ maxTokens: 128000,
8887
+ },
8888
+ "gpt-5.6-sol": {
8889
+ id: "gpt-5.6-sol",
8890
+ name: "GPT-5.6 Sol",
8891
+ api: "openai-responses",
8892
+ provider: "opencode",
8893
+ baseUrl: "https://opencode.ai/zen/v1",
8894
+ reasoning: true,
8895
+ thinkingLevelMap: { "off": null },
8896
+ input: ["text", "image"],
8897
+ cost: {
8898
+ input: 5,
8899
+ output: 30,
8900
+ cacheRead: 0.5,
8901
+ cacheWrite: 6.25,
8902
+ },
8903
+ contextWindow: 1050000,
8904
+ maxTokens: 128000,
8905
+ },
8906
+ "gpt-5.6-terra": {
8907
+ id: "gpt-5.6-terra",
8908
+ name: "GPT-5.6 Terra",
8909
+ api: "openai-responses",
8910
+ provider: "opencode",
8911
+ baseUrl: "https://opencode.ai/zen/v1",
8912
+ reasoning: true,
8913
+ thinkingLevelMap: { "off": null },
8914
+ input: ["text", "image"],
8915
+ cost: {
8916
+ input: 2.5,
8917
+ output: 15,
8918
+ cacheRead: 0.25,
8919
+ cacheWrite: 3.125,
8920
+ },
8921
+ contextWindow: 1050000,
8922
+ maxTokens: 128000,
8923
+ },
8656
8924
  "grok-4.5": {
8657
8925
  id: "grok-4.5",
8658
8926
  name: "Grok 4.5",
@@ -8701,7 +8969,7 @@ export const MODELS = {
8701
8969
  cacheRead: 0,
8702
8970
  cacheWrite: 0,
8703
8971
  },
8704
- contextWindow: 256000,
8972
+ contextWindow: 190000,
8705
8973
  maxTokens: 64000,
8706
8974
  },
8707
8975
  "kimi-k2.5": {
@@ -9552,23 +9820,6 @@ export const MODELS = {
9552
9820
  contextWindow: 262144,
9553
9821
  maxTokens: 80000,
9554
9822
  },
9555
- "arcee-ai/trinity-mini": {
9556
- id: "arcee-ai/trinity-mini",
9557
- name: "Arcee AI: Trinity Mini",
9558
- api: "openai-completions",
9559
- provider: "openrouter",
9560
- baseUrl: "https://openrouter.ai/api/v1",
9561
- reasoning: true,
9562
- input: ["text"],
9563
- cost: {
9564
- input: 0.045,
9565
- output: 0.15,
9566
- cacheRead: 0,
9567
- cacheWrite: 0,
9568
- },
9569
- contextWindow: 131072,
9570
- maxTokens: 131072,
9571
- },
9572
9823
  "arcee-ai/virtuoso-large": {
9573
9824
  id: "arcee-ai/virtuoso-large",
9574
9825
  name: "Arcee AI: Virtuoso Large",
@@ -9748,13 +9999,13 @@ export const MODELS = {
9748
9999
  reasoning: false,
9749
10000
  input: ["text"],
9750
10001
  cost: {
9751
- input: 0.24,
9752
- output: 0.8999999999999999,
10002
+ input: 0.27,
10003
+ output: 1.12,
9753
10004
  cacheRead: 0.135,
9754
10005
  cacheWrite: 0,
9755
10006
  },
9756
10007
  contextWindow: 163840,
9757
- maxTokens: 16384,
10008
+ maxTokens: 65536,
9758
10009
  },
9759
10010
  "deepseek/deepseek-chat-v3.1": {
9760
10011
  id: "deepseek/deepseek-chat-v3.1",
@@ -9765,8 +10016,8 @@ export const MODELS = {
9765
10016
  reasoning: true,
9766
10017
  input: ["text"],
9767
10018
  cost: {
9768
- input: 0.21,
9769
- output: 0.7899999999999999,
10019
+ input: 0.25,
10020
+ output: 0.95,
9770
10021
  cacheRead: 0.13,
9771
10022
  cacheWrite: 0,
9772
10023
  },
@@ -9817,11 +10068,11 @@ export const MODELS = {
9817
10068
  input: ["text"],
9818
10069
  cost: {
9819
10070
  input: 0.27,
9820
- output: 0.95,
9821
- cacheRead: 0.13,
10071
+ output: 1,
10072
+ cacheRead: 0.135,
9822
10073
  cacheWrite: 0,
9823
10074
  },
9824
- contextWindow: 163840,
10075
+ contextWindow: 131072,
9825
10076
  maxTokens: 32768,
9826
10077
  },
9827
10078
  "deepseek/deepseek-v3.2": {
@@ -9833,13 +10084,13 @@ export const MODELS = {
9833
10084
  reasoning: true,
9834
10085
  input: ["text"],
9835
10086
  cost: {
9836
- input: 0.2288,
9837
- output: 0.3432,
9838
- cacheRead: 0.02288,
10087
+ input: 0.26899999999999996,
10088
+ output: 0.39999999999999997,
10089
+ cacheRead: 0.13449999999999998,
9839
10090
  cacheWrite: 0,
9840
10091
  },
9841
- contextWindow: 131072,
9842
- maxTokens: 64000,
10092
+ contextWindow: 163840,
10093
+ maxTokens: 65536,
9843
10094
  },
9844
10095
  "deepseek/deepseek-v3.2-exp": {
9845
10096
  id: "deepseek/deepseek-v3.2-exp",
@@ -9869,13 +10120,13 @@ export const MODELS = {
9869
10120
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
9870
10121
  input: ["text"],
9871
10122
  cost: {
9872
- input: 0.09,
9873
- output: 0.18,
9874
- cacheRead: 0.018,
10123
+ input: 0.098,
10124
