@elyracode/ai 0.9.36 → 0.9.38

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.
@@ -13,11 +13,11 @@ export const MODELS = {
13
13
  cost: {
14
14
  input: 0.33,
15
15
  output: 2.75,
16
- cacheRead: 0,
17
- cacheWrite: 0,
16
+ cacheRead: 0.0825,
17
+ cacheWrite: 0.33,
18
18
  },
19
- contextWindow: 128000,
20
- maxTokens: 4096,
19
+ contextWindow: 1000000,
20
+ maxTokens: 65535,
21
21
  },
22
22
  "amazon.nova-lite-v1:0": {
23
23
  id: "amazon.nova-lite-v1:0",
@@ -31,10 +31,10 @@ export const MODELS = {
31
31
  input: 0.06,
32
32
  output: 0.24,
33
33
  cacheRead: 0.015,
34
- cacheWrite: 0,
34
+ cacheWrite: 0.06,
35
35
  },
36
36
  contextWindow: 300000,
37
- maxTokens: 8192,
37
+ maxTokens: 10000,
38
38
  },
39
39
  "amazon.nova-micro-v1:0": {
40
40
  id: "amazon.nova-micro-v1:0",
@@ -48,10 +48,10 @@ export const MODELS = {
48
48
  input: 0.035,
49
49
  output: 0.14,
50
50
  cacheRead: 0.00875,
51
- cacheWrite: 0,
51
+ cacheWrite: 0.035,
52
52
  },
53
53
  contextWindow: 128000,
54
- maxTokens: 8192,
54
+ maxTokens: 10000,
55
55
  },
56
56
  "amazon.nova-pro-v1:0": {
57
57
  id: "amazon.nova-pro-v1:0",
@@ -65,10 +65,10 @@ export const MODELS = {
65
65
  input: 0.8,
66
66
  output: 3.2,
67
67
  cacheRead: 0.2,
68
- cacheWrite: 0,
68
+ cacheWrite: 0.8,
69
69
  },
70
70
  contextWindow: 300000,
71
- maxTokens: 8192,
71
+ maxTokens: 10000,
72
72
  },
73
73
  "anthropic.claude-fable-5": {
74
74
  id: "anthropic.claude-fable-5",
@@ -173,8 +173,8 @@ export const MODELS = {
173
173
  thinkingLevelMap: { "xhigh": "max" },
174
174
  input: ["text", "image"],
175
175
  cost: {
176
- input: 5,
177
- output: 25,
176
+ input: 5.5,
177
+ output: 27.5,
178
178
  cacheRead: 0.5,
179
179
  cacheWrite: 6.25,
180
180
  },
@@ -265,13 +265,13 @@ export const MODELS = {
265
265
  thinkingType: "adaptive",
266
266
  input: ["text", "image"],
267
267
  cost: {
268
- input: 3,
269
- output: 15,
270
- cacheRead: 0.3,
271
- cacheWrite: 3.75,
268
+ input: 3.3,
269
+ output: 16.5,
270
+ cacheRead: 0.33,
271
+ cacheWrite: 4.125,
272
272
  },
273
273
  contextWindow: 1000000,
274
- maxTokens: 64000,
274
+ maxTokens: 128000,
275
275
  },
276
276
  "anthropic.claude-sonnet-5": {
277
277
  id: "anthropic.claude-sonnet-5",
@@ -291,6 +291,75 @@ export const MODELS = {
291
291
  contextWindow: 1000000,
292
292
  maxTokens: 128000,
293
293
  },
294
+ "apac.amazon.nova-lite-v1:0": {
295
+ id: "apac.amazon.nova-lite-v1:0",
296
+ name: "Nova Lite (APAC)",
297
+ api: "bedrock-converse-stream",
298
+ provider: "amazon-bedrock",
299
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
300
+ reasoning: false,
301
+ input: ["text", "image"],
302
+ cost: {
303
+ input: 0.063,
304
+ output: 0.252,
305
+ cacheRead: 0.01575,
306
+ cacheWrite: 0.063,
307
+ },
308
+ contextWindow: 300000,
309
+ maxTokens: 10000,
310
+ },
311
+ "apac.amazon.nova-micro-v1:0": {
312
+ id: "apac.amazon.nova-micro-v1:0",
313
+ name: "Nova Micro (APAC)",
314
+ api: "bedrock-converse-stream",
315
+ provider: "amazon-bedrock",
316
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
317
+ reasoning: false,
318
+ input: ["text"],
319
+ cost: {
320
+ input: 0.037,
321
+ output: 0.148,
322
+ cacheRead: 0.00925,
323
+ cacheWrite: 0.037,
324
+ },
325
+ contextWindow: 128000,
326
+ maxTokens: 10000,
327
+ },
328
+ "apac.amazon.nova-pro-v1:0": {
329
+ id: "apac.amazon.nova-pro-v1:0",
330
+ name: "Nova Pro (APAC)",
331
+ api: "bedrock-converse-stream",
332
+ provider: "amazon-bedrock",
333
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
334
+ reasoning: false,
335
+ input: ["text", "image"],
336
+ cost: {
337
+ input: 0.84,
338
+ output: 3.36,
339
+ cacheRead: 0.21,
340
+ cacheWrite: 0.84,
341
+ },
342
+ contextWindow: 300000,
343
+ maxTokens: 10000,
344
+ },
345
+ "apac.anthropic.claude-sonnet-4-20250514-v1:0": {
346
+ id: "apac.anthropic.claude-sonnet-4-20250514-v1:0",
347
+ name: "Claude Sonnet 4 (APAC)",
348
+ api: "bedrock-converse-stream",
349
+ provider: "amazon-bedrock",
350
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
351
+ reasoning: true,
352
+ thinkingType: "budget",
353
+ input: ["text", "image"],
354
+ cost: {
355
+ input: 3,
356
+ output: 15,
357
+ cacheRead: 0.3,
358
+ cacheWrite: 3.75,
359
+ },
360
+ contextWindow: 200000,
361
+ maxTokens: 64000,
362
+ },
294
363
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
295
364
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
296
365
  name: "Claude Haiku 4.5 (AU)",
@@ -301,10 +370,10 @@ export const MODELS = {
301
370
  thinkingType: "budget",
302
371
  input: ["text", "image"],
303
372
  cost: {
304
- input: 1,
305
- output: 5,
306
- cacheRead: 0.1,
307
- cacheWrite: 1.25,
373
+ input: 1.1,
374
+ output: 5.5,
375
+ cacheRead: 0.11,
376
+ cacheWrite: 1.375,
308
377
  },
309
378
  contextWindow: 200000,
310
379
  maxTokens: 64000,
@@ -320,14 +389,33 @@ export const MODELS = {
320
389
  thinkingLevelMap: { "xhigh": "max" },
321
390
  input: ["text", "image"],
322
391
  cost: {
323
- input: 16.5,
324
- output: 82.5,
392
+ input: 5.5,
393
+ output: 27.5,
325
394
  cacheRead: 0.5,
326
395
  cacheWrite: 6.25,
327
396
  },
328
397
  contextWindow: 1000000,
329
398
  maxTokens: 128000,
330
399
  },
400
+ "au.anthropic.claude-opus-4-7": {
401
+ id: "au.anthropic.claude-opus-4-7",
402
+ name: "Claude Opus 4.7 (AU)",
403
+ api: "bedrock-converse-stream",
404
+ provider: "amazon-bedrock",
405
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
406
+ reasoning: true,
407
+ thinkingType: "adaptive",
408
+ thinkingLevelMap: { "xhigh": "xhigh" },
409
+ input: ["text", "image"],
410
+ cost: {
411
+ input: 5.5,
412
+ output: 27.5,
413
+ cacheRead: 0.55,
414
+ cacheWrite: 6.875,
415
+ },
416
+ contextWindow: 1000000,
417
+ maxTokens: 128000,
418
+ },
331
419
  "au.anthropic.claude-opus-4-8": {
332
420
  id: "au.anthropic.claude-opus-4-8",
333
421
  name: "Claude Opus 4.8 (AU)",
@@ -339,10 +427,10 @@ export const MODELS = {
339
427
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
340
428
  input: ["text", "image"],
341
429
  cost: {
342
- input: 5,
343
- output: 25,
344
- cacheRead: 0.5,
345
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
346
434
  },
347
435
  contextWindow: 1000000,
348
436
  maxTokens: 128000,
@@ -357,10 +445,10 @@ export const MODELS = {
357
445
  thinkingType: "adaptive",
358
446
  input: ["text", "image"],
359
447
  cost: {
360
- input: 5,
361
- output: 25,
362
- cacheRead: 0.5,
363
- cacheWrite: 6.25,
448
+ input: 5.5,
449
+ output: 27.5,
450
+ cacheRead: 0.55,
451
+ cacheWrite: 6.875,
364
452
  },
365
453
  contextWindow: 1000000,
366
454
  maxTokens: 128000,
@@ -375,10 +463,10 @@ export const MODELS = {
375
463
  thinkingType: "budget",
376
464
  input: ["text", "image"],
377
465
  cost: {
378
- input: 3,
379
- output: 15,
380
- cacheRead: 0.3,
381
- cacheWrite: 3.75,
466
+ input: 3.3,
467
+ output: 16.5,
468
+ cacheRead: 0.33,
469
+ cacheWrite: 4.125,
382
470
  },
383
471
  contextWindow: 200000,
384
472
  maxTokens: 64000,
@@ -411,30 +499,30 @@ export const MODELS = {
411
499
  thinkingType: "adaptive",
412
500
  input: ["text", "image"],
413
501
  cost: {
414
- input: 2,
415
- output: 10,
416
- cacheRead: 0.2,
417
- cacheWrite: 2.5,
502
+ input: 2.2,
503
+ output: 11,
504
+ cacheRead: 0.22,
505
+ cacheWrite: 2.75,
418
506
  },
419
507
  contextWindow: 1000000,
420
508
  maxTokens: 128000,
421
509
  },
422
- "deepseek.r1-v1:0": {
423
- id: "deepseek.r1-v1:0",
424
- name: "DeepSeek-R1",
510
+ "ca.amazon.nova-lite-v1:0": {
511
+ id: "ca.amazon.nova-lite-v1:0",
512
+ name: "Nova Lite (CA)",
425
513
  api: "bedrock-converse-stream",
426
514
  provider: "amazon-bedrock",
427
515
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
428
