@dreb/ai 2.57.1 → 2.58.0

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",
@@ -4886,6 +4937,24 @@ export const MODELS = {
4886
4937
  contextWindow: 262144,
4887
4938
  maxTokens: 65536,
4888
4939
  },
4940
+ "Qwen/Qwen3.8-2.4T-A95B": {
4941
+ id: "Qwen/Qwen3.8-2.4T-A95B",
4942
+ name: "Qwen3.8 2.4T A95B",
4943
+ api: "openai-completions",
4944
+ provider: "huggingface",
4945
+ baseUrl: "https://router.huggingface.co/v1",
4946
+ compat: { "supportsDeveloperRole": false },
4947
+ reasoning: true,
4948
+ input: ["text"],
4949
+ cost: {
4950
+ input: 2.5,
4951
+ output: 6.25,
4952
+ cacheRead: 0,
4953
+ cacheWrite: 0,
4954
+ },
4955
+ contextWindow: 262144,
4956
+ maxTokens: 131072,
4957
+ },
4889
4958
  "XiaomiMiMo/MiMo-V2-Flash": {
4890
4959
  id: "XiaomiMiMo/MiMo-V2-Flash",
4891
4960
  name: "MiMo-V2-Flash",
@@ -8281,16 +8350,16 @@ export const MODELS = {
8281
8350
  "opencode-go": {
8282
8351
  "deepseek-v4-flash": {
8283
8352
  id: "deepseek-v4-flash",
8284
- name: "DeepSeek V4 Flash (2x usage)",
8353
+ name: "DeepSeek V4 Flash",
8285
8354
  api: "openai-completions",
8286
8355
  provider: "opencode-go",
8287
8356
  baseUrl: "https://opencode.ai/zen/go/v1",
8288
8357
  reasoning: true,
8289
8358
  input: ["text"],
8290
8359
  cost: {
8291
- input: 0.07,
8292
- output: 0.14,
8293
- cacheRead: 0.0014,
8360
+ input: 0.22,
8361
+ output: 0.66,
8362
+ cacheRead: 0.007,
8294
8363
  cacheWrite: 0,
8295
8364
  },
8296
8365
  contextWindow: 1000000,
@@ -8305,9 +8374,9 @@ export const MODELS = {
8305
8374
  reasoning: true,
8306
8375
  input: ["text"],
8307
8376
  cost: {
8308
- input: 0.435,
8309
- output: 0.87,
8310
- cacheRead: 0.003625,
8377
+ input: 0.66,
8378
+ output: 1.98,
8379
+ cacheRead: 0.022,
8311
8380
  cacheWrite: 0,
8312
8381
  },
8313
8382
  contextWindow: 1000000,
@@ -8537,9 +8606,9 @@ export const MODELS = {
8537
8606
  "qwen3.6-plus": {
8538
8607
  id: "qwen3.6-plus",
8539
8608
  name: "Qwen3.6 Plus",
8540
- api: "anthropic-messages",
8609
+ api: "openai-completions",
8541
8610
  provider: "opencode-go",
8542
- baseUrl: "https://opencode.ai/zen/go",
8611
+ baseUrl: "https://opencode.ai/zen/go/v1",
8543
8612
  reasoning: true,
8544
8613
  input: ["text", "image"],
8545
8614
  cost: {
@@ -8554,9 +8623,9 @@ export const MODELS = {
8554
8623
  "qwen3.7-max": {
8555
8624
  id: "qwen3.7-max",
8556
8625
  name: "Qwen3.7 Max",
8557
- api: "anthropic-messages",
8626
+ api: "openai-completions",
8558
8627
  provider: "opencode-go",
8559
- baseUrl: "https://opencode.ai/zen/go",
8628
+ baseUrl: "https://opencode.ai/zen/go/v1",
8560
8629
  reasoning: true,
8561
8630
  input: ["text"],
8562
8631
  cost: {
@@ -8571,9 +8640,9 @@ export const MODELS = {
8571
8640
  "qwen3.7-plus": {
8572
8641
  id: "qwen3.7-plus",
8573
8642
  name: "Qwen3.7 Plus",
8574
- api: "anthropic-messages",
8643
+ api: "openai-completions",
8575
8644
  provider: "opencode-go",
8576
- baseUrl: "https://opencode.ai/zen/go",
8645
+ baseUrl: "https://opencode.ai/zen/go/v1",
8577
8646
  reasoning: true,
8578
8647
  input: ["text", "image"],
8579
8648
