@dreb/ai 2.57.1 → 2.58.1

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.
@@ -767,6 +767,57 @@ export const MODELS = {
767
767
  contextWindow: 1000000,
768
768
  maxTokens: 128000,
769
769
  },
770
+ "global.openai.gpt-5.6-luna": {
771
+ id: "global.openai.gpt-5.6-luna",
772
+ name: "GPT-5.6 Luna (Global)",
773
+ api: "bedrock-converse-stream",
774
+ provider: "amazon-bedrock",
775
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
776
+ reasoning: true,
777
+ input: ["text", "image"],
778
+ cost: {
779
+ input: 0.22,
780
+ output: 1.32,
781
+ cacheRead: 0.022,
782
+ cacheWrite: 0.275,
783
+ },
784
+ contextWindow: 1050000,
785
+ maxTokens: 128000,
786
+ },
787
+ "global.openai.gpt-5.6-sol": {
788
+ id: "global.openai.gpt-5.6-sol",
789
+ name: "GPT-5.6 Sol (Global)",
790
+ api: "bedrock-converse-stream",
791
+ provider: "amazon-bedrock",
792
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
793
+ reasoning: true,
794
+ input: ["text", "image"],
795
+ cost: {
796
+ input: 5.5,
797
+ output: 33,
798
+ cacheRead: 0.55,
799
+ cacheWrite: 6.875,
800
+ },
801
+ contextWindow: 1050000,
802
+ maxTokens: 128000,
803
+ },
804
+ "global.openai.gpt-5.6-terra": {
805
+ id: "global.openai.gpt-5.6-terra",
806
+ name: "GPT-5.6 Terra (Global)",
807
+ api: "bedrock-converse-stream",
808
+ provider: "amazon-bedrock",
809
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
810
+ reasoning: true,
811
+ input: ["text", "image"],
812
+ cost: {
813
+ input: 2.2,
814
+ output: 13.2,
815
+ cacheRead: 0.22,
816
+ cacheWrite: 2.75,
817
+ },
818
+ contextWindow: 1050000,
819
+ maxTokens: 128000,
820
+ },
770
821
  "google.gemma-3-27b-it": {
771
822
  id: "google.gemma-3-27b-it",
772
823
  name: "Google Gemma 3 27B Instruct",
@@ -4760,6 +4811,42 @@ export const MODELS = {
4760
4811
  contextWindow: 262144,
4761
4812
  maxTokens: 131072,
4762
4813
  },
4814
+ "Qwen/Qwen3-VL-235B-A22B-Instruct": {
4815
+ id: "Qwen/Qwen3-VL-235B-A22B-Instruct",
4816
+ name: "Qwen3 VL 235B A22B Instruct",
4817
+ api: "openai-completions",
4818
+ provider: "huggingface",
4819
+ baseUrl: "https://router.huggingface.co/v1",
4820
+ compat: { "supportsDeveloperRole": false },
4821
+ reasoning: false,
4822
+ input: ["text", "image"],
4823
+ cost: {
4824
+ input: 0.3,
4825
+ output: 1.5,
4826
+ cacheRead: 0,
4827
+ cacheWrite: 0,
4828
+ },
4829
+ contextWindow: 131072,
4830
+ maxTokens: 32768,
4831
+ },
4832
+ "Qwen/Qwen3-VL-235B-A22B-Thinking": {
4833
+ id: "Qwen/Qwen3-VL-235B-A22B-Thinking",
4834
+ name: "Qwen3 VL 235B A22B Thinking",
4835
+ api: "openai-completions",
4836
+ provider: "huggingface",
4837
+ baseUrl: "https://router.huggingface.co/v1",
4838
+ compat: { "supportsDeveloperRole": false },
4839
+ reasoning: true,
4840
+ input: ["text", "image"],
4841
+ cost: {
4842
+ input: 0.98,
4843
+ output: 3.95,
4844
+ cacheRead: 0,
4845
+ cacheWrite: 0,
4846
+ },
4847
+ contextWindow: 131072,
4848
+ maxTokens: 32768,
4849
+ },
4763
4850
  "Qwen/Qwen3.5-122B-A10B": {
4764
4851
  id: "Qwen/Qwen3.5-122B-A10B",
4765
4852
  name: "Qwen3.5 122B-A10B",
@@ -4886,6 +4973,24 @@ export const MODELS = {
4886
4973
  contextWindow: 262144,
4887
4974
  maxTokens: 65536,
4888
4975
  },
4976
+ "Qwen/Qwen3.8-2.4T-A95B": {
4977
+ id: "Qwen/Qwen3.8-2.4T-A95B",
4978
+ name: "Qwen3.8 2.4T A95B",
4979
+ api: "openai-completions",
4980
+ provider: "huggingface",
4981
+ baseUrl: "https://router.huggingface.co/v1",
4982
+ compat: { "supportsDeveloperRole": false },
4983
+ reasoning: true,
4984
+ input: ["text"],
4985
+ cost: {
4986
+ input: 2.5,
4987
+ output: 6.25,
4988
+ cacheRead: 0,
4989
+ cacheWrite: 0,
4990
+ },
4991
+ contextWindow: 262144,
4992
+ maxTokens: 131072,
4993
+ },
4889
4994
  "XiaomiMiMo/MiMo-V2-Flash": {
4890
4995
  id: "XiaomiMiMo/MiMo-V2-Flash",
4891
4996
  name: "MiMo-V2-Flash",
@@ -8281,16 +8386,16 @@ export const MODELS = {
8281
8386
  "opencode-go": {
8282
8387
  "deepseek-v4-flash": {
8283
8388
  id: "deepseek-v4-flash",
8284
- name: "DeepSeek V4 Flash (2x usage)",
8389
+ name: "DeepSeek V4 Flash",
8285
8390
  api: "openai-completions",
8286
8391
  provider: "opencode-go",
8287
8392
  baseUrl: "https://opencode.ai/zen/go/v1",
8288
8393
  reasoning: true,
8289
8394
  input: ["text"],
8290
8395
  cost: {
8291
- input: 0.07,
8292
- output: 0.14,
8293
- cacheRead: 0.0014,
8396
+ input: 0.22,
8397
+ output: 0.66,
8398
+ cacheRead: 0.007,
8294
8399
  cacheWrite: 0,