- reasoning: true,
429
- input: ["text"],
516
+ reasoning: false,
517
+ input: ["text", "image"],
430
518
  cost: {
431
- input: 1.35,
432
- output: 5.4,
433
- cacheRead: 0,
434
- cacheWrite: 0,
519
+ input: 0.064,
520
+ output: 0.256,
521
+ cacheRead: 0.016,
522
+ cacheWrite: 0.064,
435
523
  },
436
- contextWindow: 128000,
437
- maxTokens: 32768,
524
+ contextWindow: 300000,
525
+ maxTokens: 10000,
438
526
  },
439
527
  "deepseek.v3-v1:0": {
440
528
  id: "deepseek.v3-v1:0",
@@ -455,7 +543,7 @@ export const MODELS = {
455
543
  },
456
544
  "deepseek.v3.2": {
457
545
  id: "deepseek.v3.2",
458
- name: "DeepSeek-V3.2",
546
+ name: "DeepSeek V3.2",
459
547
  api: "bedrock-converse-stream",
460
548
  provider: "amazon-bedrock",
461
549
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -470,6 +558,74 @@ export const MODELS = {
470
558
  contextWindow: 163840,
471
559
  maxTokens: 81920,
472
560
  },
561
+ "eu.amazon.nova-2-lite-v1:0": {
562
+ id: "eu.amazon.nova-2-lite-v1:0",
563
+ name: "Nova 2 Lite (EU)",
564
+ api: "bedrock-converse-stream",
565
+ provider: "amazon-bedrock",
566
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
567
+ reasoning: true,
568
+ input: ["text", "image"],
569
+ cost: {
570
+ input: 0.374,
571
+ output: 3.157,
572
+ cacheRead: 0.0935,
573
+ cacheWrite: 0.374,
574
+ },
575
+ contextWindow: 1000000,
576
+ maxTokens: 65535,
577
+ },
578
+ "eu.amazon.nova-lite-v1:0": {
579
+ id: "eu.amazon.nova-lite-v1:0",
580
+ name: "Nova Lite (EU)",
581
+ api: "bedrock-converse-stream",
582
+ provider: "amazon-bedrock",
583
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
584
+ reasoning: false,
585
+ input: ["text", "image"],
586
+ cost: {
587
+ input: 0.069,
588
+ output: 0.276,
589
+ cacheRead: 0.01725,
590
+ cacheWrite: 0.069,
591
+ },
592
+ contextWindow: 300000,
593
+ maxTokens: 10000,
594
+ },
595
+ "eu.amazon.nova-micro-v1:0": {
596
+ id: "eu.amazon.nova-micro-v1:0",
597
+ name: "Nova Micro (EU)",
598
+ api: "bedrock-converse-stream",
599
+ provider: "amazon-bedrock",
600
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
601
+ reasoning: false,
602
+ input: ["text"],
603
+ cost: {
604
+ input: 0.04,
605
+ output: 0.16,
606
+ cacheRead: 0.01,
607
+ cacheWrite: 0.04,
608
+ },
609
+ contextWindow: 128000,
610
+ maxTokens: 10000,
611
+ },
612
+ "eu.amazon.nova-pro-v1:0": {
613
+ id: "eu.amazon.nova-pro-v1:0",
614
+ name: "Nova Pro (EU)",
615
+ api: "bedrock-converse-stream",
616
+ provider: "amazon-bedrock",
617
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
618
+ reasoning: false,
619
+ input: ["text", "image"],
620
+ cost: {
621
+ input: 0.92,
622
+ output: 3.68,
623
+ cacheRead: 0.23,
624
+ cacheWrite: 0.92,
625
+ },
626
+ contextWindow: 300000,
627
+ maxTokens: 10000,
628
+ },
473
629
  "eu.anthropic.claude-fable-5": {
474
630
  id: "eu.anthropic.claude-fable-5",
475
631
  name: "Claude Fable 5 (EU)",
@@ -600,6 +756,24 @@ export const MODELS = {
600
756
  contextWindow: 1000000,
601
757
  maxTokens: 128000,
602
758
  },
759
+ "eu.anthropic.claude-sonnet-4-20250514-v1:0": {
760
+ id: "eu.anthropic.claude-sonnet-4-20250514-v1:0",
761
+ name: "Claude Sonnet 4 (EU)",
762
+ api: "bedrock-converse-stream",
763
+ provider: "amazon-bedrock",
764
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
765
+ reasoning: true,
766
+ thinkingType: "budget",
767
+ input: ["text", "image"],
768
+ cost: {
769
+ input: 3,
770
+ output: 15,
771
+ cacheRead: 0.3,
772
+ cacheWrite: 3.75,
773
+ },
774
+ contextWindow: 200000,
775
+ maxTokens: 64000,
776
+ },
603
777
  "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
604
778
  id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
605
779
  name: "Claude Sonnet 4.5 (EU)",
@@ -634,7 +808,7 @@ export const MODELS = {
634
808
  cacheWrite: 4.125,
635
809
  },
636
810
  contextWindow: 1000000,
637
- maxTokens: 64000,
811
+ maxTokens: 128000,
638
812
  },
639
813
  "eu.anthropic.claude-sonnet-5": {
640
814
  id: "eu.anthropic.claude-sonnet-5",
@@ -654,6 +828,40 @@ export const MODELS = {
654
828
  contextWindow: 1000000,
655
829
  maxTokens: 128000,
656
830
  },
831
+ "eu.mistral.pixtral-large-2502-v1:0": {
832
+ id: "eu.mistral.pixtral-large-2502-v1:0",
833
+ name: "Pixtral Large (25.02) (EU)",
834
+ api: "bedrock-converse-stream",
835
+ provider: "amazon-bedrock",
836
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
837
+ reasoning: false,
838
+ input: ["text", "image"],
839
+ cost: {
840
+ input: 2,
841
+ output: 6,
842
+ cacheRead: 0,
843
+ cacheWrite: 0,
844
+ },
845
+ contextWindow: 128000,
846
+ maxTokens: 8192,
847
+ },
848
+ "global.amazon.nova-2-lite-v1:0": {
849
+ id: "global.amazon.nova-2-lite-v1:0",
850
+ name: "Nova 2 Lite (Global)",
851
+ api: "bedrock-converse-stream",
852
+ provider: "amazon-bedrock",
853
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
854
+ reasoning: true,
855
+ input: ["text", "image"],
856
+ cost: {
857
+ input: 0.3,
858
+ output: 2.5,
859
+ cacheRead: 0.075,
860
+ cacheWrite: 0.3,
861
+ },
862
+ contextWindow: 1000000,
863
+ maxTokens: 65535,
864
+ },
657
865
  "global.anthropic.claude-fable-5": {
658
866
  id: "global.anthropic.claude-fable-5",
659
867
  name: "Claude Fable 5 (Global)",
@@ -803,6 +1011,24 @@ export const MODELS = {
803
1011
  contextWindow: 1000000,
804
1012
  maxTokens: 128000,
805
1013
  },
1014
+ "global.anthropic.claude-sonnet-4-20250514-v1:0": {
1015
+ id: "global.anthropic.claude-sonnet-4-20250514-v1:0",
1016
+ name: "Claude Sonnet 4 (Global)",
1017
+ api: "bedrock-converse-stream",
1018
+ provider: "amazon-bedrock",
1019
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1020
+ reasoning: true,
1021
+ thinkingType: "budget",
1022
+ input: ["text", "image"],
1023
+ cost: {
1024
+ input: 3,
1025
+ output: 15,
1026
+ cacheRead: 0.3,
1027
+ cacheWrite: 3.75,
1028
+ },
1029
+ contextWindow: 200000,
1030
+ maxTokens: 64000,
1031
+ },
806
1032
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
807
1033
  id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
808
1034
  name: "Claude Sonnet 4.5 (Global)",
@@ -837,7 +1063,7 @@ export const MODELS = {
837
1063
  cacheWrite: 3.75,
838
1064
  },
839
1065
  contextWindow: 1000000,
840
- maxTokens: 64000,
1066
+ maxTokens: 128000,
841
1067
  },
842
1068
  "global.anthropic.claude-sonnet-5": {
843
1069
  id: "global.anthropic.claude-sonnet-5",
@@ -908,30 +1134,81 @@ export const MODELS = {
908
1134
  contextWindow: 1050000,
909
1135
  maxTokens: 128000,
910
1136
  },
911
- "google.gemma-3-27b-it": {
912
- id: "google.gemma-3-27b-it",
913
- name: "Google Gemma 3 27B Instruct",
1137
+ "global.openai.gpt-6-astra": {
1138
+ id: "global.openai.gpt-6-astra",
1139
+ name: "GPT-6 Astra (Global)",
914
1140
  api: "bedrock-converse-stream",
915
1141
  provider: "amazon-bedrock",
916
1142
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
917
- reasoning: false,
1143
+ reasoning: true,
918
1144
  input: ["text", "image"],
919
1145
  cost: {
920
- input: 0.12,
921
- output: 0.2,
922
- cacheRead: 0,
923
- cacheWrite: 0,
1146
+ input: 10,
1147
+ output: 50,
1148
+ cacheRead: 1,
1149
+ cacheWrite: 12.5,
924
1150
  },
925
- contextWindow: 202752,
926
- maxTokens: 8192,
1151
+ contextWindow: 1050000,
1152
+ maxTokens: 128000,
927
1153
  },
928
- "google.gemma-3-4b-it": {
929
- id: "google.gemma-3-4b-it",
930
- name: "Gemma 3 4B IT",
1154
+ "global.xai.grok-4.6": {
1155
+ id: "global.xai.grok-4.6",
1156
+ name: "Grok 4.6 (Global)",
931
1157
  api: "bedrock-converse-stream",
932
1158
  provider: "amazon-bedrock",
933
1159
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
934
- reasoning: false,
1160
+ reasoning: true,
1161
+ input: ["text", "image"],
1162
+ cost: {
1163
+ input: 2,
1164