  cost: {
@@ -8588,9 +8657,9 @@ export const MODELS = {
8588
8657
  "qwen3.8-max": {
8589
8658
  id: "qwen3.8-max",
8590
8659
  name: "Qwen3.8 Max",
8591
- api: "anthropic-messages",
8660
+ api: "openai-completions",
8592
8661
  provider: "opencode-go",
8593
- baseUrl: "https://opencode.ai/zen/go",
8662
+ baseUrl: "https://opencode.ai/zen/go/v1",
8594
8663
  reasoning: true,
8595
8664
  input: ["text", "image"],
8596
8665
  cost: {
@@ -9532,12 +9601,12 @@ export const MODELS = {
9532
9601
  input: ["text"],
9533
9602
  cost: {
9534
9603
  input: 0.27,
9535
- output: 0.95,
9536
- cacheRead: 0.13,
9604
+ output: 1,
9605
+ cacheRead: 0,
9537
9606
  cacheWrite: 0,
9538
9607
  },
9539
9608
  contextWindow: 163840,
9540
- maxTokens: 32768,
9609
+ maxTokens: 163840,
9541
9610
  },
9542
9611
  "deepseek/deepseek-v3.2": {
9543
9612
  id: "deepseek/deepseek-v3.2",
@@ -9582,9 +9651,9 @@ export const MODELS = {
9582
9651
  reasoning: true,
9583
9652
  input: ["text"],
9584
9653
  cost: {
9585
- input: 0.06426,
9586
- output: 0.12852,
9587
- cacheRead: 0.012852,
9654
+ input: 0.077,
9655
+ output: 0.154,
9656
+ cacheRead: 0.015399999999999999,
9588
9657
  cacheWrite: 0,
9589
9658
  },
9590
9659
  contextWindow: 1048576,
@@ -9604,25 +9673,25 @@ export const MODELS = {
9604
9673
  cacheRead: 0.028,
9605
9674
  cacheWrite: 0,
9606
9675
  },
9607
- contextWindow: 1048576,
9676
+ contextWindow: 1310720,
9608
9677
  maxTokens: 393216,
9609
9678
  },
9610
9679
  "deepseek/deepseek-v4-pro": {
9611
9680
  id: "deepseek/deepseek-v4-pro",
9612
- name: "DeepSeek: DeepSeek V4 Pro",
9681
+ name: "DeepSeek: DeepSeek V4 Pro 0423",
9613
9682
  api: "openai-completions",
9614
9683
  provider: "openrouter",
9615
9684
  baseUrl: "https://openrouter.ai/api/v1",
9616
9685
  reasoning: true,
9617
9686
  input: ["text"],
9618
9687
  cost: {
9619
- input: 1.1680000000000001,
9620
- output: 2.3360000000000003,
9621
- cacheRead: 0.09855000000000001,
9688
+ input: 0.66,
9689
+ output: 1.9800000000000002,
9690
+ cacheRead: 0.022,
9622
9691
  cacheWrite: 0,
9623
9692
  },
9624
9693
  contextWindow: 1048576,
9625
- maxTokens: 393216,
9694
+ maxTokens: 384000,
9626
9695
  },
9627
9696
  "deepseek/deepseek-v4-pro-0813": {
9628
9697
  id: "deepseek/deepseek-v4-pro-0813",
@@ -9633,9 +9702,9 @@ export const MODELS = {
9633
9702
  reasoning: true,
9634
9703
  input: ["text"],
9635
9704
  cost: {
9636
- input: 0.435,
9637
- output: 0.87,
9638
- cacheRead: 0.003625,
9705
+ input: 0.66,
9706
+ output: 1.9800000000000002,
9707
+ cacheRead: 0.022,
9639
9708
  cacheWrite: 0,
9640
9709
  },
9641
9710
  contextWindow: 1048576,
@@ -10126,13 +10195,13 @@ export const MODELS = {
10126
10195
  reasoning: true,
10127
10196
  input: ["text", "image"],
10128
10197
  cost: {
10129
- input: 0.12,
10130
- output: 0.39999999999999997,
10131
- cacheRead: 0.049999999999999996,
10198
+ input: 0.07,
10199
+ output: 0.33999999999999997,
10200
+ cacheRead: 0,
10132
10201
  cacheWrite: 0,
10133
10202
  },
10134
10203
  contextWindow: 262144,
10135
- maxTokens: 262144,
10204
+ maxTokens: 16384,
10136