8295
8400
  },
8296
8401
  contextWindow: 1000000,
@@ -8305,9 +8410,9 @@ export const MODELS = {
8305
8410
  reasoning: true,
8306
8411
  input: ["text"],
8307
8412
  cost: {
8308
- input: 0.435,
8309
- output: 0.87,
8310
- cacheRead: 0.003625,
8413
+ input: 0.66,
8414
+ output: 1.98,
8415
+ cacheRead: 0.022,
8311
8416
  cacheWrite: 0,
8312
8417
  },
8313
8418
  contextWindow: 1000000,
@@ -8537,9 +8642,9 @@ export const MODELS = {
8537
8642
  "qwen3.6-plus": {
8538
8643
  id: "qwen3.6-plus",
8539
8644
  name: "Qwen3.6 Plus",
8540
- api: "anthropic-messages",
8645
+ api: "openai-completions",
8541
8646
  provider: "opencode-go",
8542
- baseUrl: "https://opencode.ai/zen/go",
8647
+ baseUrl: "https://opencode.ai/zen/go/v1",
8543
8648
  reasoning: true,
8544
8649
  input: ["text", "image"],
8545
8650
  cost: {
@@ -8554,9 +8659,9 @@ export const MODELS = {
8554
8659
  "qwen3.7-max": {
8555
8660
  id: "qwen3.7-max",
8556
8661
  name: "Qwen3.7 Max",
8557
- api: "anthropic-messages",
8662
+ api: "openai-completions",
8558
8663
  provider: "opencode-go",
8559
- baseUrl: "https://opencode.ai/zen/go",
8664
+ baseUrl: "https://opencode.ai/zen/go/v1",
8560
8665
  reasoning: true,
8561
8666
  input: ["text"],
8562
8667
  cost: {
@@ -8571,9 +8676,9 @@ export const MODELS = {
8571
8676
  "qwen3.7-plus": {
8572
8677
  id: "qwen3.7-plus",
8573
8678
  name: "Qwen3.7 Plus",
8574
- api: "anthropic-messages",
8679
+ api: "openai-completions",
8575
8680
  provider: "opencode-go",
8576
- baseUrl: "https://opencode.ai/zen/go",
8681
+ baseUrl: "https://opencode.ai/zen/go/v1",
8577
8682
  reasoning: true,
8578
8683
  input: ["text", "image"],
8579
8684
  cost: {
@@ -8588,9 +8693,9 @@ export const MODELS = {
8588
8693
  "qwen3.8-max": {
8589
8694
  id: "qwen3.8-max",
8590
8695
  name: "Qwen3.8 Max",
8591
- api: "anthropic-messages",
8696
+ api: "openai-completions",
8592
8697
  provider: "opencode-go",
8593
- baseUrl: "https://opencode.ai/zen/go",
8698
+ baseUrl: "https://opencode.ai/zen/go/v1",
8594
8699
  reasoning: true,
8595
8700
  input: ["text", "image"],
8596
8701
  cost: {
@@ -9532,12 +9637,12 @@ export const MODELS = {
9532
9637
  input: ["text"],
9533
9638
  cost: {
9534
9639
  input: 0.27,
9535
- output: 0.95,
9536
- cacheRead: 0.13,
9640
+ output: 1,
9641
+ cacheRead: 0,
9537
9642
  cacheWrite: 0,
9538
9643
  },
9539
9644
  contextWindow: 163840,
9540
- maxTokens: 32768,
9645
+ maxTokens: 163840,
9541
9646
  },
9542
9647
  "deepseek/deepseek-v3.2": {
9543
9648
  id: "deepseek/deepseek-v3.2",
@@ -9582,9 +9687,9 @@ export const MODELS = {
9582
9687
  reasoning: true,
9583
9688
  input: ["text"],
9584
9689
  cost: {
9585
- input: 0.06426,
9586
- output: 0.12852,
9587
- cacheRead: 0.012852,
9690
+ input: 0.08259999999999999,
9691
+ output: 0.16519999999999999,
9692
+ cacheRead: 0.01652,
9588
9693
  cacheWrite: 0,
9589
9694
  },
9590
9695
  contextWindow: 1048576,
@@ -9604,25 +9709,25 @@ export const MODELS = {
9604
9709
  cacheRead: 0.028,
9605
9710
  cacheWrite: 0,
9606
9711
  },
9607
- contextWindow: 1048576,
9712
+ contextWindow: 1310720,
9608
9713
  maxTokens: 393216,
9609
9714
  },
9610
9715
  "deepseek/deepseek-v4-pro": {
9611
9716
  id: "deepseek/deepseek-v4-pro",
9612
- name: "DeepSeek: DeepSeek V4 Pro",
9717
+ name: "DeepSeek: DeepSeek V4 Pro 0423",
9613
9718
  api: "openai-completions",
9614
9719
  provider: "openrouter",
9615
9720
  baseUrl: "https://openrouter.ai/api/v1",
9616
9721
  reasoning: true,
9617
9722
  input: ["text"],
9618
9723
  cost: {
9619
- input: 1.1680000000000001,
9620
- output: 2.3360000000000003,
9621
- cacheRead: 0.09855000000000001,
9724
+ input: 0.66,
9725
+ output: 1.9800000000000002,
9726
+ cacheRead: 0.022,
9622
9727
  cacheWrite: 0,
9623
9728
  },
9624
9729
  contextWindow: 1048576,
9625
- maxTokens: 393216,
9730
+ maxTokens: 384000,
9626
9731
  },
9627
9732
  "deepseek/deepseek-v4-pro-0813": {
9628
9733
  id: "deepseek/deepseek-v4-pro-0813",
@@ -9633,9 +9738,9 @@ export const MODELS = {
9633
9738
  reasoning: true,
9634
9739
  input: ["text"],
9635
9740
  cost: {
9636
- input: 0.435,
9637
- output: 0.87,
9638
- cacheRead: 0.003625,
9741
+ input: 0.66,
9742
+ output: 1.9800000000000002,
9743
+ cacheRead: 0.022,
9639
9744
  cacheWrite: 0,
9640
9745
  },
9641
9746
  contextWindow: 1048576,
@@ -10126,13 +10231,13 @@ export const MODELS = {
10126
10231
  reasoning: true,
10127
10232
  input: ["text", "image"],