+ output: 0.196,
10125
+ cacheRead: 0.02,
9875
10126
  cacheWrite: 0,
9876
10127
  },
9877
10128
  contextWindow: 1048576,
9878
- maxTokens: 65536,
10129
+ maxTokens: 4096,
9879
10130
  },
9880
10131
  "deepseek/deepseek-v4-pro": {
9881
10132
  id: "deepseek/deepseek-v4-pro",
@@ -10127,12 +10378,12 @@ export const MODELS = {
10127
10378
  input: ["text", "image"],
10128
10379
  cost: {
10129
10380
  input: 0.08,
10130
- output: 0.16,
10131
- cacheRead: 0,
10381
+ output: 0.44999999999999996,
10382
+ cacheRead: 0.04,
10132
10383
  cacheWrite: 0,
10133
10384
  },
10134
10385
  contextWindow: 131072,
10135
- maxTokens: 16384,
10386
+ maxTokens: 131072,
10136
10387
  },
10137
10388
  "google/gemma-4-26b-a4b-it": {
10138
10389
  id: "google/gemma-4-26b-a4b-it",
@@ -10143,13 +10394,13 @@ export const MODELS = {
10143
10394
  reasoning: true,
10144
10395
  input: ["text", "image"],
10145
10396
  cost: {
10146
- input: 0.06,
10147
- output: 0.33,
10397
+ input: 0.09999999999999999,
10398
+ output: 0.3,
10148
10399
  cacheRead: 0,
10149
10400
  cacheWrite: 0,
10150
10401
  },
10151
10402
  contextWindow: 262144,
10152
- maxTokens: 4096,
10403
+ maxTokens: 256000,
10153
10404
  },
10154
10405
  "google/gemma-4-26b-a4b-it:free": {
10155
10406
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10178,12 +10429,12 @@ export const MODELS = {
10178
10429
  input: ["text", "image"],
10179
10430
  cost: {
10180
10431
  input: 0.12,
10181
- output: 0.35,
10182
- cacheRead: 0.09,
10432
+ output: 0.37,
10433
+ cacheRead: 0,
10183
10434
  cacheWrite: 0,
10184
10435
  },
10185
10436
  contextWindow: 262144,
10186
- maxTokens: 262144,
10437
+ maxTokens: 16384,
10187
10438
  },
10188
10439
  "google/gemma-4-31b-it:free": {
10189
10440
  id: "google/gemma-4-31b-it:free",
@@ -10200,7 +10451,7 @@ export const MODELS = {
10200
10451
  cacheWrite: 0,
10201
10452
  },
10202
10453
  contextWindow: 262144,
10203
- maxTokens: 8192,
10454
+ maxTokens: 32768,
10204
10455
  },
10205
10456
  "ibm-granite/granite-4.1-8b": {
10206
10457
  id: "ibm-granite/granite-4.1-8b",
@@ -10287,6 +10538,23 @@ export const MODELS = {
10287
10538
  contextWindow: 262144,
10288
10539
  maxTokens: 65536,
10289
10540
  },
10541
+ "kwaipilot/kat-coder-air-v2.5": {
10542
+ id: "kwaipilot/kat-coder-air-v2.5",
10543
+ name: "Kwaipilot: KAT-Coder-Air V2.5",
10544
+ api: "openai-completions",
10545
+ provider: "openrouter",
10546
+ baseUrl: "https://openrouter.ai/api/v1",
10547
+ reasoning: false,
10548
+ input: ["text"],
10549
+ cost: {
10550
+ input: 0.15,
10551
+ output: 0.6,
10552
+ cacheRead: 0.03,
10553
+ cacheWrite: 0,
10554
+ },
10555
+ contextWindow: 256000,
10556
+ maxTokens: 80000,
10557
+ },
10290
10558
  "kwaipilot/kat-coder-pro-v2": {
10291
10559
  id: "kwaipilot/kat-coder-pro-v2",
10292
10560
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -10304,22 +10572,22 @@ export const MODELS = {
10304
10572
  contextWindow: 256000,
10305
10573
  maxTokens: 80000,
10306
10574
  },
10307
- "liquid/lfm-2.5-1.2b-thinking:free": {
10308
- id: "liquid/lfm-2.5-1.2b-thinking:free",
10309
- name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
10575
+ "kwaipilot/kat-coder-pro-v2.5": {
10576
+ id: "kwaipilot/kat-coder-pro-v2.5",
10577
+ name: "Kwaipilot: KAT-Coder-Pro V2.5",
10310
10578
  api: "openai-completions",
10311
10579
  provider: "openrouter",
10312
10580
  baseUrl: "https://openrouter.ai/api/v1",
10313
- reasoning: true,
10581
+ reasoning: false,
10314
10582
  input: ["text"],
10315
10583
  cost: {
10316
- input: 0,
10317
- output: 0,
10318
- cacheRead: 0,
10584
+ input: 0.74,
10585
+ output: 2.96,
10586
+ cacheRead: 0.15,
10319
10587
  cacheWrite: 0,
10320
10588
  },
10321
- contextWindow: 32768,
10322
- maxTokens: 4096,
10589
+ contextWindow: 256000,
10590
+ maxTokens: 80000,
10323
10591
  },
10324
10592
  "meta-llama/llama-3.1-70b-instruct": {
10325
10593
  id: "meta-llama/llama-3.1-70b-instruct",
@@ -10347,13 +10615,13 @@ export const MODELS = {
10347
10615
  reasoning: false,
10348
10616
  input: ["text"],