+ output: 6,
1165
+ cacheRead: 0.5,
1166
+ cacheWrite: 0,
1167
+ },
1168
+ contextWindow: 500000,
1169
+ maxTokens: 500000,
1170
+ },
1171
+ "google.gemma-4-26b-a4b": {
1172
+ id: "google.gemma-4-26b-a4b",
1173
+ name: "Gemma 4 26B A4B IT",
1174
+ api: "bedrock-converse-stream",
1175
+ provider: "amazon-bedrock",
1176
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1177
+ reasoning: true,
1178
+ input: ["text", "image"],
1179
+ cost: {
1180
+ input: 0.13,
1181
+ output: 0.4,
1182
+ cacheRead: 0,
1183
+ cacheWrite: 0,
1184
+ },
1185
+ contextWindow: 262144,
1186
+ maxTokens: 32768,
1187
+ },
1188
+ "google.gemma-4-31b": {
1189
+ id: "google.gemma-4-31b",
1190
+ name: "Gemma 4 31B IT",
1191
+ api: "bedrock-converse-stream",
1192
+ provider: "amazon-bedrock",
1193
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1194
+ reasoning: true,
1195
+ input: ["text", "image"],
1196
+ cost: {
1197
+ input: 0.14,
1198
+ output: 0.4,
1199
+ cacheRead: 0,
1200
+ cacheWrite: 0,
1201
+ },
1202
+ contextWindow: 262144,
1203
+ maxTokens: 32768,
1204
+ },
1205
+ "google.gemma-4-e2b": {
1206
+ id: "google.gemma-4-e2b",
1207
+ name: "Gemma 4 E2B IT",
1208
+ api: "bedrock-converse-stream",
1209
+ provider: "amazon-bedrock",
1210
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1211
+ reasoning: true,
935
1212
  input: ["text", "image"],
936
1213
  cost: {
937
1214
  input: 0.04,
@@ -939,8 +1216,59 @@ export const MODELS = {
939
1216
  cacheRead: 0,
940
1217
  cacheWrite: 0,
941
1218
  },
942
- contextWindow: 128000,
943
- maxTokens: 4096,
1219
+ contextWindow: 131072,
1220
+ maxTokens: 8192,
1221
+ },
1222
+ "in.openai.gpt-5.6-luna": {
1223
+ id: "in.openai.gpt-5.6-luna",
1224
+ name: "GPT-5.6 Luna (India)",
1225
+ api: "bedrock-converse-stream",
1226
+ provider: "amazon-bedrock",
1227
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1228
+ reasoning: true,
1229
+ input: ["text", "image"],
1230
+ cost: {
1231
+ input: 0.22,
1232
+ output: 1.32,
1233
+ cacheRead: 0.022,
1234
+ cacheWrite: 0.275,
1235
+ },
1236
+ contextWindow: 1050000,
1237
+ maxTokens: 128000,
1238
+ },
1239
+ "in.openai.gpt-5.6-terra": {
1240
+ id: "in.openai.gpt-5.6-terra",
1241
+ name: "GPT-5.6 Terra (India)",
1242
+ api: "bedrock-converse-stream",
1243
+ provider: "amazon-bedrock",
1244
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1245
+ reasoning: true,
1246
+ input: ["text", "image"],
1247
+ cost: {
1248
+ input: 2.2,
1249
+ output: 13.2,
1250
+ cacheRead: 0.22,
1251
+ cacheWrite: 2.75,
1252
+ },
1253
+ contextWindow: 1050000,
1254
+ maxTokens: 128000,
1255
+ },
1256
+ "jp.amazon.nova-2-lite-v1:0": {
1257
+ id: "jp.amazon.nova-2-lite-v1:0",
1258
+ name: "Nova 2 Lite (JP)",
1259
+ api: "bedrock-converse-stream",
1260
+ provider: "amazon-bedrock",
1261
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1262
+ reasoning: true,
1263
+ input: ["text", "image"],
1264
+ cost: {
1265
+ input: 0.396,
1266
+ output: 3.311,
1267
+ cacheRead: 0.099,
1268
+ cacheWrite: 0.396,
1269
+ },
1270
+ contextWindow: 1000000,
1271
+ maxTokens: 65535,
944
1272
  },
945
1273
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
946
1274
  id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
@@ -952,10 +1280,10 @@ export const MODELS = {
952
1280
  thinkingType: "budget",
953
1281
  input: ["text", "image"],
954
1282
  cost: {
955
- input: 1,
956
- output: 5,
957
- cacheRead: 0.1,
958
- cacheWrite: 1.25,
1283
+ input: 1.1,
1284
+ output: 5.5,
1285
+ cacheRead: 0.11,
1286
+ cacheWrite: 1.375,
959
1287
  },
960
1288
  contextWindow: 200000,
961
1289
  maxTokens: 64000,
@@ -971,10 +1299,10 @@ export const MODELS = {
971
1299
  thinkingLevelMap: { "xhigh": "xhigh" },
972
1300
  input: ["text", "image"],
973
1301
  cost: {
974
- input: 5,
975
- output: 25,
976
- cacheRead: 0.5,
977
- cacheWrite: 6.25,
1302
+ input: 5.5,
1303
+ output: 27.5,
1304
+ cacheRead: 0.55,
1305
+ cacheWrite: 6.875,
978
1306
  },
979
1307
  contextWindow: 1000000,
980
1308
  maxTokens: 128000,
@@ -990,10 +1318,10 @@ export const MODELS = {
990
1318
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
991
1319
  input: ["text", "image"],
992
1320
  cost: {
993
- input: 5,
994
- output: 25,
995
- cacheRead: 0.5,
996
- cacheWrite: 6.25,
1321
+ input: 5.5,
1322
+ output: 27.5,
1323
+ cacheRead: 0.55,
1324
+ cacheWrite: 6.875,
997
1325
  },
998
1326
  contextWindow: 1000000,
999
1327
  maxTokens: 128000,
@@ -1008,10 +1336,10 @@ export const MODELS = {
1008
1336
  thinkingType: "adaptive",
1009
1337
  input: ["text", "image"],
1010
1338
  cost: {
1011
- input: 5,
1012
- output: 25,
1013
- cacheRead: 0.5,
1014
- cacheWrite: 6.25,
1339
+ input: 5.5,
1340
+ output: 27.5,
1341
+ cacheRead: 0.55,
1342
+ cacheWrite: 6.875,
1015
1343
  },
1016
1344
  contextWindow: 1000000,
1017
1345
  maxTokens: 128000,
@@ -1026,10 +1354,10 @@ export const MODELS = {
1026
1354
  thinkingType: "budget",
1027
1355
  input: ["text", "image"],
1028
1356
  cost: {
1029
- input: 3,
1030
- output: 15,
1031
- cacheRead: 0.3,
1032
- cacheWrite: 3.75,
1357
+ input: 3.3,
1358
+ output: 16.5,
1359
+ cacheRead: 0.33,
1360
+ cacheWrite: 4.125,
1033
1361
  },
1034
1362
  contextWindow: 200000,
1035
1363
  maxTokens: 64000,
@@ -1044,13 +1372,13 @@ export const MODELS = {
1044
1372
  thinkingType: "adaptive",
1045
1373
  input: ["text", "image"],
1046
1374
  cost: {
1047
- input: 3,
1048
- output: 15,
1049
- cacheRead: 0.3,
1050
- cacheWrite: 3.75,
1375
+ input: 3.3,
1376
+ output: 16.5,
1377
+ cacheRead: 0.33,
1378
+ cacheWrite: 4.125,
1051
1379
  },
1052
1380
  contextWindow: 1000000,
1053
- maxTokens: 64000,
1381
+ maxTokens: 128000,
1054
1382
  },
1055
1383
  "jp.anthropic.claude-sonnet-5": {
1056
1384
  id: "jp.anthropic.claude-sonnet-5",
@@ -1062,10 +1390,10 @@ export const MODELS = {
1062
1390
  thinkingType: "adaptive",
1063
1391
  input: ["text", "image"],
1064
1392
  cost: {
1065
- input: 2,
1066
- output: 10,
1067
- cacheRead: 0.2,
1068
- cacheWrite: 2.5,
1393
+ input: 2.2,
1394
+ output: 11,
1395
+ cacheRead: 0.22,
1396
+ cacheWrite: 2.75,
1069
1397
  },
1070
1398
  contextWindow: 1000000,
1071
1399
  maxTokens: 128000,
@@ -1136,7 +1464,7 @@ export const MODELS = {
1136
1464
  cacheWrite: 0,
1137
1465
  },
1138
1466
  contextWindow: 1000000,
1139
- maxTokens: 16384,
1467
+ maxTokens: 8192,
1140
1468
  },
1141
1469
  "meta.llama4-scout-17b-instruct-v1:0": {
1142
1470
  id: "meta.llama4-scout-17b-instruct-v1:0",
@@ -1152,12 +1480,12 @@ export const MODELS = {
1152
1480
  cacheRead: 0,
1153
1481
  cacheWrite: 0,
1154
1482
  },
1155
- contextWindow: 3500000,
1156
- maxTokens: 16384,
1483
+ contextWindow: 10000000,
1484
+ maxTokens: 8192,
1157
1485
  },
1158
1486
  "minimax.minimax-m2": {
1159
1487
  id: "minimax.minimax-m2",
1160
- name: "MiniMax M2",
1488
+ name: "MiniMax-M2",
1161
1489
  api: "bedrock-converse-stream",
1162
1490
  provider: "amazon-bedrock",
1163
1491
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1174,7 +1502,7 @@ export const MODELS = {
1174
1502
  },
1175
1503
  "minimax.minimax-m2.1": {
1176
1504
  id: "minimax.minimax-m2.1",
1177
- name: "MiniMax M2.1",
1505
+ name: "MiniMax-M2.1",
1178
1506
  api: "bedrock-converse-stream",
1179
1507
  provider: "amazon-bedrock",
1180
1508
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1191,7 +1519,7 @@ export const MODELS = {
1191
1519
  },
1192
1520
  "minimax.minimax-m2.5": {
1193
1521
  id: "minimax.minimax-m2.5",
1194
- name: "MiniMax M2.5",
1522
+ name: "MiniMax-M2.5",
1195
1523
  api: "bedrock-converse-stream",
1196
1524
  provider: "amazon-bedrock",
1197
1525
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1247,7 +1575,7 @@ export const MODELS = {
1247
1575
  provider: "amazon-bedrock",
1248