10128
10233
  cost: {
10129
- input: 0.12,
10130
- output: 0.39999999999999997,
10131
- cacheRead: 0.049999999999999996,
10234
+ input: 0.07,
10235
+ output: 0.33999999999999997,
10236
+ cacheRead: 0,
10132
10237
  cacheWrite: 0,
10133
10238
  },
10134
10239
  contextWindow: 262144,
10135
- maxTokens: 262144,
10240
+ maxTokens: 16384,
10136
10241
  },
10137
10242
  "google/gemma-4-26b-a4b-it:free": {
10138
10243
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10338,23 +10443,6 @@ export const MODELS = {
10338
10443
  contextWindow: 256000,
10339
10444
  maxTokens: 80000,
10340
10445
  },
10341
- "liquid/lfm-2.5-2.6b:free": {
10342
- id: "liquid/lfm-2.5-2.6b:free",
10343
- name: "LiquidAI: LFM2.5-2.6B (free)",
10344
- api: "openai-completions",
10345
- provider: "openrouter",
10346
- baseUrl: "https://openrouter.ai/api/v1",
10347
- reasoning: true,
10348
- input: ["text"],
10349
- cost: {
10350
- input: 0,
10351
- output: 0,
10352
- cacheRead: 0,
10353
- cacheWrite: 0,
10354
- },
10355
- contextWindow: 128000,
10356
- maxTokens: 8192,
10357
- },
10358
10446
  "meituan/longcat-2.0": {
10359
10447
  id: "meituan/longcat-2.0",
10360
10448
  name: "Meituan: LongCat 2.0",
@@ -10976,9 +11064,9 @@ export const MODELS = {
10976
11064
  reasoning: true,
10977
11065
  input: ["text", "image"],
10978
11066
  cost: {
10979
- input: 0.5415,
10980
- output: 2.2800000000000002,
10981
- cacheRead: 0.09119999999999999,
11067
+ input: 0.95,
11068
+ output: 4,
11069
+ cacheRead: 0.16,
10982
11070
  cacheWrite: 0,
10983
11071
  },
10984
11072
  contextWindow: 262144,
@@ -11080,11 +11168,11 @@ export const MODELS = {
11080
11168
  cost: {
11081
11169
  input: 0.049999999999999996,
11082
11170
  output: 0.19999999999999998,
11083
- cacheRead: 0.024999999999999998,
11171
+ cacheRead: 0.03,
11084
11172
  cacheWrite: 0,
11085
11173
  },
11086
11174
  contextWindow: 262144,
11087
- maxTokens: 228000,
11175
+ maxTokens: 262144,
11088
11176
  },
11089
11177
  "nvidia/nemotron-3-nano-30b-a3b:free": {
11090
11178
  id: "nvidia/nemotron-3-nano-30b-a3b:free",
@@ -11214,13 +11302,13 @@ export const MODELS = {
11214
11302
  reasoning: true,
11215
11303
  input: ["text"],
11216
11304
  cost: {
11217
- input: 0.09999999999999999,
11218
- output: 0.25,
11219
- cacheRead: 0.049999999999999996,
11305
+ input: 0.08,
11306
+ output: 0.19999999999999998,
11307
+ cacheRead: 0.04,
11220
11308
  cacheWrite: 0,
11221
11309
  },
11222
11310
  contextWindow: 1000000,
11223
- maxTokens: 262144,
11311
+ maxTokens: 131072,
11224
11312
  },
11225
11313
  "nvidia/nemotron-3.5-lightning:free": {
11226
11314
  id: "nvidia/nemotron-3.5-lightning:free",
@@ -12200,10 +12288,10 @@ export const MODELS = {
12200
12288
  reasoning: true,
12201
12289
  input: ["text", "image"],
12202
12290
  cost: {
12203
- input: 0.09999999999999999,
12204
- output: 0.6,
12205
- cacheRead: 0.01,
12206
- cacheWrite: 0.125,
12291
+ input: 0.19999999999999998,
12292
+ output: 1.2,
12293
+ cacheRead: 0.02,
12294
+ cacheWrite: 0.25,
12207
12295
  },
12208
12296
  contextWindow: 1050000,
12209
12297
  maxTokens: 128000,
@@ -12217,10 +12305,10 @@ export const MODELS = {
12217
12305
  reasoning: true,
12218
12306
  input: ["text", "image"],
12219
12307
  cost: {
12220
- input: 0.09999999999999999,
12221
- output: 0.6,
12222
- cacheRead: 0.01,
12223
- cacheWrite: 0.125,
12308
+ input: 0.19999999999999998,
12309
+ output: 1.2,
12310
+ cacheRead: 0.02,
12311
+ cacheWrite: 0.25,
12224
12312
  },
12225
12313
  contextWindow: 1050000,
12226
12314
  maxTokens: 128000,
@@ -12268,10 +12356,10 @@ export const MODELS = {
12268
12356
  reasoning: true,
12269
12357
  input: ["text", "image"],
12270
12358
  cost: {
12271
- input: 5,
12272
- output: 30,
12273
- cacheRead: 0.5,
12274
- cacheWrite: 6.25,
12359
+ input: 2.5,
12360
+ output: 15,
12361
+ cacheRead: 0.25,
12362
+ cacheWrite: 3.125,
12275
12363
  },
12276
12364
  contextWindow: 1050000,
12277
12365
  maxTokens: 128000,
@@ -12285,10 +12373,10 @@ export const MODELS = {
12285
12373
  reasoning: true,
12286
12374
  input: ["text", "image"],
12287
12375
  cost: {
12288
- input: 5,
12289
- output: 30,
12290
- cacheRead: 0.5,
12291
- cacheWrite: 6.25,
12376
+ input: 2.5,
12377
+ output: 15,
12378
+ cacheRead: 0.25,
12379
+ cacheWrite: 3.125,
12292
12380
  },
12293
12381
  contextWindow: 1050000,
12294
12382
  maxTokens: 128000,
@@ -12302,9 +12390,9 @@ export const MODELS = {