10349
10617
  cost: {
10350
- input: 0.02,
10351
- output: 0.03,
10352
- cacheRead: 0,
10618
+ input: 0.049999999999999996,
10619
+ output: 0.08,
10620
+ cacheRead: 0.024999999999999998,
10353
10621
  cacheWrite: 0,
10354
10622
  },
10355
10623
  contextWindow: 131072,
10356
- maxTokens: 16384,
10624
+ maxTokens: 131072,
10357
10625
  },
10358
10626
  "meta-llama/llama-3.3-70b-instruct": {
10359
10627
  id: "meta-llama/llama-3.3-70b-instruct",
@@ -10364,13 +10632,13 @@ export const MODELS = {
10364
10632
  reasoning: false,
10365
10633
  input: ["text"],
10366
10634
  cost: {
10367
- input: 0.09999999999999999,
10368
- output: 0.32,
10635
+ input: 0.13,
10636
+ output: 0.39999999999999997,
10369
10637
  cacheRead: 0,
10370
10638
  cacheWrite: 0,
10371
10639
  },
10372
10640
  contextWindow: 131072,
10373
- maxTokens: 16384,
10641
+ maxTokens: 128000,
10374
10642
  },
10375
10643
  "meta-llama/llama-3.3-70b-instruct:free": {
10376
10644
  id: "meta-llama/llama-3.3-70b-instruct:free",
@@ -10398,8 +10666,8 @@ export const MODELS = {
10398
10666
  reasoning: false,
10399
10667
  input: ["text", "image"],
10400
10668
  cost: {
10401
- input: 0.15,
10402
- output: 0.6,
10669
+ input: 0.19999999999999998,
10670
+ output: 0.7999999999999999,
10403
10671
  cacheRead: 0,
10404
10672
  cacheWrite: 0,
10405
10673
  },
@@ -10432,7 +10700,7 @@ export const MODELS = {
10432
10700
  reasoning: true,
10433
10701
  input: ["text"],
10434
10702
  cost: {
10435
- input: 0.39999999999999997,
10703
+ input: 0.55,
10436
10704
  output: 2.2,
10437
10705
  cacheRead: 0,
10438
10706
  cacheWrite: 0,
@@ -10500,13 +10768,13 @@ export const MODELS = {
10500
10768
  reasoning: true,
10501
10769
  input: ["text"],
10502
10770
  cost: {
10503
- input: 0.24,
10504
- output: 0.96,
10505
- cacheRead: 0,
10771
+ input: 0.3,
10772
+ output: 1.2,
10773
+ cacheRead: 0.06,
10506
10774
  cacheWrite: 0,
10507
10775
  },
10508
10776
  contextWindow: 204800,
10509
- maxTokens: 196608,
10777
+ maxTokens: 131072,
10510
10778
  },
10511
10779
  "minimax/minimax-m3": {
10512
10780
  id: "minimax/minimax-m3",
@@ -10722,12 +10990,12 @@ export const MODELS = {
10722
10990
  input: ["text"],
10723
10991
  cost: {
10724
10992
  input: 0.02,
10725
- output: 0.03,
10993
+ output: 0.04,
10726
10994
  cacheRead: 0,
10727
10995
  cacheWrite: 0,
10728
10996
  },
10729
10997
  contextWindow: 131072,
10730
- maxTokens: 4096,
10998
+ maxTokens: 16384,
10731
10999
  },
10732
11000
  "mistralai/mistral-saba": {
10733
11001
  id: "mistralai/mistral-saba",
@@ -10772,13 +11040,13 @@ export const MODELS = {
10772
11040
  reasoning: false,
10773
11041
  input: ["text", "image"],
10774
11042
  cost: {
10775
- input: 0.075,
10776
- output: 0.19999999999999998,
10777
- cacheRead: 0,
11043
+ input: 0.09999999999999999,
11044
+ output: 0.3,
11045
+ cacheRead: 0.01,
10778
11046
  cacheWrite: 0,
10779
11047
  },
10780
- contextWindow: 128000,
10781
- maxTokens: 16384,
11048
+ contextWindow: 131072,
11049
+ maxTokens: 4096,
10782
11050
  },
10783
11051
  "mistralai/mixtral-8x22b-instruct": {
10784
11052
  id: "mistralai/mixtral-8x22b-instruct",
@@ -10859,11 +11127,11 @@ export const MODELS = {
10859
11127
  cost: {
10860
11128
  input: 0.6,
10861
11129
  output: 2.5,
10862
- cacheRead: 0.15,
11130
+ cacheRead: 0,
10863
11131
  cacheWrite: 0,
10864
11132
  },
10865
11133
  contextWindow: 262144,
10866
- maxTokens: 100352,
11134
+ maxTokens: 262144,
10867
11135
  },
10868
11136
  "moonshotai/kimi-k2.5": {
10869
11137
  id: "moonshotai/kimi-k2.5",
@@ -10891,9 +11159,9 @@ export const MODELS = {
10891
11159
  reasoning: true,
10892
11160
  input: ["text", "image"],
10893
11161
  cost: {
10894
- input: 0.65,
11162
+ input: 0.66,
10895
11163
  output: 3.41,
10896
- cacheRead: 0.14,
11164
+ cacheRead: 0.144,
10897
11165
  cacheWrite: 0,
10898
11166
  },
10899
11167
  contextWindow: 262144,
@@ -10908,9 +11176,9 @@ export const MODELS = {
10908
11176
  reasoning: true,
10909
11177
  input: ["text", "image"],
10910
11178
  cost: {
10911
- input: 0.72,