1576
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1249
1577
  reasoning: false,
1250
- input: ["text"],
1578
+ input: ["text", "image"],
1251
1579
  cost: {
1252
1580
  input: 0.2,
1253
1581
  output: 0.2,
@@ -1281,7 +1609,7 @@ export const MODELS = {
1281
1609
  provider: "amazon-bedrock",
1282
1610
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1283
1611
  reasoning: false,
1284
- input: ["text"],
1612
+ input: ["text", "image"],
1285
1613
  cost: {
1286
1614
  input: 0.15,
1287
1615
  output: 0.15,
@@ -1339,7 +1667,7 @@ export const MODELS = {
1339
1667
  cacheRead: 0,
1340
1668
  cacheWrite: 0,
1341
1669
  },
1342
- contextWindow: 128000,
1670
+ contextWindow: 32768,
1343
1671
  maxTokens: 4096,
1344
1672
  },
1345
1673
  "mistral.voxtral-small-24b-2507": {
@@ -1351,12 +1679,12 @@ export const MODELS = {
1351
1679
  reasoning: false,
1352
1680
  input: ["text"],
1353
1681
  cost: {
1354
- input: 0.15,
1355
- output: 0.35,
1682
+ input: 0.1,
1683
+ output: 0.3,
1356
1684
  cacheRead: 0,
1357
1685
  cacheWrite: 0,
1358
1686
  },
1359
- contextWindow: 32000,
1687
+ contextWindow: 32768,
1360
1688
  maxTokens: 8192,
1361
1689
  },
1362
1690
  "moonshot.kimi-k2-thinking": {
@@ -1391,7 +1719,7 @@ export const MODELS = {
1391
1719
  cacheWrite: 0,
1392
1720
  },
1393
1721
  contextWindow: 262143,
1394
- maxTokens: 16000,
1722
+ maxTokens: 16384,
1395
1723
  },
1396
1724
  "nvidia.nemotron-nano-12b-v2": {
1397
1725
  id: "nvidia.nemotron-nano-12b-v2",
@@ -1408,7 +1736,7 @@ export const MODELS = {
1408
1736
  cacheWrite: 0,
1409
1737
  },
1410
1738
  contextWindow: 128000,
1411
- maxTokens: 4096,
1739
+ maxTokens: 8192,
1412
1740
  },
1413
1741
  "nvidia.nemotron-nano-3-30b": {
1414
1742
  id: "nvidia.nemotron-nano-3-30b",
@@ -1424,8 +1752,8 @@ export const MODELS = {
1424
1752
  cacheRead: 0,
1425
1753
  cacheWrite: 0,
1426
1754
  },
1427
- contextWindow: 128000,
1428
- maxTokens: 4096,
1755
+ contextWindow: 262144,
1756
+ maxTokens: 8192,
1429
1757
  },
1430
1758
  "nvidia.nemotron-nano-9b-v2": {
1431
1759
  id: "nvidia.nemotron-nano-9b-v2",
@@ -1441,8 +1769,8 @@ export const MODELS = {
1441
1769
  cacheRead: 0,
1442
1770
  cacheWrite: 0,
1443
1771
  },
1444
- contextWindow: 128000,
1445
- maxTokens: 4096,
1772
+ contextWindow: 131072,
1773
+ maxTokens: 8192,
1446
1774
  },
1447
1775
  "nvidia.nemotron-super-3-120b": {
1448
1776
  id: "nvidia.nemotron-super-3-120b",
@@ -1548,6 +1876,23 @@ export const MODELS = {
1548
1876
  contextWindow: 1050000,
1549
1877
  maxTokens: 128000,
1550
1878
  },
1879
+ "openai.gpt-6-astra": {
1880
+ id: "openai.gpt-6-astra",
1881
+ name: "GPT-6 Astra",
1882
+ api: "bedrock-converse-stream",
1883
+ provider: "amazon-bedrock",
1884
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1885
+ reasoning: true,
1886
+ input: ["text", "image"],
1887
+ cost: {
1888
+ input: 11,
1889
+ output: 55,
1890
+ cacheRead: 1.1,
1891
+ cacheWrite: 13.75,
1892
+ },
1893
+ contextWindow: 1050000,
1894
+ maxTokens: 128000,
1895
+ },
1551
1896
  "openai.gpt-oss-120b": {
1552
1897
  id: "openai.gpt-oss-120b",
1553
1898
  name: "gpt-oss-120b",
@@ -1622,7 +1967,7 @@ export const MODELS = {
1622
1967
  api: "bedrock-converse-stream",
1623
1968
  provider: "amazon-bedrock",
1624
1969
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1625
- reasoning: false,
1970
+ reasoning: true,
1626
1971
  input: ["text"],
1627
1972
  cost: {
1628
1973
  input: 0.15,
@@ -1639,7 +1984,7 @@ export const MODELS = {
1639
1984
  api: "bedrock-converse-stream",
1640
1985
  provider: "amazon-bedrock",
1641
1986
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1642
- reasoning: false,
1987
+ reasoning: true,
1643
1988
  input: ["text"],
1644
1989
  cost: {
1645
1990
  input: 0.07,
@@ -1652,7 +1997,7 @@ export const MODELS = {
1652
1997
  },
1653
1998
  "qwen.qwen3-235b-a22b-2507-v1:0": {
1654
1999
  id: "qwen.qwen3-235b-a22b-2507-v1:0",
1655
- name: "Qwen3 235B A22B 2507",
2000
+ name: "Qwen3 235B-A22B Instruct 2507",
1656
2001
  api: "bedrock-converse-stream",
1657
2002
  provider: "amazon-bedrock",
1658
2003
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1669,7 +2014,7 @@ export const MODELS = {
1669
2014
  },
1670
2015
  "qwen.qwen3-32b-v1:0": {
1671
2016
  id: "qwen.qwen3-32b-v1:0",
1672
- name: "Qwen3 32B (dense)",
2017
+ name: "Qwen3 32B",
1673
2018
  api: "bedrock-converse-stream",
1674
2019
  provider: "amazon-bedrock",
1675
2020
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1681,12 +2026,12 @@ export const MODELS = {
1681
2026
  cacheRead: 0,
1682
2027
  cacheWrite: 0,
1683
2028
  },
1684
- contextWindow: 16384,
2029
+ contextWindow: 32768,
1685
2030
  maxTokens: 16384,
1686
2031
  },
1687
2032
  "qwen.qwen3-coder-30b-a3b-v1:0": {
1688
2033
  id: "qwen.qwen3-coder-30b-a3b-v1:0",
1689
- name: "Qwen3 Coder 30B A3B Instruct",
2034
+ name: "Qwen3-Coder 30B-A3B Instruct",
1690
2035
  api: "bedrock-converse-stream",
1691
2036
  provider: "amazon-bedrock",
1692
2037
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1703,14 +2048,14 @@ export const MODELS = {
1703
2048
  },
1704
2049
  "qwen.qwen3-coder-480b-a35b-v1:0": {
1705
2050
  id: "qwen.qwen3-coder-480b-a35b-v1:0",
1706
- name: "Qwen3 Coder 480B A35B Instruct",
2051
+ name: "Qwen3-Coder 480B-A35B Instruct",
1707
2052
  api: "bedrock-converse-stream",
1708
2053
  provider: "amazon-bedrock",
1709
2054
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1710
2055
  reasoning: false,
1711
2056
  input: ["text"],
1712
2057
  cost: {
1713
- input: 0.22,
2058
+ input: 0.45,
1714
2059
  output: 1.8,
1715
2060
  cacheRead: 0,
1716
2061
  cacheWrite: 0,
@@ -1724,82 +2069,201 @@ export const MODELS = {
1724
2069
  api: "bedrock-converse-stream",
1725
2070
  provider: "amazon-bedrock",
1726
2071
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1727
- reasoning: true,
2072
+ reasoning: false,
1728
2073
  input: ["text"],
1729
2074
  cost: {
1730
- input: 0.22,
1731
- output: 1.8,
2075
+ input: 0.5,
2076
+ output: 1.2,
1732
2077
  cacheRead: 0,
1733
2078
  cacheWrite: 0,
1734
2079
  },
1735
- contextWindow: 131072,
2080
+ contextWindow: 262144,
1736
2081
  maxTokens: 65536,
1737
2082
  },
1738
2083
  "qwen.qwen3-next-80b-a3b": {
1739
2084
  id: "qwen.qwen3-next-80b-a3b",
1740
- name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
2085
+ name: "Qwen3-Next 80B-A3B Instruct",
1741
2086
  api: "bedrock-converse-stream",
1742
2087
  provider: "amazon-bedrock",
1743
2088
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1744
2089
  reasoning: false,
1745
2090
  input: ["text"],
1746
2091
  cost: {
1747
- input: 0.14,
1748
- output: 1.4,
2092
+ input: 0.15,
2093
+ output: 1.2,
1749
2094
  cacheRead: 0,
1750
2095
  cacheWrite: 0,
1751
2096
  },
1752
- contextWindow: 262000,
2097
+ contextWindow: 262144,
1753
2098
  maxTokens: 262000,
1754
2099
  },
1755
2100
  "qwen.qwen3-vl-235b-a22b": {
1756
2101
  id: "qwen.qwen3-vl-235b-a22b",
1757
- name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
2102
+ name: "Qwen3 VL 235B A22B Instruct",
1758
2103
  api: "bedrock-converse-stream",
1759
2104
  provider: "amazon-bedrock",
1760
2105
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1761
2106
  reasoning: false,
1762
2107
  input: ["text", "image"],
1763
2108
  cost: {
1764
- input: 0.3,
1765
- output: 1.5,
2109
+ input: 0.53,
2110
+ output: 2.66,
1766
2111
  cacheRead: 0,
1767
2112
  cacheWrite: 0,
1768
2113
  },
1769
- contextWindow: 262000,
2114
+ contextWindow: 262144,
1770
2115
  maxTokens: 262000,
1771
2116
  },
1772
- "us.anthropic.claude-fable-5": {
1773
- id: "us.anthropic.claude-fable-5",
1774
- name: "Claude Fable 5 (US)",
2117
+ "us-gov.openai.gpt-oss-120b-1:0": {
2118