12302
12390
  reasoning: true,
12303
12391
  input: ["text", "image"],
12304
12392
  cost: {
12305
- input: 2.5,
12306
- output: 15,
12307
- cacheRead: 0.25,
12393
+ input: 1.25,
12394
+ output: 7.5,
12395
+ cacheRead: 0.125,
12308
12396
  cacheWrite: 0,
12309
12397
  },
12310
12398
  contextWindow: 1050000,
@@ -12319,9 +12407,9 @@ export const MODELS = {
12319
12407
  reasoning: true,
12320
12408
  input: ["text", "image"],
12321
12409
  cost: {
12322
- input: 2.5,
12323
- output: 15,
12324
- cacheRead: 0.25,
12410
+ input: 1.25,
12411
+ output: 7.5,
12412
+ cacheRead: 0.125,
12325
12413
  cacheWrite: 0,
12326
12414
  },
12327
12415
  contextWindow: 1050000,
@@ -12336,10 +12424,10 @@ export const MODELS = {
12336
12424
  reasoning: true,
12337
12425
  input: ["text", "image"],
12338
12426
  cost: {
12339
- input: 1,
12340
- output: 6,
12341
- cacheRead: 0.09999999999999999,
12342
- cacheWrite: 1.25,
12427
+ input: 2,
12428
+ output: 12,
12429
+ cacheRead: 0.19999999999999998,
12430
+ cacheWrite: 2.5,
12343
12431
  },
12344
12432
  contextWindow: 1050000,
12345
12433
  maxTokens: 128000,
@@ -12353,10 +12441,10 @@ export const MODELS = {
12353
12441
  reasoning: true,
12354
12442
  input: ["text", "image"],
12355
12443
  cost: {
12356
- input: 1,
12357
- output: 6,
12358
- cacheRead: 0.09999999999999999,
12359
- cacheWrite: 1.25,
12444
+ input: 2,
12445
+ output: 12,
12446
+ cacheRead: 0.19999999999999998,
12447
+ cacheWrite: 2.5,
12360
12448
  },
12361
12449
  contextWindow: 1050000,
12362
12450
  maxTokens: 128000,
@@ -13050,13 +13138,13 @@ export const MODELS = {
13050
13138
  reasoning: true,
13051
13139
  input: ["text"],
13052
13140
  cost: {
13053
- input: 0.12,
13054
- output: 0.5,
13141
+ input: 0.13,
13142
+ output: 0.52,
13055
13143
  cacheRead: 0,
13056
13144
  cacheWrite: 0,
13057
13145
  },
13058
13146
  contextWindow: 131072,
13059
- maxTokens: 16384,
13147
+ maxTokens: 8192,
13060
13148
  },
13061
13149
  "qwen/qwen3-30b-a3b-instruct-2507": {
13062
13150
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -13288,9 +13376,9 @@ export const MODELS = {
13288
13376
  reasoning: false,
13289
13377
  input: ["text", "image"],
13290
13378
  cost: {
13291
- input: 0.26,
13292
- output: 1.04,
13293
- cacheRead: 0,
13379
+ input: 0.21,
13380
+ output: 1.9,
13381
+ cacheRead: 0.09999999999999999,
13294
13382
  cacheWrite: 0,
13295
13383
  },
13296
13384
  contextWindow: 262144,
@@ -13543,13 +13631,13 @@ export const MODELS = {
13543
13631
  reasoning: true,
13544
13632
  input: ["text", "image"],
13545
13633
  cost: {
13546
- input: 0.6,
13547
- output: 3.5999999999999996,
13548
- cacheRead: 0.12,
13634
+ input: 0.28900000000000003,
13635
+ output: 2.4,
13636
+ cacheRead: 0,
13549
13637
  cacheWrite: 0,
13550
13638
  },
13551
13639
  contextWindow: 262144,
13552
- maxTokens: 262144,
13640
+ maxTokens: 131072,
13553
13641
  },
13554
13642
  "qwen/qwen3.6-35b-a3b": {
13555
13643
  id: "qwen/qwen3.6-35b-a3b",
@@ -13560,7 +13648,7 @@ export const MODELS = {
13560
13648
  reasoning: true,
13561
13649
  input: ["text", "image"],
13562
13650
  cost: {
13563
- input: 0.15,
13651
+ input: 0.14,
13564
13652
  output: 1,
13565
13653
  cacheRead: 0.049999999999999996,
13566
13654
  cacheWrite: 0,
@@ -13698,7 +13786,7 @@ export const MODELS = {
13698
13786
  cost: {
13699
13787
  input: 0.44999999999999996,
13700
13788
  output: 3.1999999999999997,
13701
- cacheRead: 0,
13789
+ cacheRead: 0.049999999999999996,
13702
13790
  cacheWrite: 0,
13703
13791
  },
13704
13792
  contextWindow: 262144,
@@ -14155,9 +14243,9 @@ export const MODELS = {
14155
14243
  reasoning: true,
14156
14244
  input: ["text"],
14157
14245
  cost: {
14158
- input: 0.55,
14159
- output: 2.2,
14160
- cacheRead: 0.11,
14246
+ input: 0.5,
14247
+ output: 2,
14248
+ cacheRead: 0.09999999999999999,
14161
14249
  cacheWrite: 0,
14162
14250
  },
14163
14251
  contextWindow: 204800,
@@ -14274,13 +14362,13 @@ export const MODELS = {
14274
14362
  reasoning: true,
14275
14363
  input: ["text"],
14276
14364
  cost: {
14277
- input: 0.46199999999999997,
14278
- output: 1.452,
14279
- cacheRead: 0.0858,
14365
+ input: 1.19,
14366
+ output: 3.74,
14367
+ cacheRead: 0.221,
14280
14368
  cacheWrite: 0,
14281
14369
  },
14282
14370
  contextWindow: 1048576,
14283
- maxTokens: 131072,
14371
+ maxTokens: 262144,
14284
14372
  },
14285
14373
  "z-ai/glm-5.2:batch": {