11179
+ input: 0.719,
10912
11180
  output: 3.49,
10913
- cacheRead: 0.159,
11181
+ cacheRead: 0.149,
10914
11182
  cacheWrite: 0,
10915
11183
  },
10916
11184
  contextWindow: 262144,
@@ -11027,9 +11295,9 @@ export const MODELS = {
11027
11295
  reasoning: true,
11028
11296
  input: ["text"],
11029
11297
  cost: {
11030
- input: 0.08,
11031
- output: 0.44999999999999996,
11032
- cacheRead: 0,
11298
+ input: 0.21,
11299
+ output: 0.45499999999999996,
11300
+ cacheRead: 0.06,
11033
11301
  cacheWrite: 0,
11034
11302
  },
11035
11303
  contextWindow: 1000000,
@@ -11061,13 +11329,13 @@ export const MODELS = {
11061
11329
  reasoning: true,
11062
11330
  input: ["text"],
11063
11331
  cost: {
11064
- input: 0.5,
11065
- output: 2.2,
11066
- cacheRead: 0.09999999999999999,
11332
+ input: 0.6,
11333
+ output: 3.5999999999999996,
11334
+ cacheRead: 0.19999999999999998,
11067
11335
  cacheWrite: 0,
11068
11336
  },
11069
11337
  contextWindow: 1000000,
11070
- maxTokens: 16384,
11338
+ maxTokens: 4096,
11071
11339
  },
11072
11340
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
11073
11341
  id: "nvidia/nemotron-3-ultra-550b-a55b:free",
@@ -11203,7 +11471,7 @@ export const MODELS = {
11203
11471
  cacheWrite: 0,
11204
11472
  },
11205
11473
  contextWindow: 1047576,
11206
- maxTokens: 4096,
11474
+ maxTokens: 32768,
11207
11475
  },
11208
11476
  "openai/gpt-4.1-mini": {
11209
11477
  id: "openai/gpt-4.1-mini",
@@ -11250,7 +11518,7 @@ export const MODELS = {
11250
11518
  cost: {
11251
11519
  input: 2.5,
11252
11520
  output: 10,
11253
- cacheRead: 0,
11521
+ cacheRead: 1.25,
11254
11522
  cacheWrite: 0,
11255
11523
  },
11256
11524
  contextWindow: 128000,
@@ -11335,11 +11603,11 @@ export const MODELS = {
11335
11603
  cost: {
11336
11604
  input: 0.049999999999999996,
11337
11605
  output: 0.39999999999999997,
11338
- cacheRead: 0.01,
11606
+ cacheRead: 0.005,
11339
11607
  cacheWrite: 0,
11340
11608
  },
11341
11609
  contextWindow: 400000,
11342
- maxTokens: 4096,
11610
+ maxTokens: 128000,
11343
11611
  },
11344
11612
  "openai/gpt-5-pro": {
11345
11613
  id: "openai/gpt-5-pro",
@@ -11369,7 +11637,7 @@ export const MODELS = {
11369
11637
  cost: {
11370
11638
  input: 1.25,
11371
11639
  output: 10,
11372
- cacheRead: 0.13,
11640
+ cacheRead: 0.125,
11373
11641
  cacheWrite: 0,
11374
11642
  },
11375
11643
  contextWindow: 400000,
@@ -11386,11 +11654,11 @@ export const MODELS = {
11386
11654
  cost: {
11387
11655
  input: 1.25,
11388
11656
  output: 10,
11389
- cacheRead: 0.13,
11657
+ cacheRead: 0.125,
11390
11658
  cacheWrite: 0,
11391
11659
  },
11392
11660
  contextWindow: 128000,
11393
- maxTokens: 32000,
11661
+ maxTokens: 16384,
11394
11662
  },
11395
11663
  "openai/gpt-5.1-codex": {
11396
11664
  id: "openai/gpt-5.1-codex",
@@ -11403,7 +11671,7 @@ export const MODELS = {
11403
11671
  cost: {
11404
11672
  input: 1.25,
11405
11673
  output: 10,
11406
- cacheRead: 0.13,
11674
+ cacheRead: 0.125,
11407
11675
  cacheWrite: 0,
11408
11676
  },
11409
11677
  contextWindow: 400000,
@@ -11821,25 +12089,8 @@ export const MODELS = {
11821
12089
  reasoning: true,
11822
12090
  input: ["text"],
11823
12091
  cost: {
11824
- input: 0.036,
11825
- output: 0.18,
11826
- cacheRead: 0,
11827
- cacheWrite: 0,
11828
- },
11829
- contextWindow: 131072,
11830
- maxTokens: 4096,
11831
- },
11832
- "openai/gpt-oss-120b:free": {
11833
- id: "openai/gpt-oss-120b:free",
11834
- name: "OpenAI: gpt-oss-120b (free)",
11835
- api: "openai-completions",
11836
- provider: "openrouter",
11837
- baseUrl: "https://openrouter.ai/api/v1",
11838
- reasoning: true,
11839
- input: ["text"],
11840
- cost: {
11841
- input: 0,
11842
- output: 0,
12092
+ input: 0.037,
12093
+ output: 0.16999999999999998,
11843
12094
  cacheRead: 0,
11844
12095
  cacheWrite: 0,
11845
12096
  },
@@ -11855,13 +12106,13 @@ export const MODELS = {
11855
12106
  reasoning: true,
11856
12107
  input: ["text"],
11857
12108
  cost: {
11858
- input: 0.029,
11859
- output: 0.14,
11860
- cacheRead: 0,
12109
+ input: 0.03,