+ id: "us-gov.openai.gpt-oss-120b-1:0",
2119
+ name: "gpt-oss-120b (GovCloud)",
1775
2120
  api: "bedrock-converse-stream",
1776
2121
  provider: "amazon-bedrock",
1777
2122
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1778
2123
  reasoning: true,
1779
- thinkingType: "adaptive",
1780
- thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
1781
- input: ["text", "image"],
2124
+ input: ["text"],
1782
2125
  cost: {
1783
- input: 10,
1784
- output: 50,
1785
- cacheRead: 1,
1786
- cacheWrite: 12.5,
2126
+ input: 0.18,
2127
+ output: 0.72,
2128
+ cacheRead: 0,
2129
+ cacheWrite: 0,
1787
2130
  },
1788
- contextWindow: 1000000,
1789
- maxTokens: 128000,
2131
+ contextWindow: 128000,
2132
+ maxTokens: 16384,
1790
2133
  },
1791
- "us.anthropic.claude-fable-5-1": {
1792
- id: "us.anthropic.claude-fable-5-1",
1793
- name: "Claude Fable 5.1 (US)",
2134
+ "us-gov.openai.gpt-oss-20b-1:0": {
2135
+ id: "us-gov.openai.gpt-oss-20b-1:0",
2136
+ name: "gpt-oss-20b (GovCloud)",
1794
2137
  api: "bedrock-converse-stream",
1795
2138
  provider: "amazon-bedrock",
1796
2139
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1797
2140
  reasoning: true,
1798
- thinkingType: "adaptive",
1799
- thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
1800
- input: ["text", "image"],
2141
+ input: ["text"],
1801
2142
  cost: {
1802
- input: 11,
2143
+ input: 0.084,
2144
+ output: 0.36,
2145
+ cacheRead: 0,
2146
+ cacheWrite: 0,
2147
+ },
2148
+ contextWindow: 128000,
2149
+ maxTokens: 16384,
2150
+ },
2151
+ "us.amazon.nova-2-lite-v1:0": {
2152
+ id: "us.amazon.nova-2-lite-v1:0",
2153
+ name: "Nova 2 Lite (US)",
2154
+ api: "bedrock-converse-stream",
2155
+ provider: "amazon-bedrock",
2156
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2157
+ reasoning: true,
2158
+ input: ["text", "image"],
2159
+ cost: {
2160
+ input: 0.33,
2161
+ output: 2.75,
2162
+ cacheRead: 0.0825,
2163
+ cacheWrite: 0.33,
2164
+ },
2165
+ contextWindow: 1000000,
2166
+ maxTokens: 65535,
2167
+ },
2168
+ "us.amazon.nova-lite-v1:0": {
2169
+ id: "us.amazon.nova-lite-v1:0",
2170
+ name: "Nova Lite (US)",
2171
+ api: "bedrock-converse-stream",
2172
+ provider: "amazon-bedrock",
2173
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2174
+ reasoning: false,
2175
+ input: ["text", "image"],
2176
+ cost: {
2177
+ input: 0.06,
2178
+ output: 0.24,
2179
+ cacheRead: 0.015,
2180
+ cacheWrite: 0.06,
2181
+ },
2182
+ contextWindow: 300000,
2183
+ maxTokens: 10000,
2184
+ },
2185
+ "us.amazon.nova-micro-v1:0": {
2186
+ id: "us.amazon.nova-micro-v1:0",
2187
+ name: "Nova Micro (US)",
2188
+ api: "bedrock-converse-stream",
2189
+ provider: "amazon-bedrock",
2190
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2191
+ reasoning: false,
2192
+ input: ["text"],
2193
+ cost: {
2194
+ input: 0.035,
2195
+ output: 0.14,
2196
+ cacheRead: 0.00875,
2197
+ cacheWrite: 0.035,
2198
+ },
2199
+ contextWindow: 128000,
2200
+ maxTokens: 10000,
2201
+ },
2202
+ "us.amazon.nova-premier-v1:0": {
2203
+ id: "us.amazon.nova-premier-v1:0",
2204
+ name: "Nova Premier (US)",
2205
+ api: "bedrock-converse-stream",
2206
+ provider: "amazon-bedrock",
2207
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2208
+ reasoning: false,
2209
+ input: ["text", "image"],
2210
+ cost: {
2211
+ input: 2.5,
2212
+ output: 12.5,
2213
+ cacheRead: 0.625,
2214
+ cacheWrite: 2.5,
2215
+ },
2216
+ contextWindow: 1000000,
2217
+ maxTokens: 10000,
2218
+ },
2219
+ "us.amazon.nova-pro-v1:0": {
2220
+ id: "us.amazon.nova-pro-v1:0",
2221
+ name: "Nova Pro (US)",
2222
+ api: "bedrock-converse-stream",
2223
+ provider: "amazon-bedrock",
2224
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2225
+ reasoning: false,
2226
+ input: ["text", "image"],
2227
+ cost: {
2228
+ input: 0.8,
2229
+ output: 3.2,
2230
+ cacheRead: 0.2,
2231
+ cacheWrite: 0.8,
2232
+ },
2233
+ contextWindow: 300000,
2234
+ maxTokens: 10000,
2235
+ },
2236
+ "us.anthropic.claude-fable-5": {
2237
+ id: "us.anthropic.claude-fable-5",
2238
+ name: "Claude Fable 5 (US)",
2239
+ api: "bedrock-converse-stream",
2240
+ provider: "amazon-bedrock",
2241
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2242
+ reasoning: true,
2243
+ thinkingType: "adaptive",
2244
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
2245
+ input: ["text", "image"],
2246
+ cost: {
2247
+ input: 11,
2248
+ output: 55,
2249
+ cacheRead: 1.1,
2250
+ cacheWrite: 13.75,
2251
+ },
2252
+ contextWindow: 1000000,
2253
+ maxTokens: 128000,
2254
+ },
2255
+ "us.anthropic.claude-fable-5-1": {
2256
+ id: "us.anthropic.claude-fable-5-1",
2257
+ name: "Claude Fable 5.1 (US)",
2258
+ api: "bedrock-converse-stream",
2259
+ provider: "amazon-bedrock",
2260
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2261
+ reasoning: true,
2262
+ thinkingType: "adaptive",
2263
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
2264
+ input: ["text", "image"],
2265
+ cost: {
2266
+ input: 11,
1803
2267
  output: 55,
1804
2268
  cacheRead: 0.275,
1805
2269
  cacheWrite: 13.75,
@@ -1817,10 +2281,10 @@ export const MODELS = {
1817
2281
  thinkingType: "budget",
1818
2282
  input: ["text", "image"],
1819
2283
  cost: {
1820
- input: 1,
1821
- output: 5,
1822
- cacheRead: 0.1,
1823
- cacheWrite: 1.25,
2284
+ input: 1.1,
2285
+ output: 5.5,
2286
+ cacheRead: 0.11,
2287
+ cacheWrite: 1.375,
1824
2288
  },
1825
2289
  contextWindow: 200000,
1826
2290
  maxTokens: 64000,
@@ -1853,10 +2317,10 @@ export const MODELS = {
1853
2317
  thinkingType: "budget",
1854
2318
  input: ["text", "image"],
1855
2319
  cost: {
1856
- input: 5,
1857
- output: 25,
1858
- cacheRead: 0.5,
1859
- cacheWrite: 6.25,
2320
+ input: 5.5,
2321
+ output: 27.5,
2322
+ cacheRead: 0.55,
2323
+ cacheWrite: 6.875,
1860
2324
  },
1861
2325
  contextWindow: 200000,
1862
2326
  maxTokens: 64000,
@@ -1872,8 +2336,8 @@ export const MODELS = {
1872
2336
  thinkingLevelMap: { "xhigh": "max" },
1873
2337
  input: ["text", "image"],
1874
2338
  cost: {
1875
- input: 5,
1876
- output: 25,
2339
+ input: 5.5,
2340
+ output: 27.5,
1877
2341
  cacheRead: 0.5,
1878
2342
  cacheWrite: 6.25,
1879
2343
  },
@@ -1891,10 +2355,10 @@ export const MODELS = {
1891
2355
  thinkingLevelMap: { "xhigh": "xhigh" },
1892
2356
  input: ["text", "image"],
1893
2357
  cost: {
1894
- input: 5,
1895
- output: 25,
1896
- cacheRead: 0.5,
1897
- cacheWrite: 6.25,
2358
+ input: 5.5,
2359
+ output: 27.5,
2360
+ cacheRead: 0.55,
2361
+ cacheWrite: 6.875,
1898
2362
  },
1899
2363
  contextWindow: 1000000,
1900
2364
  maxTokens: 128000,
@@ -1910,10 +2374,10 @@ export const MODELS = {
1910
2374
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
1911
2375
  input: ["text", "image"],
1912
2376
  cost: {
1913
- input: 5,
1914
- output: 25,
1915
- cacheRead: 0.5,
1916
- cacheWrite: 6.25,
2377
+ input: 5.5,
2378
+ output: 27.5,
2379
+ cacheRead: 0.55,
2380
+ cacheWrite: 6.875,
1917
2381
  },
1918
2382
  contextWindow: 1000000,
1919
2383
  maxTokens: 128000,
@@ -1928,17 +2392,17 @@ export const MODELS = {
1928
2392
  thinkingType: "adaptive",
1929
2393
  input: ["text", "image"],
1930
2394
  cost: {
1931
- input: 5,
1932
- output: 25,
1933
- cacheRead: 0.5,
1934
- cacheWrite: 6.25,
2395
+ input: 5.5,
2396
+ output: 27.5,
2397
+ cacheRead: 0.55,
2398
+ cacheWrite: 6.875,
1935
2399
  },
1936
2400
  contextWindow: 1000000,
1937
2401
  maxTokens: 128000,
1938
2402
  },
1939
- "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
1940
- id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
1941
- name: "Claude Sonnet 4.5 (US)",
2403
+ "us.anthropic.claude-sonnet-4-20250514-v1:0": {
2404
+ id: "us.anthropic.claude-sonnet-4-20250514-v1:0",
2405
+ name: "Claude Sonnet 4 (US)",
1942
2406
  api: "bedrock-converse-stream",
1943