14286
14374
  id: "z-ai/glm-5.2:batch",
@@ -14393,13 +14481,13 @@ export const MODELS = {
14393
14481
  reasoning: true,
14394
14482
  input: ["text"],
14395
14483
  cost: {
14396
- input: 0.0675,
14397
- output: 0.135,
14398
- cacheRead: 0.0135,
14484
+ input: 0.078596,
14485
+ output: 0.157192,
14486
+ cacheRead: 0.015719200000000003,
14399
14487
  cacheWrite: 0,
14400
14488
  },
14401
- contextWindow: 1048576,
14402
- maxTokens: 262144,
14489
+ contextWindow: 1310720,
14490
+ maxTokens: 384000,
14403
14491
  },
14404
14492
  "~google/gemini-flash-latest": {
14405
14493
  id: "~google/gemini-flash-latest",
@@ -14444,13 +14532,13 @@ export const MODELS = {
14444
14532
  reasoning: true,
14445
14533
  input: ["text", "image"],
14446
14534
  cost: {
14447
- input: 2.8,
14448
- output: 14,
14535
+ input: 2.6,
14536
+ output: 13,
14449
14537
  cacheRead: 0.29,
14450
14538
  cacheWrite: 0,
14451
14539
  },
14452
14540
  contextWindow: 1048576,
14453
- maxTokens: 1048576,
14541
+ maxTokens: 974842,
14454
14542
  },
14455
14543
  "~openai/gpt-latest": {
14456
14544
  id: "~openai/gpt-latest",
@@ -14461,10 +14549,10 @@ export const MODELS = {
14461
14549
  reasoning: true,
14462
14550
  input: ["text", "image"],
14463
14551
  cost: {
14464
- input: 5,
14465
- output: 30,
14466
- cacheRead: 0.5,
14467
- cacheWrite: 6.25,
14552
+ input: 2.5,
14553
+ output: 15,
14554
+ cacheRead: 0.25,
14555
+ cacheWrite: 3.125,
14468
14556
  },
14469
14557
  contextWindow: 1050000,
14470
14558
  maxTokens: 128000,
@@ -14941,12 +15029,29 @@ export const MODELS = {
14941
15029
  cost: {
14942
15030
  input: 2,
14943
15031
  output: 6,
14944
- cacheRead: 0.25,
15032
+ cacheRead: 0.19999999999999998,
14945
15033
  cacheWrite: 0,
14946
15034
  },
14947
15035
  contextWindow: 262144,
14948
15036
  maxTokens: 131072,
14949
15037
  },
15038
+ "alibaba/qwen3.8-27b": {
15039
+ id: "alibaba/qwen3.8-27b",
15040
+ name: "Qwen3.8 27B",
15041
+ api: "anthropic-messages",
15042
+ provider: "vercel-ai-gateway",
15043
+ baseUrl: "https://ai-gateway.vercel.sh",
15044
+ reasoning: true,
15045
+ input: ["text", "image"],
15046
+ cost: {
15047
+ input: 0.09999999999999999,
15048
+ output: 0.39999999999999997,
15049
+ cacheRead: 0.01,
15050
+ cacheWrite: 0,
15051
+ },
15052
+ contextWindow: 262144,
15053
+ maxTokens: 262133,
15054
+ },
14950
15055
  "alibaba/qwen3.8-max": {
14951
15056
  id: "alibaba/qwen3.8-max",
14952
15057
  name: "Qwen 3.8 Max",
@@ -15483,9 +15588,9 @@ export const MODELS = {
15483
15588
  reasoning: true,
15484
15589
  input: ["text"],
15485
15590
  cost: {
15486
- input: 0.19999999999999998,
15487
- output: 0.39999999999999997,
15488
- cacheRead: 0.04,
15591
+ input: 0.13,
15592
+ output: 0.26,
15593
+ cacheRead: 0.028,
15489
15594
  cacheWrite: 0,
15490
15595
  },
15491
15596
  contextWindow: 1000000,
@@ -15500,9 +15605,9 @@ export const MODELS = {
15500
15605
  reasoning: true,
15501
15606
  input: ["text"],
15502
15607
  cost: {
15503
- input: 0.19999999999999998,
15504
- output: 0.39999999999999997,
15505
- cacheRead: 0.04,
15608
+ input: 0.13,
15609
+ output: 0.26,
15610
+ cacheRead: 0.028,
15506
15611
  cacheWrite: 0,
15507
15612
  },
15508
15613
  contextWindow: 1000000,
@@ -15687,9 +15792,9 @@ export const MODELS = {
15687
15792
  reasoning: true,
15688
15793
  input: ["text", "image"],
15689
15794
  cost: {
15690
- input: 1.5,
15691
- output: 7.5,
15692
- cacheRead: 0.15,
15795
+ input: 0.75,
15796
+ output: 3.75,
15797
+ cacheRead: 0.075,
15693
15798
  cacheWrite: 0,
15694
15799
  },
15695
15800
  contextWindow: 1000000,
@@ -16596,6 +16701,23 @@ export const MODELS = {
16596
16701
  contextWindow: 1000000,
16597
16702
  maxTokens: 65000,
16598
16703
  },
16704
+ "nvidia/nemotron-3.5-lightning": {
16705
+ id: "nvidia/nemotron-3.5-lightning",
16706
+ name: "Nemotron 3.5 Lightning 30B",
16707
+ api: "anthropic-messages",
16708
+ provider: "vercel-ai-gateway",
16709
+ baseUrl: "https://ai-gateway.vercel.sh",
16710
+ reasoning: true,
16711
+ input: ["text"],
16712
+ cost: {
16713
+ input: 0,
16714
+ output: 0,
16715
+ cacheRead: 0,
16716
+ cacheWrite: 0,
16717
+ },
16718
+ contextWindow: 262144,
16719
+ maxTokens: 131072,
16720
+ },
16599
16721
  "nvidia/nemotron-nano-12b-v2-vl": {
16600
16722
  id: "nvidia/nemotron-nano-12b-v2-vl",
16601
16723
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -16681,6 +16803,23 @@ export const MODELS = {