10205
  },
10137
10206
  "google/gemma-4-26b-a4b-it:free": {
10138
10207
  id: "google/gemma-4-26b-a4b-it:free",
@@ -10976,9 +11045,9 @@ export const MODELS = {
10976
11045
  reasoning: true,
10977
11046
  input: ["text", "image"],
10978
11047
  cost: {
10979
- input: 0.5415,
10980
- output: 2.2800000000000002,
10981
- cacheRead: 0.09119999999999999,
11048
+ input: 0.5605,
11049
+ output: 2.36,
11050
+ cacheRead: 0.0944,
10982
11051
  cacheWrite: 0,
10983
11052
  },
10984
11053
  contextWindow: 262144,
@@ -11080,11 +11149,11 @@ export const MODELS = {
11080
11149
  cost: {
11081
11150
  input: 0.049999999999999996,
11082
11151
  output: 0.19999999999999998,
11083
- cacheRead: 0.024999999999999998,
11152
+ cacheRead: 0.03,
11084
11153
  cacheWrite: 0,
11085
11154
  },
11086
11155
  contextWindow: 262144,
11087
- maxTokens: 228000,
11156
+ maxTokens: 262144,
11088
11157
  },
11089
11158
  "nvidia/nemotron-3-nano-30b-a3b:free": {
11090
11159
  id: "nvidia/nemotron-3-nano-30b-a3b:free",
@@ -11214,13 +11283,13 @@ export const MODELS = {
11214
11283
  reasoning: true,
11215
11284
  input: ["text"],
11216
11285
  cost: {
11217
- input: 0.09999999999999999,
11218
- output: 0.25,
11219
- cacheRead: 0.049999999999999996,
11286
+ input: 0.08,
11287
+ output: 0.19999999999999998,
11288
+ cacheRead: 0.04,
11220
11289
  cacheWrite: 0,
11221
11290
  },
11222
11291
  contextWindow: 1000000,
11223
- maxTokens: 262144,
11292
+ maxTokens: 131072,
11224
11293
  },
11225
11294
  "nvidia/nemotron-3.5-lightning:free": {
11226
11295
  id: "nvidia/nemotron-3.5-lightning:free",
@@ -12200,10 +12269,10 @@ export const MODELS = {
12200
12269
  reasoning: true,
12201
12270
  input: ["text", "image"],
12202
12271
  cost: {
12203
- input: 0.09999999999999999,
12204
- output: 0.6,
12205
- cacheRead: 0.01,
12206
- cacheWrite: 0.125,
12272
+ input: 0.19999999999999998,
12273
+ output: 1.2,
12274
+ cacheRead: 0.02,
12275
+ cacheWrite: 0.25,
12207
12276
  },
12208
12277
  contextWindow: 1050000,
12209
12278
  maxTokens: 128000,
@@ -12217,10 +12286,10 @@ export const MODELS = {
12217
12286
  reasoning: true,
12218
12287
  input: ["text", "image"],
12219
12288
  cost: {
12220
- input: 0.09999999999999999,
12221
- output: 0.6,
12222
- cacheRead: 0.01,
12223
- cacheWrite: 0.125,
12289
+ input: 0.19999999999999998,
12290
+ output: 1.2,
12291
+ cacheRead: 0.02,
12292
+ cacheWrite: 0.25,
12224
12293
  },
12225
12294
  contextWindow: 1050000,
12226
12295
  maxTokens: 128000,
@@ -12268,10 +12337,10 @@ export const MODELS = {
12268
12337
  reasoning: true,
12269
12338
  input: ["text", "image"],
12270
12339
  cost: {
12271
- input: 5,
12272
- output: 30,
12273
- cacheRead: 0.5,
12274
- cacheWrite: 6.25,
12340
+ input: 2.5,
12341
+ output: 15,
12342
+ cacheRead: 0.25,
12343
+ cacheWrite: 3.125,
12275
12344
  },
12276
12345
  contextWindow: 1050000,
12277
12346
  maxTokens: 128000,
@@ -12285,10 +12354,10 @@ export const MODELS = {
12285
12354
  reasoning: true,
12286
12355
  input: ["text", "image"],
12287
12356
  cost: {
12288
- input: 5,
12289
- output: 30,
12290