12110
+ output: 0.13,
12111
+ cacheRead: 0.03,
11861
12112
  cacheWrite: 0,
11862
12113
  },
11863
12114
  contextWindow: 131072,
11864
- maxTokens: 4096,
12115
+ maxTokens: 131072,
11865
12116
  },
11866
12117
  "openai/gpt-oss-20b:free": {
11867
12118
  id: "openai/gpt-oss-20b:free",
@@ -12281,7 +12532,7 @@ export const MODELS = {
12281
12532
  input: ["text"],
12282
12533
  cost: {
12283
12534
  input: 0.09,
12284
- output: 0.09999999999999999,
12535
+ output: 0.55,
12285
12536
  cacheRead: 0,
12286
12537
  cacheWrite: 0,
12287
12538
  },
@@ -12331,13 +12582,13 @@ export const MODELS = {
12331
12582
  reasoning: false,
12332
12583
  input: ["text"],
12333
12584
  cost: {
12334
- input: 0.04815,
12335
- output: 0.19305,
12585
+ input: 0.09999999999999999,
12586
+ output: 0.3,
12336
12587
  cacheRead: 0,
12337
12588
  cacheWrite: 0,
12338
12589
  },
12339
- contextWindow: 131072,
12340
- maxTokens: 32000,
12590
+ contextWindow: 262144,
12591
+ maxTokens: 4096,
12341
12592
  },
12342
12593
  "qwen/qwen3-30b-a3b-thinking-2507": {
12343
12594
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -12399,9 +12650,9 @@ export const MODELS = {
12399
12650
  reasoning: false,
12400
12651
  input: ["text"],
12401
12652
  cost: {
12402
- input: 0.22,
12403
- output: 1.7999999999999998,
12404
- cacheRead: 0,
12653
+ input: 0.3,
12654
+ output: 1,
12655
+ cacheRead: 0.09999999999999999,
12405
12656
  cacheWrite: 0,
12406
12657
  },
12407
12658
  contextWindow: 1048576,
@@ -12450,7 +12701,7 @@ export const MODELS = {
12450
12701
  reasoning: false,
12451
12702
  input: ["text"],
12452
12703
  cost: {
12453
- input: 0.11,
12704
+ input: 0.12,
12454
12705
  output: 0.7999999999999999,
12455
12706
  cacheRead: 0.07,
12456
12707
  cacheWrite: 0,
@@ -12535,13 +12786,13 @@ export const MODELS = {
12535
12786
  reasoning: false,
12536
12787
  input: ["text"],
12537
12788
  cost: {
12538
- input: 0.09,
12789
+ input: 0.09999999999999999,
12539
12790
  output: 1.1,
12540
- cacheRead: 0,
12791
+ cacheRead: 0.07,
12541
12792
  cacheWrite: 0,
12542
12793
  },
12543
12794
  contextWindow: 262144,
12544
- maxTokens: 16384,
12795
+ maxTokens: 262144,
12545
12796
  },
12546
12797
  "qwen/qwen3-next-80b-a3b-instruct:free": {
12547
12798
  id: "qwen/qwen3-next-80b-a3b-instruct:free",
@@ -12586,13 +12837,13 @@ export const MODELS = {
12586
12837
  reasoning: false,
12587
12838
  input: ["text", "image"],
12588
12839
  cost: {
12589
- input: 0.19999999999999998,
12590
- output: 0.88,
12591
- cacheRead: 0.11,
12840
+ input: 0.21,
12841
+ output: 1.9,
12842
+ cacheRead: 0.09999999999999999,
12592
12843
  cacheWrite: 0,
12593
12844
  },
12594
- contextWindow: 262144,
12595
- maxTokens: 16384,
12845
+ contextWindow: 131072,
12846
+ maxTokens: 32768,
12596
12847
  },
12597
12848
  "qwen/qwen3-vl-235b-a22b-thinking": {
12598
12849
  id: "qwen/qwen3-vl-235b-a22b-thinking",
@@ -12711,7 +12962,7 @@ export const MODELS = {
12711
12962
  cacheWrite: 0,
12712
12963
  },
12713
12964
  contextWindow: 262144,
12714
- maxTokens: 262144,
12965
+ maxTokens: 65536,
12715
12966
  },
12716
12967
  "qwen/qwen3.5-27b": {
12717
12968
  id: "qwen/qwen3.5-27b",
@@ -12741,11 +12992,11 @@ export const MODELS = {
12741
12992
  cost: {
12742
12993
  input: 0.14,
12743
12994
  output: 1,
12744
- cacheRead: 0.049999999999999996,
12995
+ cacheRead: 0,
12745
12996
  cacheWrite: 0,
12746
12997
  },
12747
12998
  contextWindow: 262144,
12748
- maxTokens: 81920,
12999
+ maxTokens: 262144,
12749
13000
  },
12750
13001
  "qwen/qwen3.5-397b-a17b": {
12751
13002
  id: "qwen/qwen3.5-397b-a17b",
@@ -12756,13 +13007,13 @@ export const MODELS = {
12756
13007
  reasoning: true,
12757
13008
  input: ["text", "image"],
12758
13009
  cost: {
12759
- input: 0.385,
12760
- output: 2.4499999999999997,
12761
- cacheRead: 0.111,
13010
+ input: 0.44999999999999996,
13011
+ output: 3,
13012
+ cacheRead: 0.22499999999999998,
12762
13013
  cacheWrite: 0,
12763
13014
  },
12764
- contextWindow: 256000,
12765
- maxTokens: 4096,
13015
+ contextWindow: 262144,