2407
  provider: "amazon-bedrock",
1944
2408
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1954,6 +2418,24 @@ export const MODELS = {
1954
2418
  contextWindow: 200000,
1955
2419
  maxTokens: 64000,
1956
2420
  },
2421
+ "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2422
+ id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
2423
+ name: "Claude Sonnet 4.5 (US)",
2424
+ api: "bedrock-converse-stream",
2425
+ provider: "amazon-bedrock",
2426
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2427
+ reasoning: true,
2428
+ thinkingType: "budget",
2429
+ input: ["text", "image"],
2430
+ cost: {
2431
+ input: 3.3,
2432
+ output: 16.5,
2433
+ cacheRead: 0.33,
2434
+ cacheWrite: 4.125,
2435
+ },
2436
+ contextWindow: 200000,
2437
+ maxTokens: 64000,
2438
+ },
1957
2439
  "us.anthropic.claude-sonnet-4-6": {
1958
2440
  id: "us.anthropic.claude-sonnet-4-6",
1959
2441
  name: "Claude Sonnet 4.6 (US)",
@@ -1964,13 +2446,13 @@ export const MODELS = {
1964
2446
  thinkingType: "adaptive",
1965
2447
  input: ["text", "image"],
1966
2448
  cost: {
1967
- input: 3,
1968
- output: 15,
1969
- cacheRead: 0.3,
1970
- cacheWrite: 3.75,
2449
+ input: 3.3,
2450
+ output: 16.5,
2451
+ cacheRead: 0.33,
2452
+ cacheWrite: 4.125,
1971
2453
  },
1972
2454
  contextWindow: 1000000,
1973
- maxTokens: 64000,
2455
+ maxTokens: 128000,
1974
2456
  },
1975
2457
  "us.anthropic.claude-sonnet-5": {
1976
2458
  id: "us.anthropic.claude-sonnet-5",
@@ -1982,30 +2464,64 @@ export const MODELS = {
1982
2464
  thinkingType: "adaptive",
1983
2465
  input: ["text", "image"],
1984
2466
  cost: {
1985
- input: 2,
1986
- output: 10,
1987
- cacheRead: 0.2,
1988
- cacheWrite: 2.5,
2467
+ input: 2.2,
2468
+ output: 11,
2469
+ cacheRead: 0.22,
2470
+ cacheWrite: 2.75,
1989
2471
  },
1990
2472
  contextWindow: 1000000,
1991
2473
  maxTokens: 128000,
1992
2474
  },
1993
- "us.deepseek.r1-v1:0": {
1994
- id: "us.deepseek.r1-v1:0",
1995
- name: "DeepSeek-R1 (US)",
2475
+ "us.meta.llama3-1-70b-instruct-v1:0": {
2476
+ id: "us.meta.llama3-1-70b-instruct-v1:0",
2477
+ name: "Llama 3.1 70B Instruct (US)",
1996
2478
  api: "bedrock-converse-stream",
1997
2479
  provider: "amazon-bedrock",
1998
2480
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1999
- reasoning: true,
2481
+ reasoning: false,
2000
2482
  input: ["text"],
2001
2483
  cost: {
2002
- input: 1.35,
2003
- output: 5.4,
2484
+ input: 0.72,
2485
+ output: 0.72,
2004
2486
  cacheRead: 0,
2005
2487
  cacheWrite: 0,
2006
2488
  },
2007
2489
  contextWindow: 128000,
2008
- maxTokens: 32768,
2490
+ maxTokens: 4096,
2491
+ },
2492
+ "us.meta.llama3-1-8b-instruct-v1:0": {
2493
+ id: "us.meta.llama3-1-8b-instruct-v1:0",
2494
+ name: "Llama 3.1 8B Instruct (US)",
2495
+ api: "bedrock-converse-stream",
2496
+ provider: "amazon-bedrock",
2497
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2498
+ reasoning: false,
2499
+ input: ["text"],
2500
+ cost: {
2501
+ input: 0.22,
2502
+ output: 0.22,
2503
+ cacheRead: 0,
2504
+ cacheWrite: 0,
2505
+ },
2506
+ contextWindow: 128000,
2507
+ maxTokens: 4096,
2508
+ },
2509
+ "us.meta.llama3-3-70b-instruct-v1:0": {
2510
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
2511
+ name: "Llama 3.3 70B Instruct (US)",
2512
+ api: "bedrock-converse-stream",
2513
+ provider: "amazon-bedrock",
2514
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2515
+ reasoning: false,
2516
+ input: ["text"],
2517
+ cost: {
2518
+ input: 0.72,
2519
+ output: 0.72,
2520
+ cacheRead: 0,
2521
+ cacheWrite: 0,
2522
+ },
2523
+ contextWindow: 128000,
2524
+ maxTokens: 4096,
2009
2525
  },
2010
2526
  "us.meta.llama4-maverick-17b-instruct-v1:0": {
2011
2527
  id: "us.meta.llama4-maverick-17b-instruct-v1:0",
@@ -2022,24 +2538,160 @@ export const MODELS = {
2022
2538
  cacheWrite: 0,
2023
2539
  },
2024
2540
  contextWindow: 1000000,
2025
- maxTokens: 16384,
2541
+ maxTokens: 8192,
2542
+ },
2543
+ "us.meta.llama4-scout-17b-instruct-v1:0": {
2544
+ id: "us.meta.llama4-scout-17b-instruct-v1:0",
2545
+ name: "Llama 4 Scout 17B Instruct (US)",
2546
+ api: "bedrock-converse-stream",
2547
+ provider: "amazon-bedrock",
2548
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2549
+ reasoning: false,
2550
+ input: ["text", "image"],
2551
+ cost: {
2552
+ input: 0.17,
2553
+ output: 0.66,
2554
+ cacheRead: 0,
2555
+ cacheWrite: 0,
2556
+ },
2557
+ contextWindow: 10000000,
2558
+ maxTokens: 8192,
2559
+ },
2560
+ "us.mistral.pixtral-large-2502-v1:0": {
2561
+ id: "us.mistral.pixtral-large-2502-v1:0",
2562
+ name: "Pixtral Large (25.02) (US)",
2563
+ api: "bedrock-converse-stream",
2564
+ provider: "amazon-bedrock",
2565
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2566
+ reasoning: false,
2567
+ input: ["text", "image"],
2568
+ cost: {
2569
+ input: 2,
2570
+ output: 6,
2571
+ cacheRead: 0,
2572
+ cacheWrite: 0,
2573
+ },
2574
+ contextWindow: 128000,
2575
+ maxTokens: 8192,
2576
+ },
2577
+ "us.openai.gpt-5.6-luna": {
2578
+ id: "us.openai.gpt-5.6-luna",
2579
+ name: "GPT-5.6 Luna (US)",
2580
+ api: "bedrock-converse-stream",
2581
+ provider: "amazon-bedrock",
2582
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2583
+ reasoning: true,
2584
+ input: ["text", "image"],
2585
+ cost: {
2586
+ input: 0.22,
2587
+ output: 1.32,
2588
+ cacheRead: 0.022,
2589
+ cacheWrite: 0.275,
2590
+ },
2591
+ contextWindow: 1050000,
2592
+ maxTokens: 128000,
2593
+ },
2594
+ "us.openai.gpt-5.6-sol": {
2595
+ id: "us.openai.gpt-5.6-sol",
2596
+ name: "GPT-5.6 Sol (US)",
2597
+ api: "bedrock-converse-stream",
2598
+ provider: "amazon-bedrock",
2599
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2600
+ reasoning: true,
2601
+ input: ["text", "image"],
2602
+ cost: {
2603
+ input: 4.4,
2604
+ output: 22,
2605
+ cacheRead: 0.44,
2606
+ cacheWrite: 5.5,
2607
+ },
2608
+ contextWindow: 1050000,
2609
+ maxTokens: 128000,
2610
+ },
2611
+ "us.openai.gpt-5.6-terra": {
2612
+ id: "us.openai.gpt-5.6-terra",
2613
+ name: "GPT-5.6 Terra (US)",
2614
+ api: "bedrock-converse-stream",
2615
+ provider: "amazon-bedrock",
2616
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2617
+ reasoning: true,
2618
+ input: ["text", "image"],
2619
+ cost: {
2620
+ input: 2.2,
2621
+ output: 13.2,
2622
+ cacheRead: 0.22,
2623
+ cacheWrite: 2.75,
2624
+ },
2625
+ contextWindow: 1050000,
2626
+ maxTokens: 128000,
2627
+ },
2628
+ "us.openai.gpt-6-astra": {
2629
+ id: "us.openai.gpt-6-astra",
2630
+ name: "GPT-6 Astra (US)",
2631
+ api: "bedrock-converse-stream",
2632
+ provider: "amazon-bedrock",
2633
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2634
+ reasoning: true,
2635
+ input: ["text", "image"],
2636
+ cost: {
2637
+ input: 11,
2638
+ output: 55,
2639
+ cacheRead: 1.1,
2640
+ cacheWrite: 13.75,
2641
+ },
2642
+ contextWindow: 1050000,
2643
+ maxTokens: 128000,
2644
+ },
2645
+ "us.writer.palmyra-x4-v1:0": {
2646
+ id: "us.writer.palmyra-x4-v1:0",
2647
+ name: "Palmyra X4 (US)",
2648
+ api: "bedrock-converse-stream",
2649
+ provider: "amazon-bedrock",
2650
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2651
+ reasoning: true,
2652
+ input: ["text"],
2653
+ cost: {
2654
+ input: 2.5,
2655
+ output: 10,
2656
+ cacheRead: 0,
2657
+ cacheWrite: 0,
2658
+ },
2659
+ contextWindow: 122880,
2660
+ maxTokens: 8192,
2661
+ },
2662
+ "us.writer.palmyra-x5-v1:0": {
2663
+ id: "us.writer.palmyra-x5-v1:0",
2664
+ name: "Palmyra X5 (US)",
2665
+ api: "bedrock-converse-stream",
2666
+ provider: "amazon-bedrock",
2667
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2668
+ reasoning: true,
2669
+ input: ["text"],
2670
+ cost: {
2671
+ input: 0.6,
2672
+ output: 6,
2673
+ cacheRead: 0,
2674
+ cacheWrite: 0,
2675
+ },
2676
+ contextWindow: 1040000,
2677