- cacheRead: 0.5,
12291
- cacheWrite: 6.25,
12357
+ input: 2.5,
12358
+ output: 15,
12359
+ cacheRead: 0.25,
12360
+ cacheWrite: 3.125,
12292
12361
  },
12293
12362
  contextWindow: 1050000,
12294
12363
  maxTokens: 128000,
@@ -12302,9 +12371,9 @@ export const MODELS = {
12302
12371
  reasoning: true,
12303
12372
  input: ["text", "image"],
12304
12373
  cost: {
12305
- input: 2.5,
12306
- output: 15,
12307
- cacheRead: 0.25,
12374
+ input: 1.25,
12375
+ output: 7.5,
12376
+ cacheRead: 0.125,
12308
12377
  cacheWrite: 0,
12309
12378
  },
12310
12379
  contextWindow: 1050000,
@@ -12319,9 +12388,9 @@ export const MODELS = {
12319
12388
  reasoning: true,
12320
12389
  input: ["text", "image"],
12321
12390
  cost: {
12322
- input: 2.5,
12323
- output: 15,
12324
- cacheRead: 0.25,
12391
+ input: 1.25,
12392
+ output: 7.5,
12393
+ cacheRead: 0.125,
12325
12394
  cacheWrite: 0,
12326
12395
  },
12327
12396
  contextWindow: 1050000,
@@ -12336,10 +12405,10 @@ export const MODELS = {
12336
12405
  reasoning: true,
12337
12406
  input: ["text", "image"],
12338
12407
  cost: {
12339
- input: 1,
12340
- output: 6,
12341
- cacheRead: 0.09999999999999999,
12342
- cacheWrite: 1.25,
12408
+ input: 2,
12409
+ output: 12,
12410
+ cacheRead: 0.19999999999999998,
12411
+ cacheWrite: 2.5,
12343
12412
  },
12344
12413
  contextWindow: 1050000,
12345
12414
  maxTokens: 128000,
@@ -12353,10 +12422,10 @@ export const MODELS = {
12353
12422
  reasoning: true,
12354
12423
  input: ["text", "image"],
12355
12424
  cost: {
12356
- input: 1,
12357
- output: 6,
12358
- cacheRead: 0.09999999999999999,
12359
- cacheWrite: 1.25,
12425
+ input: 2,
12426
+ output: 12,
12427
+ cacheRead: 0.19999999999999998,
12428
+ cacheWrite: 2.5,
12360
12429
  },
12361
12430
  contextWindow: 1050000,
12362
12431
  maxTokens: 128000,
@@ -13050,13 +13119,13 @@ export const MODELS = {
13050
13119
  reasoning: true,
13051
13120
  input: ["text"],
13052
13121
  cost: {
13053
- input: 0.12,
13054
- output: 0.5,
13122
+ input: 0.13,
13123
+ output: 0.52,
13055
13124
  cacheRead: 0,
13056
13125
  cacheWrite: 0,
13057
13126
  },
13058
13127
  contextWindow: 131072,
13059
- maxTokens: 16384,
13128
+ maxTokens: 8192,
13060
13129
  },
13061
13130
  "qwen/qwen3-30b-a3b-instruct-2507": {
13062
13131
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -13288,9 +13357,9 @@ export const MODELS = {
13288
13357
  reasoning: false,
13289
13358
  input: ["text", "image"],
13290
13359
  cost: {
13291
- input: 0.26,
13292
- output: 1.04,
13293
- cacheRead: 0,
13360
+ input: 0.21,
13361
+ output: 1.9,
13362
+ cacheRead: 0.09999999999999999,
13294
13363
  cacheWrite: 0,
13295
13364
  },
13296
13365
  contextWindow: 262144,
@@ -13543,13 +13612,13 @@ export const MODELS = {
13543
13612
  reasoning: true,
13544
13613
  input: ["text", "image"],
13545
13614
  cost: {
13546
- input: 0.6,
13547
- output: 3.5999999999999996,
13548
- cacheRead: 0.12,
13615
+ input: 0.28900000000000003,
13616
+ output: 2.4,
13617
+ cacheRead: 0,
13549
13618
  cacheWrite: 0,
13550
13619
  },
13551
13620
  contextWindow: 262144,