16681
16803
  contextWindow: 1047576,
16682
16804
  maxTokens: 32768,
16683
16805
  },
16806
+ "openai/gpt-4.1-fast": {
16807
+ id: "openai/gpt-4.1-fast",
16808
+ name: "GPT-4.1 (Fast)",
16809
+ api: "anthropic-messages",
16810
+ provider: "vercel-ai-gateway",
16811
+ baseUrl: "https://ai-gateway.vercel.sh",
16812
+ reasoning: false,
16813
+ input: ["text", "image"],
16814
+ cost: {
16815
+ input: 3.5,
16816
+ output: 14,
16817
+ cacheRead: 0.875,
16818
+ cacheWrite: 0,
16819
+ },
16820
+ contextWindow: 1047576,
16821
+ maxTokens: 32768,
16822
+ },
16684
16823
  "openai/gpt-4.1-mini": {
16685
16824
  id: "openai/gpt-4.1-mini",
16686
16825
  name: "GPT-4.1 mini",
@@ -16698,6 +16837,23 @@ export const MODELS = {
16698
16837
  contextWindow: 1047576,
16699
16838
  maxTokens: 32768,
16700
16839
  },
16840
+ "openai/gpt-4.1-mini-fast": {
16841
+ id: "openai/gpt-4.1-mini-fast",
16842
+ name: "GPT-4.1 mini (Fast)",
16843
+ api: "anthropic-messages",
16844
+ provider: "vercel-ai-gateway",
16845
+ baseUrl: "https://ai-gateway.vercel.sh",
16846
+ reasoning: false,
16847
+ input: ["text", "image"],
16848
+ cost: {
16849
+ input: 0.7,
16850
+ output: 2.8,
16851
+ cacheRead: 0.175,
16852
+ cacheWrite: 0,
16853
+ },
16854
+ contextWindow: 1047576,
16855
+ maxTokens: 32768,
16856
+ },
16701
16857
  "openai/gpt-4.1-nano": {
16702
16858
  id: "openai/gpt-4.1-nano",
16703
16859
  name: "GPT-4.1 nano",
@@ -16715,6 +16871,23 @@ export const MODELS = {
16715
16871
  contextWindow: 1047576,
16716
16872
  maxTokens: 32768,
16717
16873
  },
16874
+ "openai/gpt-4.1-nano-fast": {
16875
+ id: "openai/gpt-4.1-nano-fast",
16876
+ name: "GPT-4.1 nano (Fast)",
16877
+ api: "anthropic-messages",
16878
+ provider: "vercel-ai-gateway",
16879
+ baseUrl: "https://ai-gateway.vercel.sh",
16880
+ reasoning: false,
16881
+ input: ["text", "image"],
16882
+ cost: {
16883
+ input: 0.19999999999999998,
16884
+ output: 0.7999999999999999,
16885
+ cacheRead: 0.049999999999999996,
16886
+ cacheWrite: 0,
16887
+ },
16888
+ contextWindow: 1047576,
16889
+ maxTokens: 32768,
16890
+ },
16718
16891
  "openai/gpt-4o": {
16719
16892
  id: "openai/gpt-4o",
16720
16893
  name: "GPT-4o",
@@ -16732,6 +16905,23 @@ export const MODELS = {
16732
16905
  contextWindow: 128000,
16733
16906
  maxTokens: 16384,
16734
16907
  },
16908
+ "openai/gpt-4o-fast": {
16909
+ id: "openai/gpt-4o-fast",
16910
+ name: "GPT-4o (Fast)",
16911
+ api: "anthropic-messages",
16912
+ provider: "vercel-ai-gateway",
16913
+ baseUrl: "https://ai-gateway.vercel.sh",
16914
+ reasoning: false,
16915
+ input: ["text", "image"],
16916
+ cost: {
16917
+ input: 4.25,
16918
+ output: 17,
16919
+ cacheRead: 2.125,
16920
+ cacheWrite: 0,
16921
+ },
16922
+ contextWindow: 128000,
16923
+ maxTokens: 16384,
16924
+ },
16735
16925
  "openai/gpt-4o-mini": {
16736
16926
  id: "openai/gpt-4o-mini",
16737
16927
  name: "GPT-4o mini",
@@ -16749,6 +16939,23 @@ export const MODELS = {
16749
16939
  contextWindow: 128000,
16750
16940
  maxTokens: 16384,
16751
16941
  },
16942
+ "openai/gpt-4o-mini-fast": {
16943
+ id: "openai/gpt-4o-mini-fast",
16944
+ name: "GPT-4o mini (Fast)",
16945
+ api: "anthropic-messages",
16946
+ provider: "vercel-ai-gateway",
16947
+ baseUrl: "https://ai-gateway.vercel.sh",
16948
+ reasoning: false,
16949
+ input: ["text", "image"],
16950
+ cost: {
16951
+ input: 0.25,
16952
+ output: 1,
16953
+ cacheRead: 0.125,
16954
+ cacheWrite: 0,
16955
+ },
16956
+ contextWindow: 128000,
16957
+ maxTokens: 16384,
16958
+ },
16752
16959
  "openai/gpt-5": {
16753
16960
  id: "openai/gpt-5",
16754
16961
  name: "GPT-5",
@@ -16783,6 +16990,23 @@ export const MODELS = {
16783
16990
  contextWindow: 400000,
16784
16991
  maxTokens: 128000,
16785
16992
  },
16993
+ "openai/gpt-5-fast": {
16994
+ id: "openai/gpt-5-fast",
16995
+ name: "GPT-5 (Fast)",
16996
+ api: "anthropic-messages",
16997
+ provider: "vercel-ai-gateway",
16998
+ baseUrl: "https://ai-gateway.vercel.sh",
16999
+ reasoning: true,
17000
+ input: ["text", "image"],
17001
+ cost: {
17002
+ input: 2.5,
17003
+ output: 20,
17004
+ cacheRead: 0.25,
17005
+ cacheWrite: 0,
17006
+ },
17007
+ contextWindow: 400000,
17008
+ maxTokens: 128000,
17009
+ },
16786
17010
  "openai/gpt-5-mini": {
16787
17011
  id: "openai/gpt-5-mini",
16788
17012
  name: "GPT-5 mini",
@@ -16800,6 +17024,23 @@ export const MODELS = {
16800
17024
  contextWindow: 400000,
16801