13016
+ maxTokens: 65536,
12766
13017
  },
12767
13018
  "qwen/qwen3.5-9b": {
12768
13019
  id: "qwen/qwen3.5-9b",
@@ -12841,13 +13092,13 @@ export const MODELS = {
12841
13092
  reasoning: true,
12842
13093
  input: ["text", "image"],
12843
13094
  cost: {
12844
- input: 0.28500000000000003,
12845
- output: 2.4,
12846
- cacheRead: 0.15,
13095
+ input: 0.44999999999999996,
13096
+ output: 2.7,
13097
+ cacheRead: 0,
12847
13098
  cacheWrite: 0,
12848
13099
  },
12849
13100
  contextWindow: 262144,
12850
- maxTokens: 262140,
13101
+ maxTokens: 65536,
12851
13102
  },
12852
13103
  "qwen/qwen3.6-35b-a3b": {
12853
13104
  id: "qwen/qwen3.6-35b-a3b",
@@ -13062,13 +13313,13 @@ export const MODELS = {
13062
13313
  reasoning: true,
13063
13314
  input: ["text"],
13064
13315
  cost: {
13065
- input: 0.14,
13066
- output: 0.58,
13067
- cacheRead: 0.035,
13316
+ input: 0.19999999999999998,
13317
+ output: 0.7999999999999999,
13318
+ cacheRead: 0.049999999999999996,
13068
13319
  cacheWrite: 0,
13069
13320
  },
13070
13321
  contextWindow: 262144,
13071
- maxTokens: 4096,
13322
+ maxTokens: 131072,
13072
13323
  },
13073
13324
  "tencent/hy3:free": {
13074
13325
  id: "tencent/hy3:free",
@@ -13198,13 +13449,13 @@ export const MODELS = {
13198
13449
  reasoning: true,
13199
13450
  input: ["text", "image"],
13200
13451
  cost: {
13201
- input: 0.105,
13452
+ input: 0.14,
13202
13453
  output: 0.28,
13203
- cacheRead: 0.028,
13454
+ cacheRead: 0.0028,
13204
13455
  cacheWrite: 0,
13205
13456
  },
13206
13457
  contextWindow: 1048576,
13207
- maxTokens: 4096,
13458
+ maxTokens: 131072,
13208
13459
  },
13209
13460
  "xiaomi/mimo-v2.5-pro": {
13210
13461
  id: "xiaomi/mimo-v2.5-pro",
@@ -13283,9 +13534,9 @@ export const MODELS = {
13283
13534
  reasoning: true,
13284
13535
  input: ["text"],
13285
13536
  cost: {
13286
- input: 0.43,
13287
- output: 1.74,
13288
- cacheRead: 0.08,
13537
+ input: 0.5,
13538
+ output: 2,
13539
+ cacheRead: 0.09999999999999999,
13289
13540
  cacheWrite: 0,
13290
13541
  },
13291
13542
  contextWindow: 202752,
@@ -13334,13 +13585,13 @@ export const MODELS = {
13334
13585
  reasoning: true,
13335
13586
  input: ["text"],
13336
13587
  cost: {
13337
- input: 0.06,
13588
+ input: 0.060500000000000005,
13338
13589
  output: 0.39999999999999997,
13339
- cacheRead: 0.01,
13590
+ cacheRead: 0,
13340
13591
  cacheWrite: 0,
13341
13592
  },
13342
- contextWindow: 202752,
13343
- maxTokens: 16384,
13593
+ contextWindow: 200000,
13594
+ maxTokens: 131072,
13344
13595
  },
13345
13596
  "z-ai/glm-5": {
13346
13597
  id: "z-ai/glm-5",
@@ -13357,7 +13608,7 @@ export const MODELS = {
13357
13608
  cacheWrite: 0,
13358
13609
  },
13359
13610
  contextWindow: 202752,
13360
- maxTokens: 4096,
13611
+ maxTokens: 202752,
13361
13612
  },
13362
13613
  "z-ai/glm-5-turbo": {
13363
13614
  id: "z-ai/glm-5-turbo",
@@ -13373,7 +13624,7 @@ export const MODELS = {
13373
13624
  cacheRead: 0.24,
13374
13625
  cacheWrite: 0,
13375
13626
  },
13376
- contextWindow: 262144,
13627
+ contextWindow: 202752,
13377
13628
  maxTokens: 131072,
13378
13629
  },
13379
13630
  "z-ai/glm-5.1": {
@@ -13402,13 +13653,13 @@ export const MODELS = {
13402
13653
  reasoning: true,
13403
13654
  input: ["text"],
13404
13655
  cost: {
13405
- input: 0.54,
13406
- output: 1.76,
13407
- cacheRead: 0.09999999999999999,
13656
+ input: 0.8932,
13657
+ output: 2.8072,
13658
+ cacheRead: 0.16588,
13408
13659
  cacheWrite: 0,
13409
13660
  },
13410
13661
  contextWindow: 1048576,
13411
- maxTokens: 101376,
13662
+ maxTokens: 131072,
13412
13663
  },
13413
13664
  "z-ai/glm-5v-turbo": {
13414
13665
  id: "z-ai/glm-5v-turbo",
@@ -13539,9 +13790,9 @@ export const MODELS = {
13539
13790
  reasoning: true,
13540
13791
  input: ["text", "image"],
13541
13792
  cost: {
13542
- input: 0.65,
13793
+ input: 0.66,
13543
13794
  output: 3.41,
13544
- cacheRead: 0.14,
13795
+ cacheRead: 0.15,
13545
13796
  cacheWrite: 0,
13546
13797
  },
13547
13798
  contextWindow: 262144,
@@ -14520,7 +14771,7 @@ export const MODELS = {