13552
- maxTokens: 262144,
13621
+ maxTokens: 131072,
13553
13622
  },
13554
13623
  "qwen/qwen3.6-35b-a3b": {
13555
13624
  id: "qwen/qwen3.6-35b-a3b",
@@ -13560,7 +13629,7 @@ export const MODELS = {
13560
13629
  reasoning: true,
13561
13630
  input: ["text", "image"],
13562
13631
  cost: {
13563
- input: 0.15,
13632
+ input: 0.14,
13564
13633
  output: 1,
13565
13634
  cacheRead: 0.049999999999999996,
13566
13635
  cacheWrite: 0,
@@ -13698,7 +13767,7 @@ export const MODELS = {
13698
13767
  cost: {
13699
13768
  input: 0.44999999999999996,
13700
13769
  output: 3.1999999999999997,
13701
- cacheRead: 0,
13770
+ cacheRead: 0.049999999999999996,
13702
13771
  cacheWrite: 0,
13703
13772
  },
13704
13773
  contextWindow: 262144,
@@ -14155,9 +14224,9 @@ export const MODELS = {
14155
14224
  reasoning: true,
14156
14225
  input: ["text"],
14157
14226
  cost: {
14158
- input: 0.55,
14159
- output: 2.2,
14160
- cacheRead: 0.11,
14227
+ input: 0.5,
14228
+ output: 2,
14229
+ cacheRead: 0.09999999999999999,
14161
14230
  cacheWrite: 0,
14162
14231
  },
14163
14232
  contextWindow: 204800,
@@ -14274,9 +14343,9 @@ export const MODELS = {
14274
14343
  reasoning: true,
14275
14344
  input: ["text"],
14276
14345
  cost: {
14277
- input: 0.46199999999999997,
14278
- output: 1.452,
14279
- cacheRead: 0.0858,
14346
+ input: 0.476,
14347
+ output: 1.496,
14348
+ cacheRead: 0.08839999999999999,
14280
14349
  cacheWrite: 0,
14281
14350
  },
14282
14351
  contextWindow: 1048576,
@@ -14393,13 +14462,13 @@ export const MODELS = {
14393
14462
  reasoning: true,
14394
14463
  input: ["text"],
14395
14464
  cost: {
14396
- input: 0.0675,
14397
- output: 0.135,
14398
- cacheRead: 0.0135,
14465
+ input: 0.0784,
14466
+ output: 0.1568,
14467
+ cacheRead: 0.01568,
14399
14468
  cacheWrite: 0,
14400
14469
  },
14401
- contextWindow: 1048576,
14402
- maxTokens: 262144,
14470
+ contextWindow: 1310720,
14471
+ maxTokens: 384000,
14403
14472
  },
14404
14473
  "~google/gemini-flash-latest": {
14405
14474
  id: "~google/gemini-flash-latest",
@@ -14444,13 +14513,13 @@ export const MODELS = {
14444
14513
  reasoning: true,
14445
14514
  input: ["text", "image"],
14446
14515
  cost: {
14447
- input: 2.8,
14448
- output: 14,
14516
+ input: 2.6,
14517
+ output: 13,
14449
14518
  cacheRead: 0.29,
14450
14519
  cacheWrite: 0,
14451
14520
  },
14452
14521
  contextWindow: 1048576,
14453
- maxTokens: 1048576,
14522
+ maxTokens: 974842,
14454
14523
  },
14455
14524
  "~openai/gpt-latest": {
14456
14525
  id: "~openai/gpt-latest",
@@ -14461,10 +14530,10 @@ export const MODELS = {
14461
14530
  reasoning: true,
14462
14531
  input: ["text", "image"],
14463
14532
  cost: {
14464
- input: 5,
14465
- output: 30,
14466
- cacheRead: 0.5,
14467
- cacheWrite: 6.25,
14533
+ input: 2.5,
14534
+ output: 15,
14535
+ cacheRead: 0.25,
14536
+ cacheWrite: 3.125,
14468
14537
  },
14469
14538
  contextWindow: 1050000,
14470
14539
  maxTokens: 128000,
@@ -14941,12 +15010,29 @@ export const MODELS = {
14941
15010
  cost: {
14942
15011
  input: 2,
14943
15012
  output: 6,
14944