+ maxTokens: 8192,
2026
2678
  },
2027
- "us.meta.llama4-scout-17b-instruct-v1:0": {
2028
- id: "us.meta.llama4-scout-17b-instruct-v1:0",
2029
- name: "Llama 4 Scout 17B Instruct (US)",
2679
+ "us.xai.grok-4.6": {
2680
+ id: "us.xai.grok-4.6",
2681
+ name: "Grok 4.6 (US)",
2030
2682
  api: "bedrock-converse-stream",
2031
2683
  provider: "amazon-bedrock",
2032
2684
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2033
- reasoning: false,
2685
+ reasoning: true,
2034
2686
  input: ["text", "image"],
2035
2687
  cost: {
2036
- input: 0.17,
2037
- output: 0.66,
2038
- cacheRead: 0,
2688
+ input: 2.2,
2689
+ output: 6.6,
2690
+ cacheRead: 0.55,
2039
2691
  cacheWrite: 0,
2040
2692
  },
2041
- contextWindow: 3500000,
2042
- maxTokens: 16384,
2693
+ contextWindow: 500000,
2694
+ maxTokens: 500000,
2043
2695
  },
2044
2696
  "writer.palmyra-x4-v1:0": {
2045
2697
  id: "writer.palmyra-x4-v1:0",
@@ -2158,7 +2810,7 @@ export const MODELS = {
2158
2810
  cacheWrite: 0,
2159
2811
  },
2160
2812
  contextWindow: 202752,
2161
- maxTokens: 101376,
2813
+ maxTokens: 131072,
2162
2814
  },
2163
2815
  },
2164
2816
  "anthropic": {
@@ -3093,39 +3745,39 @@ export const MODELS = {
3093
3745
  },
3094
3746
  },
3095
3747
  "cerebras": {
3096
- "gemma-4-31b": {
3097
- id: "gemma-4-31b",
3098
- name: "Gemma 4 31B IT",
3748
+ "gpt-oss-120b": {
3749
+ id: "gpt-oss-120b",
3750
+ name: "GPT OSS 120B",
3099
3751
  api: "openai-completions",
3100
3752
  provider: "cerebras",
3101
3753
  baseUrl: "https://api.cerebras.ai/v1",
3102
3754
  reasoning: true,
3103
- input: ["text", "image"],
3755
+ input: ["text"],
3104
3756
  cost: {
3105
- input: 0.99,
3106
- output: 1.49,
3757
+ input: 0.35,
3758
+ output: 0.75,
3107
3759
  cacheRead: 0,
3108
3760
  cacheWrite: 0,
3109
3761
  },
3110
3762
  contextWindow: 131072,
3111
3763
  maxTokens: 40960,
3112
3764
  },
3113
- "gpt-oss-120b": {
3114
- id: "gpt-oss-120b",
3115
- name: "GPT OSS 120B",
3765
+ "qwen-3.8-27b": {
3766
+ id: "qwen-3.8-27b",
3767
+ name: "Qwen3.8 27B",
3116
3768
  api: "openai-completions",
3117
3769
  provider: "cerebras",
3118
3770
  baseUrl: "https://api.cerebras.ai/v1",
3119
3771
  reasoning: true,
3120
- input: ["text"],
3772
+ input: ["text", "image"],
3121
3773
  cost: {
3122
- input: 0.35,
3123
- output: 0.75,
3774
+ input: 0.99,
3775
+ output: 1.49,
3124
3776
  cacheRead: 0,
3125
3777
  cacheWrite: 0,
3126
3778
  },
3127
- contextWindow: 131072,
3128
- maxTokens: 40960,
3779
+ contextWindow: 65536,
3780
+ maxTokens: 32768,
3129
3781
  },
3130
3782
  },
3131
3783
  "cloudflare-ai-gateway": {
@@ -4095,6 +4747,23 @@ export const MODELS = {
4095
4747
  contextWindow: 1000000,
4096
4748
  maxTokens: 384000,
4097
4749
  },
4750
+ "accounts/fireworks/models/deepseek-v4-pro": {
4751
+ id: "accounts/fireworks/models/deepseek-v4-pro",
4752
+ name: "DeepSeek V4 Pro",
4753
+ api: "anthropic-messages",
4754
+ provider: "fireworks",
4755
+ baseUrl: "https://api.fireworks.ai/inference",
4756
+ reasoning: true,
4757
+ input: ["text"],
4758
+ cost: {
4759
+ input: 1.2,
4760
+ output: 1.2,
4761
+ cacheRead: 0.6,
4762
+ cacheWrite: 0,
4763
+ },
4764
+ contextWindow: 1000000,
4765
+ maxTokens: 384000,
4766
+ },
4098
4767
  "accounts/fireworks/models/deepseek-v4-pro-0813": {
4099
4768
  id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4100
4769
  name: "DeepSeek V4 Pro 0813",
@@ -4112,6 +4781,23 @@ export const MODELS = {
4112
4781
  contextWindow: 1000000,
4113
4782
  maxTokens: 384000,
4114
4783
  },
4784
+ "accounts/fireworks/models/deepseek-v4p1-flash": {
4785
+ id: "accounts/fireworks/models/deepseek-v4p1-flash",
4786
+ name: "DeepSeek V4.1 Flash",
4787
+ api: "anthropic-messages",
4788
+ provider: "fireworks",
4789
+ baseUrl: "https://api.fireworks.ai/inference",
4790
+ reasoning: true,
4791
+ input: ["text", "image"],
4792
+ cost: {
4793
+ input: 0.22,
4794
+ output: 0.66,
4795
+ cacheRead: 0.007,
4796
+ cacheWrite: 0,
4797
+ },
4798
+ contextWindow: 1000000,
4799
+ maxTokens: 384000,
4800
+ },
4115
4801
  "accounts/fireworks/models/glm-5p2": {
4116
4802
  id: "accounts/fireworks/models/glm-5p2",
4117
4803
  name: "GLM 5.2",
@@ -4143,8 +4829,8 @@ export const MODELS = {
4143
4829
  cacheRead: 0.26,
4144
4830
  cacheWrite: 0,
4145
4831
  },
4146
- contextWindow: 1000000,
4147
- maxTokens: 131072,
4832
+ contextWindow: 1048573,
4833
+ maxTokens: 262144,
4148
4834
  },
4149
4835
  "accounts/fireworks/models/glm-5p3-flash": {
4150
4836
  id: "accounts/fireworks/models/glm-5p3-flash",
@@ -4160,7 +4846,7 @@ export const MODELS = {
4160
4846
  cacheRead: 0.03,
4161
4847
  cacheWrite: 0,
4162
4848
  },
4163
- contextWindow: 1000000,
4849
+ contextWindow: 1048573,
4164
4850
  maxTokens: 131072,
4165
4851
  },
4166
4852
  "accounts/fireworks/models/gpt-oss-120b": {
@@ -4248,6 +4934,23 @@ export const MODELS = {
4248
4934
  contextWindow: 1048576,
4249
4935
  maxTokens: 131072,
4250
4936
  },
4937
+ "accounts/fireworks/models/minimax-m2p7": {
4938
+ id: "accounts/fireworks/models/minimax-m2p7",
4939
+ name: "MiniMax-M2.7",
4940
+ api: "anthropic-messages",
4941
+ provider: "fireworks",
4942
+ baseUrl: "https://api.fireworks.ai/inference",
4943
+ reasoning: true,
4944
+ input: ["text"],
4945
+ cost: {
4946
+ input: 1.2,
4947
+ output: 1.2,
4948
+ cacheRead: 0.6,
4949
+ cacheWrite: 0,
4950
+ },
4951
+ contextWindow: 196608,
4952
+ maxTokens: 131072,
4953
+ },
4251
4954
  "accounts/fireworks/models/minimax-m3": {
4252
4955
  id: "accounts/fireworks/models/minimax-m3",
4253
4956
  name: "MiniMax-M3",
@@ -4255,7 +4958,7 @@ export const MODELS = {
4255
4958
  provider: "fireworks",
4256
4959
  baseUrl: "https://api.fireworks.ai/inference",
4257
4960
  reasoning: true,
4258
- input: ["text", "image"],
4961
+ input: ["text"],
4259
4962
  cost: {
4260
4963
  input: 0.3,
4261
4964
  output: 1.2,
@@ -4265,6 +4968,23 @@ export const MODELS = {
4265
4968
  contextWindow: 512000,
4266
4969
  maxTokens: 512000,
4267
4970
  },
4971
+ "accounts/fireworks/models/mistral-large-3-fp8": {
4972
+ id: "accounts/fireworks/models/mistral-large-3-fp8",
4973
+ name: "Mistral Large 3 675B Instruct 2512",
4974
+ api: "anthropic-messages",
4975
+ provider: "fireworks",
4976
+ baseUrl: "https://api.fireworks.ai/inference",
4977
+ reasoning: false,
4978
+ input: ["text", "image"],
4979
+ cost: {
4980
+ input: 0,
4981
+ output: 0,
4982
+ cacheRead: 0,
4983
+ cacheWrite: 0,
4984
+ },
4985
+ contextWindow: 262144,
4986
+ maxTokens: 262144,
4987
+ },
4268
4988
  "accounts/fireworks/models/muse-glimmer-30b": {
4269
4989
  id: "accounts/fireworks/models/muse-glimmer-30b",
4270
4990
  name: "Muse Glimmer 30B",
@@ -4293,112 +5013,282 @@ export const MODELS = {
4293
5013
  cost: {
4294
5014
  input: 0.6,
4295
5015
  output: 2.4,
4296
- cacheRead: 0.119,
5016
+ cacheRead: 0.12,
5017
+ cacheWrite: 0,
5018
+ },
5019
+ contextWindow: 262144,
5020
+ maxTokens: 128000,
5021
+ },
5022
+ "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b": {
5023
+ id: "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b",
5024
+ name: "Nemotron 3.5 Lightning 30B A3B",
5025
+ api: "anthropic-messages",
5026
+ provider: "fireworks",
5027
+ baseUrl: "https://api.fireworks.ai/inference",
5028
+ reasoning: true,
5029
+ input: ["text"],
5030
+ cost: {
5031
+ input: 0.05,
5032
+ output: 0.2,
5033
+ cacheRead: 0.01,
5034
+ cacheWrite: 0,
5035
+ },
5036
+ contextWindow: 262144,
5037
+ maxTokens: 262144,
5038
+ },
5039
+ "accounts/fireworks/models/qwen3p7-plus": {
5040
+ id: "accounts/fireworks/models/qwen3p7-plus",
5041
+ name: "Qwen 3.7 Plus",
5042
+ api: "anthropic-messages",
5043
+ provider: "fireworks",
5044
+ baseUrl: "https://api.fireworks.ai/inference",
5045
+ reasoning: true,
5046
+ input: ["text", "image"],