17025
  maxTokens: 128000,
16802
17026
  },
17027
+ "openai/gpt-5-mini-fast": {
17028
+ id: "openai/gpt-5-mini-fast",
17029
+ name: "GPT-5 mini (Fast)",
17030
+ api: "anthropic-messages",
17031
+ provider: "vercel-ai-gateway",
17032
+ baseUrl: "https://ai-gateway.vercel.sh",
17033
+ reasoning: true,
17034
+ input: ["text", "image"],
17035
+ cost: {
17036
+ input: 0.44999999999999996,
17037
+ output: 3.5999999999999996,
17038
+ cacheRead: 0.045,
17039
+ cacheWrite: 0,
17040
+ },
17041
+ contextWindow: 400000,
17042
+ maxTokens: 128000,
17043
+ },
16803
17044
  "openai/gpt-5-nano": {
16804
17045
  id: "openai/gpt-5-nano",
16805
17046
  name: "GPT-5 nano",
@@ -16902,6 +17143,23 @@ export const MODELS = {
16902
17143
  contextWindow: 400000,
16903
17144
  maxTokens: 128000,
16904
17145
  },
17146
+ "openai/gpt-5.1-thinking-fast": {
17147
+ id: "openai/gpt-5.1-thinking-fast",
17148
+ name: "GPT 5.1 Thinking (Fast)",
17149
+ api: "anthropic-messages",
17150
+ provider: "vercel-ai-gateway",
17151
+ baseUrl: "https://ai-gateway.vercel.sh",
17152
+ reasoning: true,
17153
+ input: ["text", "image"],
17154
+ cost: {
17155
+ input: 2.5,
17156
+ output: 20,
17157
+ cacheRead: 0.25,
17158
+ cacheWrite: 0,
17159
+ },
17160
+ contextWindow: 400000,
17161
+ maxTokens: 128000,
17162
+ },
16905
17163
  "openai/gpt-5.2": {
16906
17164
  id: "openai/gpt-5.2",
16907
17165
  name: "GPT 5.2",
@@ -16936,6 +17194,23 @@ export const MODELS = {
16936
17194
  contextWindow: 400000,
16937
17195
  maxTokens: 128000,
16938
17196
  },
17197
+ "openai/gpt-5.2-fast": {
17198
+ id: "openai/gpt-5.2-fast",
17199
+ name: "GPT 5.2 (Fast)",
17200
+ api: "anthropic-messages",
17201
+ provider: "vercel-ai-gateway",
17202
+ baseUrl: "https://ai-gateway.vercel.sh",
17203
+ reasoning: true,
17204
+ input: ["text", "image"],
17205
+ cost: {
17206
+ input: 3.5,
17207
+ output: 28,
17208
+ cacheRead: 0.35,
17209
+ cacheWrite: 0,
17210
+ },
17211
+ contextWindow: 400000,
17212
+ maxTokens: 128000,
17213
+ },
16939
17214
  "openai/gpt-5.2-pro": {
16940
17215
  id: "openai/gpt-5.2-pro",
16941
17216
  name: "GPT 5.2 ",
@@ -16970,6 +17245,23 @@ export const MODELS = {
16970
17245
  contextWindow: 400000,
16971
17246
  maxTokens: 128000,
16972
17247
  },
17248
+ "openai/gpt-5.3-codex-fast": {
17249
+ id: "openai/gpt-5.3-codex-fast",
17250
+ name: "GPT 5.3 Codex (Fast)",
17251
+ api: "anthropic-messages",
17252
+ provider: "vercel-ai-gateway",
17253
+ baseUrl: "https://ai-gateway.vercel.sh",
17254
+ reasoning: true,
17255
+ input: ["text", "image"],
17256
+ cost: {
17257
+ input: 3.5,
17258
+ output: 28,
17259
+ cacheRead: 0.35,
17260
+ cacheWrite: 0,
17261
+ },
17262
+ contextWindow: 400000,
17263
+ maxTokens: 128000,
17264
+ },
16973
17265
  "openai/gpt-5.4": {
16974
17266
  id: "openai/gpt-5.4",
16975
17267
  name: "GPT 5.4",
@@ -16987,6 +17279,23 @@ export const MODELS = {
16987
17279
  contextWindow: 1050000,
16988
17280
  maxTokens: 128000,
16989
17281
  },
17282
+ "openai/gpt-5.4-fast": {
17283
+ id: "openai/gpt-5.4-fast",
17284
+ name: "GPT 5.4 (Fast)",
17285
+ api: "anthropic-messages",
17286
+ provider: "vercel-ai-gateway",
17287
+ baseUrl: "https://ai-gateway.vercel.sh",
17288
+ reasoning: true,
17289
+ input: ["text", "image"],
17290
+ cost: {
17291
+ input: 5,
17292
+ output: 30,
17293
+ cacheRead: 0.5,
17294
+ cacheWrite: 0,
17295
+ },
17296
+ contextWindow: 1050000,
17297
+ maxTokens: 128000,
17298
+ },
16990
17299
  "openai/gpt-5.4-mini": {
16991
17300
  id: "openai/gpt-5.4-mini",
16992
17301
  name: "GPT 5.4 Mini",
@@ -17004,6 +17313,23 @@ export const MODELS = {
17004
17313
  contextWindow: 400000,
17005
17314
  maxTokens: 128000,
17006
17315
  },
17316
+ "openai/gpt-5.4-mini-fast": {
17317
+ id: "openai/gpt-5.4-mini-fast",
17318
+ name: "GPT 5.4 Mini (Fast)",
17319
+ api: "anthropic-messages",
17320
+ provider: "vercel-ai-gateway",
17321
+ baseUrl: "https://ai-gateway.vercel.sh",
17322
+ reasoning: true,
17323
+ input: ["text", "image"],
17324
+ cost: {
17325
+ input: 1.5,
17326
+ output: 9,
17327
+ cacheRead: 0.15,
17328
+ cacheWrite: 0,
17329
+ },
17330
+ contextWindow: 400000,
17331
+ maxTokens: 128000,
17332
+ },
17007
17333
  "openai/gpt-5.4-nano": {
17008
17334
  id: "openai/gpt-5.4-nano",
17009
17335
  name: "GPT 5.4 Nano",
@@ -17055,6 +17381,23 @@ export const MODELS = {
17055
17381
  contextWindow: 1000000,
17056
17382
  maxTokens: 128000,
17057