14520
14771
  cacheWrite: 18.75,
14521
14772
  },
14522
14773
  contextWindow: 200000,
14523
- maxTokens: 32000,
14774
+ maxTokens: 8192,
14524
14775
  },
14525
14776
  "anthropic/claude-opus-4.1": {
14526
14777
  id: "anthropic/claude-opus-4.1",
@@ -14592,6 +14843,24 @@ export const MODELS = {
14592
14843
  contextWindow: 1000000,
14593
14844
  maxTokens: 128000,
14594
14845
  },
14846
+ "anthropic/claude-opus-4.7-fast": {
14847
+ id: "anthropic/claude-opus-4.7-fast",
14848
+ name: "Claude Opus 4.7 (Fast)",
14849
+ api: "anthropic-messages",
14850
+ provider: "vercel-ai-gateway",
14851
+ baseUrl: "https://ai-gateway.vercel.sh",
14852
+ reasoning: true,
14853
+ thinkingLevelMap: { "xhigh": "xhigh" },
14854
+ input: ["text", "image"],
14855
+ cost: {
14856
+ input: 30,
14857
+ output: 150,
14858
+ cacheRead: 3,
14859
+ cacheWrite: 37.5,
14860
+ },
14861
+ contextWindow: 1000000,
14862
+ maxTokens: 128000,
14863
+ },
14595
14864
  "anthropic/claude-opus-4.8": {
14596
14865
  id: "anthropic/claude-opus-4.8",
14597
14866
  name: "Claude Opus 4.8",
@@ -14610,6 +14879,24 @@ export const MODELS = {
14610
14879
  contextWindow: 1000000,
14611
14880
  maxTokens: 128000,
14612
14881
  },
14882
+ "anthropic/claude-opus-4.8-fast": {
14883
+ id: "anthropic/claude-opus-4.8-fast",
14884
+ name: "Claude Opus 4.8 (Fast)",
14885
+ api: "anthropic-messages",
14886
+ provider: "vercel-ai-gateway",
14887
+ baseUrl: "https://ai-gateway.vercel.sh",
14888
+ reasoning: true,
14889
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
14890
+ input: ["text", "image"],
14891
+ cost: {
14892
+ input: 10,
14893
+ output: 50,
14894
+ cacheRead: 1,
14895
+ cacheWrite: 12.5,
14896
+ },
14897
+ contextWindow: 1000000,
14898
+ maxTokens: 128000,
14899
+ },
14613
14900
  "anthropic/claude-sonnet-4": {
14614
14901
  id: "anthropic/claude-sonnet-4",
14615
14902
  name: "Claude Sonnet 4",
@@ -14625,7 +14912,7 @@ export const MODELS = {
14625
14912
  cacheWrite: 3.75,
14626
14913
  },
14627
14914
  contextWindow: 1000000,
14628
- maxTokens: 64000,
14915
+ maxTokens: 8192,
14629
14916
  },
14630
14917
  "anthropic/claude-sonnet-4.5": {
14631
14918
  id: "anthropic/claude-sonnet-4.5",
@@ -14678,23 +14965,6 @@ export const MODELS = {
14678
14965
  contextWindow: 1000000,
14679
14966
  maxTokens: 128000,
14680
14967
  },
14681
- "arcee-ai/trinity-large-preview": {
14682
- id: "arcee-ai/trinity-large-preview",
14683
- name: "Trinity Large Preview",
14684
- api: "anthropic-messages",
14685
- provider: "vercel-ai-gateway",
14686
- baseUrl: "https://ai-gateway.vercel.sh",
14687
- reasoning: false,
14688
- input: ["text"],
14689
- cost: {
14690
- input: 0.25,
14691
- output: 1,
14692
- cacheRead: 0,
14693
- cacheWrite: 0,
14694
- },
14695
- contextWindow: 131000,
14696
- maxTokens: 131000,
14697
- },
14698
14968
  "arcee-ai/trinity-large-thinking": {
14699
14969
  id: "arcee-ai/trinity-large-thinking",
14700
14970
  name: "Trinity Large Thinking",
@@ -14823,12 +15093,12 @@ export const MODELS = {
14823
15093
  reasoning: true,
14824
15094
  input: ["text"],
14825
15095
  cost: {
14826
- input: 0.6,
14827
- output: 1.7,
14828
- cacheRead: 0,
15096
+ input: 0.25,
15097
+ output: 0.95,
15098
+ cacheRead: 0.13,
14829
15099
  cacheWrite: 0,
14830
15100
  },
14831
- contextWindow: 128000,
15101
+ contextWindow: 163840,
14832
15102
  maxTokens: 128000,
14833
15103
  },
14834
15104
  "deepseek/deepseek-v3.1-terminus": {
@@ -14893,7 +15163,7 @@ export const MODELS = {
14893
15163
  cost: {
14894
15164
  input: 0.14,
14895
15165
  output: 0.28,
14896
- cacheRead: 0.0028,
15166
+ cacheRead: 0.028,
14897
15167
  cacheWrite: 0,
14898
15168
  },
14899
15169
  contextWindow: 1000000,
@@ -15154,6 +15424,23 @@ export const MODELS = {
15154
15424
  contextWindow: 1000000,
15155
15425
  maxTokens: 32000,
15156
15426
  },
15427
+ "kwaipilot/kat-coder-air-v2.5": {
15428
+ id: "kwaipilot/kat-coder-air-v2.5",
15429
+ name: "Kat Coder Air V2.5",
15430
+ api: "anthropic-messages",
15431
+ provider: "vercel-ai-gateway",
15432