- cacheRead: 0.25,
15013
+ cacheRead: 0.19999999999999998,
14945
15014
  cacheWrite: 0,
14946
15015
  },
14947
15016
  contextWindow: 262144,
14948
15017
  maxTokens: 131072,
14949
15018
  },
15019
+ "alibaba/qwen3.8-27b": {
15020
+ id: "alibaba/qwen3.8-27b",
15021
+ name: "Qwen3.8 27B",
15022
+ api: "anthropic-messages",
15023
+ provider: "vercel-ai-gateway",
15024
+ baseUrl: "https://ai-gateway.vercel.sh",
15025
+ reasoning: true,
15026
+ input: ["text", "image"],
15027
+ cost: {
15028
+ input: 0,
15029
+ output: 0,
15030
+ cacheRead: 0,
15031
+ cacheWrite: 0,
15032
+ },
15033
+ contextWindow: 262144,
15034
+ maxTokens: 262133,
15035
+ },
14950
15036
  "alibaba/qwen3.8-max": {
14951
15037
  id: "alibaba/qwen3.8-max",
14952
15038
  name: "Qwen 3.8 Max",
@@ -15483,9 +15569,9 @@ export const MODELS = {
15483
15569
  reasoning: true,
15484
15570
  input: ["text"],
15485
15571
  cost: {
15486
- input: 0.19999999999999998,
15487
- output: 0.39999999999999997,
15488
- cacheRead: 0.04,
15572
+ input: 0.13,
15573
+ output: 0.26,
15574
+ cacheRead: 0.028,
15489
15575
  cacheWrite: 0,
15490
15576
  },
15491
15577
  contextWindow: 1000000,
@@ -15500,9 +15586,9 @@ export const MODELS = {
15500
15586
  reasoning: true,
15501
15587
  input: ["text"],
15502
15588
  cost: {
15503
- input: 0.19999999999999998,
15504
- output: 0.39999999999999997,
15505
- cacheRead: 0.04,
15589
+ input: 0.13,
15590
+ output: 0.26,
15591
+ cacheRead: 0.028,
15506
15592
  cacheWrite: 0,
15507
15593
  },
15508
15594
  contextWindow: 1000000,
@@ -15687,9 +15773,9 @@ export const MODELS = {
15687
15773
  reasoning: true,
15688
15774
  input: ["text", "image"],
15689
15775
  cost: {
15690
- input: 1.5,
15691
- output: 7.5,
15692
- cacheRead: 0.15,
15776
+ input: 0.75,
15777
+ output: 3.75,
15778
+ cacheRead: 0.075,
15693
15779
  cacheWrite: 0,
15694
15780
  },
15695
15781
  contextWindow: 1000000,
@@ -16596,6 +16682,23 @@ export const MODELS = {
16596
16682
  contextWindow: 1000000,
16597
16683
  maxTokens: 65000,
16598
16684
  },
16685
+ "nvidia/nemotron-3.5-lightning": {
16686
+ id: "nvidia/nemotron-3.5-lightning",
16687
+ name: "Nemotron 3.5 Lightning 30B",
16688
+ api: "anthropic-messages",
16689
+ provider: "vercel-ai-gateway",
16690
+ baseUrl: "https://ai-gateway.vercel.sh",
16691
+ reasoning: true,
16692
+ input: ["text"],
16693
+ cost: {
16694
+ input: 0,
16695
+ output: 0,
16696
+ cacheRead: 0,
16697
+ cacheWrite: 0,
16698
+ },
16699
+ contextWindow: 262144,
16700
+ maxTokens: 131072,
16701
+ },
16599
16702
  "nvidia/nemotron-nano-12b-v2-vl": {
16600
16703
  id: "nvidia/nemotron-nano-12b-v2-vl",
16601
16704
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -17098,10 +17201,10 @@ export const MODELS = {
17098
17201
  reasoning: true,
17099
17202
  input: ["text", "image"],
17100
17203
  cost: {
17101
- input: 5,
17102
- output: 30,
17103
- cacheRead: 0.5,
17104
- cacheWrite: 6.25,
17204
+ input: 2.5,
17205
+ output: 15,
17206
+ cacheRead: 0.25,
17207
+ cacheWrite: 3.125,
17105
17208
  },
17106
17209
  contextWindow: 1050000,
17107
17210
  maxTokens: 128000,