17383
  },
17384
+ "openai/gpt-5.5-fast": {
17385
+ id: "openai/gpt-5.5-fast",
17386
+ name: "GPT 5.5 (Fast)",
17387
+ api: "anthropic-messages",
17388
+ provider: "vercel-ai-gateway",
17389
+ baseUrl: "https://ai-gateway.vercel.sh",
17390
+ reasoning: true,
17391
+ input: ["text", "image"],
17392
+ cost: {
17393
+ input: 12.5,
17394
+ output: 75,
17395
+ cacheRead: 1.25,
17396
+ cacheWrite: 0,
17397
+ },
17398
+ contextWindow: 1000000,
17399
+ maxTokens: 128000,
17400
+ },
17058
17401
  "openai/gpt-5.5-pro": {
17059
17402
  id: "openai/gpt-5.5-pro",
17060
17403
  name: "GPT 5.5 Pro",
@@ -17089,6 +17432,23 @@ export const MODELS = {
17089
17432
  contextWindow: 1050000,
17090
17433
  maxTokens: 128000,
17091
17434
  },
17435
+ "openai/gpt-5.6-luna-fast": {
17436
+ id: "openai/gpt-5.6-luna-fast",
17437
+ name: "GPT 5.6 Luna (Fast)",
17438
+ api: "anthropic-messages",
17439
+ provider: "vercel-ai-gateway",
17440
+ baseUrl: "https://ai-gateway.vercel.sh",
17441
+ reasoning: true,
17442
+ input: ["text", "image"],
17443
+ cost: {
17444
+ input: 0.39999999999999997,
17445
+ output: 2.4,
17446
+ cacheRead: 0.04,
17447
+ cacheWrite: 0.25,
17448
+ },
17449
+ contextWindow: 1050000,
17450
+ maxTokens: 128000,
17451
+ },
17092
17452
  "openai/gpt-5.6-sol": {
17093
17453
  id: "openai/gpt-5.6-sol",
17094
17454
  name: "GPT 5.6 Sol",
@@ -17097,11 +17457,28 @@ export const MODELS = {
17097
17457
  baseUrl: "https://ai-gateway.vercel.sh",
17098
17458
  reasoning: true,
17099
17459
  input: ["text", "image"],
17460
+ cost: {
17461
+ input: 2.5,
17462
+ output: 15,
17463
+ cacheRead: 0.25,
17464
+ cacheWrite: 3.125,
17465
+ },
17466
+ contextWindow: 1050000,
17467
+ maxTokens: 128000,
17468
+ },
17469
+ "openai/gpt-5.6-sol-fast": {
17470
+ id: "openai/gpt-5.6-sol-fast",
17471
+ name: "GPT 5.6 Sol (Fast)",
17472
+ api: "anthropic-messages",
17473
+ provider: "vercel-ai-gateway",
17474
+ baseUrl: "https://ai-gateway.vercel.sh",
17475
+ reasoning: true,
17476
+ input: ["text", "image"],
17100
17477
  cost: {
17101
17478
  input: 5,
17102
17479
  output: 30,
17103
17480
  cacheRead: 0.5,
17104
- cacheWrite: 6.25,
17481
+ cacheWrite: 3.125,
17105
17482
  },
17106
17483
  contextWindow: 1050000,
17107
17484
  maxTokens: 128000,
@@ -17123,6 +17500,23 @@ export const MODELS = {
17123
17500
  contextWindow: 1050000,
17124
17501
  maxTokens: 128000,
17125
17502
  },
17503
+ "openai/gpt-5.6-terra-fast": {
17504
+ id: "openai/gpt-5.6-terra-fast",
17505
+ name: "GPT 5.6 Terra (Fast)",
17506
+ api: "anthropic-messages",
17507
+ provider: "vercel-ai-gateway",
17508
+ baseUrl: "https://ai-gateway.vercel.sh",
17509
+ reasoning: true,
17510
+ input: ["text", "image"],
17511
+ cost: {
17512
+ input: 4,
17513
+ output: 24,
17514
+ cacheRead: 0.39999999999999997,
17515
+ cacheWrite: 2.5,
17516
+ },
17517
+ contextWindow: 1050000,
17518
+ maxTokens: 128000,
17519
+ },
17126
17520
  "openai/gpt-oss-120b": {
17127
17521
  id: "openai/gpt-oss-120b",
17128
17522
  name: "GPT OSS 120B",
@@ -17225,6 +17619,23 @@ export const MODELS = {
17225
17619
  contextWindow: 200000,
17226
17620
  maxTokens: 100000,
17227
17621
  },
17622
+ "openai/o3-fast": {
17623
+ id: "openai/o3-fast",
17624
+ name: "o3 (Fast)",
17625
+ api: "anthropic-messages",
17626
+ provider: "vercel-ai-gateway",
17627
+ baseUrl: "https://ai-gateway.vercel.sh",
17628
+ reasoning: true,
17629
+ input: ["text", "image"],
17630
+ cost: {
17631
+ input: 3.5,
17632
+ output: 14,
17633
+ cacheRead: 0.875,
17634
+ cacheWrite: 0,
17635
+ },
17636
+ contextWindow: 200000,
17637
+ maxTokens: 100000,
17638
+ },
17228
17639
  "openai/o3-mini": {
17229
17640
  id: "openai/o3-mini",
17230
17641
  name: "o3-mini",
@@ -17276,6 +17687,23 @@ export const MODELS = {
17276
17687
  contextWindow: 200000,
17277
17688
  maxTokens: 100000,
17278
17689
  },
17690
+ "openai/o4-mini-fast": {
17691
+ id: "openai/o4-mini-fast",
17692
+ name: "o4-mini (Fast)",
17693
+ api: "anthropic-messages",
17694
+ provider: "vercel-ai-gateway",
17695
+ baseUrl: "https://ai-gateway.vercel.sh",
17696
+ reasoning: true,
17697
+ input: ["text", "image"],
17698
+ cost: {
17699
+ input: 2,
17700
+ output: 8,
17701
+ cacheRead: 0.5,
17702
+ cacheWrite: 0,
17703
+ },
17704
+ contextWindow: 200000,
17705
+ maxTokens: 100000,
17706
+ },
17279
17707
  "poolside/laguna-s-2.1": {
17280
17708
  id: "poolside/laguna-s-2.1",
17281
17709
  name: "Laguna S 2.1",