+ baseUrl: "https://ai-gateway.vercel.sh",
15433
+ reasoning: true,
15434
+ input: ["text"],
15435
+ cost: {
15436
+ input: 0.15,
15437
+ output: 0.6,
15438
+ cacheRead: 0.03,
15439
+ cacheWrite: 0,
15440
+ },
15441
+ contextWindow: 256000,
15442
+ maxTokens: 80000,
15443
+ },
15157
15444
  "kwaipilot/kat-coder-pro-v1": {
15158
15445
  id: "kwaipilot/kat-coder-pro-v1",
15159
15446
  name: "KAT-Coder-Pro V1",
@@ -15188,39 +15475,22 @@ export const MODELS = {
15188
15475
  contextWindow: 256000,
15189
15476
  maxTokens: 256000,
15190
15477
  },
15191
- "meituan/longcat-flash-chat": {
15192
- id: "meituan/longcat-flash-chat",
15193
- name: "LongCat Flash Chat",
15194
- api: "anthropic-messages",
15195
- provider: "vercel-ai-gateway",
15196
- baseUrl: "https://ai-gateway.vercel.sh",
15197
- reasoning: false,
15198
- input: ["text"],
15199
- cost: {
15200
- input: 0,
15201
- output: 0,
15202
- cacheRead: 0,
15203
- cacheWrite: 0,
15204
- },
15205
- contextWindow: 128000,
15206
- maxTokens: 100000,
15207
- },
15208
- "meituan/longcat-flash-thinking-2601": {
15209
- id: "meituan/longcat-flash-thinking-2601",
15210
- name: "LongCat Flash Thinking 2601",
15478
+ "kwaipilot/kat-coder-pro-v2.5": {
15479
+ id: "kwaipilot/kat-coder-pro-v2.5",
15480
+ name: "Kat Coder Pro V2.5",
15211
15481
  api: "anthropic-messages",
15212
15482
  provider: "vercel-ai-gateway",
15213
15483
  baseUrl: "https://ai-gateway.vercel.sh",
15214
15484
  reasoning: true,
15215
15485
  input: ["text"],
15216
15486
  cost: {
15217
- input: 0,
15218
- output: 0,
15219
- cacheRead: 0,
15487
+ input: 0.74,
15488
+ output: 2.96,
15489
+ cacheRead: 0.15,
15220
15490
  cacheWrite: 0,
15221
15491
  },
15222
- contextWindow: 32768,
15223
- maxTokens: 32768,
15492
+ contextWindow: 256000,
15493
+ maxTokens: 80000,
15224
15494
  },
15225
15495
  "meta/llama-3.1-70b": {
15226
15496
  id: "meta/llama-3.1-70b",
@@ -15348,7 +15618,7 @@ export const MODELS = {
15348
15618
  provider: "vercel-ai-gateway",
15349
15619
  baseUrl: "https://ai-gateway.vercel.sh",
15350
15620
  reasoning: true,
15351
- input: ["text"],
15621
+ input: ["text", "image"],
15352
15622
  cost: {
15353
15623
  input: 1.25,
15354
15624
  output: 4.25,
@@ -15528,23 +15798,6 @@ export const MODELS = {
15528
15798
  contextWindow: 256000,
15529
15799
  maxTokens: 256000,
15530
15800
  },
15531
- "mistral/devstral-small": {
15532
- id: "mistral/devstral-small",
15533
- name: "Devstral Small 1.1",
15534
- api: "anthropic-messages",
15535
- provider: "vercel-ai-gateway",
15536
- baseUrl: "https://ai-gateway.vercel.sh",
15537
- reasoning: false,
15538
- input: ["text"],
15539
- cost: {
15540
- input: 0.09999999999999999,
15541
- output: 0.3,
15542
- cacheRead: 0,
15543
- cacheWrite: 0,
15544
- },
15545
- contextWindow: 128000,
15546
- maxTokens: 64000,
15547
- },
15548
15801
  "mistral/devstral-small-2": {
15549
15802
  id: "mistral/devstral-small-2",
15550
15803
  name: "Devstral Small 2",
@@ -15749,23 +16002,6 @@ export const MODELS = {
15749
16002
  contextWindow: 128000,
15750
16003
  maxTokens: 4000,
15751
16004
  },
15752
- "mistral/pixtral-large": {
15753
- id: "mistral/pixtral-large",
15754
- name: "Pixtral Large",
15755
- api: "anthropic-messages",
15756
- provider: "vercel-ai-gateway",
15757
- baseUrl: "https://ai-gateway.vercel.sh",
15758
- reasoning: false,
15759
- input: ["text", "image"],
15760
- cost: {
15761
- input: 2,
15762
- output: 6,
15763
- cacheRead: 0,
15764
- cacheWrite: 0,
15765
- },
15766
- contextWindow: 128000,
15767
- maxTokens: 4000,
15768
- },
15769
16005
  "moonshotai/kimi-k2": {
15770
16006
  id: "moonshotai/kimi-k2",
15771
16007
  name: "Kimi K2 Instruct",
@@ -17164,9 +17400,9 @@ export const MODELS = {
17164
17400
  reasoning: true,
17165
17401
  input: ["text"],
17166
17402
  cost: {
17167
- input: 3,
17168
- output: 10.25,
17169
- cacheRead: 0.5,
17403
+ input: 2.0999999999999996,
17404
+ output: 6.6000000000000005,
17405
+ cacheRead: 0.21,
17170
17406
  cacheWrite: 0,
17171
17407
  },
17172
17408
  contextWindow: 1000000,