5047
+ cost: {
5048
+ input: 0.4,
5049
+ output: 1.6,
5050
+ cacheRead: 0.08,
5051
+ cacheWrite: 0,
5052
+ },
5053
+ contextWindow: 262144,
5054
+ maxTokens: 65536,
5055
+ },
5056
+ "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
5057
+ id: "accounts/fireworks/models/qwen3p8-2p4t-a95b",
5058
+ name: "Qwen3.8 2.4T A95B",
5059
+ api: "anthropic-messages",
5060
+ provider: "fireworks",
5061
+ baseUrl: "https://api.fireworks.ai/inference",
5062
+ reasoning: true,
5063
+ input: ["text"],
5064
+ cost: {
5065
+ input: 2,
5066
+ output: 6,
5067
+ cacheRead: 0.25,
5068
+ cacheWrite: 0,
5069
+ },
5070
+ contextWindow: 262144,
5071
+ maxTokens: 131072,
5072
+ },
5073
+ "accounts/fireworks/models/qwen3p8-max": {
5074
+ id: "accounts/fireworks/models/qwen3p8-max",
5075
+ name: "Qwen3.8 Max",
5076
+ api: "anthropic-messages",
5077
+ provider: "fireworks",
5078
+ baseUrl: "https://api.fireworks.ai/inference",
5079
+ reasoning: true,
5080
+ input: ["text", "image"],
5081
+ cost: {
5082
+ input: 2,
5083
+ output: 6,
5084
+ cacheRead: 0.25,
5085
+ cacheWrite: 0,
5086
+ },
5087
+ contextWindow: 262144,
5088
+ maxTokens: 131072,
5089
+ },
5090
+ "accounts/fireworks/routers/deepseek-flash-latest": {
5091
+ id: "accounts/fireworks/routers/deepseek-flash-latest",
5092
+ name: "DeepSeek Flash Latest",
5093
+ api: "anthropic-messages",
5094
+ provider: "fireworks",
5095
+ baseUrl: "https://api.fireworks.ai/inference",
5096
+ reasoning: true,
5097
+ input: ["text", "image"],
5098
+ cost: {
5099
+ input: 0.22,
5100
+ output: 0.66,
5101
+ cacheRead: 0.007,
5102
+ cacheWrite: 0,
5103
+ },
5104
+ contextWindow: 1000000,
5105
+ maxTokens: 384000,
5106
+ },
5107
+ "accounts/fireworks/routers/deepseek-pro-latest": {
5108
+ id: "accounts/fireworks/routers/deepseek-pro-latest",
5109
+ name: "DeepSeek Pro Latest",
5110
+ api: "anthropic-messages",
5111
+ provider: "fireworks",
5112
+ baseUrl: "https://api.fireworks.ai/inference",
5113
+ reasoning: true,
5114
+ input: ["text"],
5115
+ cost: {
5116
+ input: 1.32,
5117
+ output: 3.96,
5118
+ cacheRead: 0.044,
5119
+ cacheWrite: 0,
5120
+ },
5121
+ contextWindow: 1000000,
5122
+ maxTokens: 384000,
5123
+ },
5124
+ "accounts/fireworks/routers/glm-5p2-fast": {
5125
+ id: "accounts/fireworks/routers/glm-5p2-fast",
5126
+ name: "GLM 5.2 Fast",
5127
+ api: "anthropic-messages",
5128
+ provider: "fireworks",
5129
+ baseUrl: "https://api.fireworks.ai/inference",
5130
+ reasoning: true,
5131
+ input: ["text"],
5132
+ cost: {
5133
+ input: 2.1,
5134
+ output: 6.6,
5135
+ cacheRead: 0.21,
5136
+ cacheWrite: 0,
5137
+ },
5138
+ contextWindow: 1048575,
5139
+ maxTokens: 131072,
5140
+ },
5141
+ "accounts/fireworks/routers/glm-5p3-fast": {
5142
+ id: "accounts/fireworks/routers/glm-5p3-fast",
5143
+ name: "GLM 5.3 Fast",
5144
+ api: "anthropic-messages",
5145
+ provider: "fireworks",
5146
+ baseUrl: "https://api.fireworks.ai/inference",
5147
+ reasoning: true,
5148
+ input: ["text"],
5149
+ cost: {
5150
+ input: 2.1,
5151
+ output: 6.6,
5152
+ cacheRead: 0.39,
5153
+ cacheWrite: 0,
5154
+ },
5155
+ contextWindow: 1048572,
5156
+ maxTokens: 262144,
5157
+ },
5158
+ "accounts/fireworks/routers/glm-fast-latest": {
5159
+ id: "accounts/fireworks/routers/glm-fast-latest",
5160
+ name: "GLM 5.3 Fast (Latest)",
5161
+ api: "anthropic-messages",
5162
+ provider: "fireworks",
5163
+ baseUrl: "https://api.fireworks.ai/inference",
5164
+ reasoning: true,
5165
+ input: ["text"],
5166
+ cost: {
5167
+ input: 2.1,
5168
+ output: 6.6,
5169
+ cacheRead: 0.39,
5170
+ cacheWrite: 0,
5171
+ },
5172
+ contextWindow: 1048572,
5173
+ maxTokens: 262144,
5174
+ },
5175
+ "accounts/fireworks/routers/glm-flash-latest": {
5176
+ id: "accounts/fireworks/routers/glm-flash-latest",
5177
+ name: "GLM Flash Latest (GLM 5.3 Flash)",
5178
+ api: "anthropic-messages",
5179
+ provider: "fireworks",
5180
+ baseUrl: "https://api.fireworks.ai/inference",
5181
+ reasoning: true,
5182
+ input: ["text", "image"],
5183
+ cost: {
5184
+ input: 0.15,
5185
+ output: 0.5,
5186
+ cacheRead: 0.03,
4297
5187
  cacheWrite: 0,
4298
5188
  },
4299
- contextWindow: 262144,
4300
- maxTokens: 128000,
5189
+ contextWindow: 1048573,
5190
+ maxTokens: 131072,
4301
5191
  },
4302
- "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b": {
4303
- id: "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b",
4304
- name: "Nemotron 3.5 Lightning 30B A3B",
5192
+ "accounts/fireworks/routers/glm-latest": {
5193
+ id: "accounts/fireworks/routers/glm-latest",
5194
+ name: "GLM Latest",
4305
5195
  api: "anthropic-messages",
4306
5196
  provider: "fireworks",
4307
5197
  baseUrl: "https://api.fireworks.ai/inference",
4308
5198
  reasoning: true,
4309
5199
  input: ["text"],
4310
5200
  cost: {
4311
- input: 0.05,
4312
- output: 0.2,
4313
- cacheRead: 0.01,
5201
+ input: 1.4,
5202
+ output: 4.4,
5203
+ cacheRead: 0.26,
4314
5204
  cacheWrite: 0,
4315
5205
  },
4316
- contextWindow: 262144,
5206
+ contextWindow: 1048573,
4317
5207
  maxTokens: 262144,
4318
5208
  },
4319
- "accounts/fireworks/models/qwen3p7-plus": {
4320
- id: "accounts/fireworks/models/qwen3p7-plus",
4321
- name: "Qwen 3.7 Plus",
5209
+ "accounts/fireworks/routers/kimi-fast-latest": {
5210
+ id: "accounts/fireworks/routers/kimi-fast-latest",
5211
+ name: "Kimi Fast Latest",
4322
5212
  api: "anthropic-messages",
4323
5213
  provider: "fireworks",
4324
5214
  baseUrl: "https://api.fireworks.ai/inference",
4325
5215
  reasoning: true,
4326
5216
  input: ["text", "image"],
4327
5217
  cost: {
4328
- input: 0.4,
4329
- output: 1.6,
4330
- cacheRead: 0.08,
5218
+ input: 4.5,
5219
+ output: 22.5,
5220
+ cacheRead: 0.45,
4331
5221
  cacheWrite: 0,
4332
5222
  },
4333
- contextWindow: 262144,
4334
- maxTokens: 65536,
5223
+ contextWindow: 1048576,
5224
+ maxTokens: 131072,
4335
5225
  },
4336
- "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
4337
- id: "accounts/fireworks/models/qwen3p8-2p4t-a95b",
4338
- name: "Qwen3.8 2.4T A95B",
5226
+ "accounts/fireworks/routers/kimi-k3-fast": {
5227
+ id: "accounts/fireworks/routers/kimi-k3-fast",
5228
+ name: "Kimi K3 Fast",
4339
5229
  api: "anthropic-messages",
4340
5230
  provider: "fireworks",
4341
5231
  baseUrl: "https://api.fireworks.ai/inference",
4342
5232
  reasoning: true,
4343
- input: ["text"],
5233
+ input: ["text", "image"],
4344
5234
  cost: {
4345
- input: 2,
4346
- output: 6,
4347
- cacheRead: 0.25,
5235
+ input: 4.5,
5236
+ output: 22.5,
5237
+ cacheRead: 0.45,
4348
5238
  cacheWrite: 0,
4349
5239
  },
4350
- contextWindow: 262144,
5240
+ contextWindow: 1048576,
4351
5241
  maxTokens: 131072,
4352
5242
  },
4353
- "accounts/fireworks/models/qwen3p8-max": {
4354
- id: "accounts/fireworks/models/qwen3p8-max",
4355
- name: "Qwen3.8 Max",
5243
+ "accounts/fireworks/routers/kimi-latest": {
5244
+ id: "accounts/fireworks/routers/kimi-latest",
5245
+ name: "Kimi Latest",
4356
5246
  api: "anthropic-messages",
4357
5247
  provider: "fireworks",
4358
5248
  baseUrl: "https://api.fireworks.ai/inference",
4359
5249
  reasoning: true,
4360
- input: ["text"],
5250
+ input: ["text", "image"],
4361
5251
  cost: {
4362
- input: 2,
4363
- output: 6,
4364
- cacheRead: 0.25,
5252
+ input: 3,
5253
+ output: 15,
5254
+ cacheRead: 0.3,
4365
5255
  cacheWrite: 0,
4366
5256
  },
4367
- contextWindow: 262144,
5257
+ contextWindow: 1048576,
4368
5258
  maxTokens: 131072,
4369
5259
  },
4370
- "accounts/fireworks/routers/glm-5p2-fast": {
4371
- id: "accounts/fireworks/routers/glm-5p2-fast",
4372
- name: "GLM 5.2 Fast",
5260
+ "accounts/fireworks/routers/minimax-latest": {
5261
+ id: "accounts/fireworks/routers/minimax-latest",
5262
+ name: "MiniMax Latest",
4373
5263
  api: "anthropic-messages",
4374
5264
  provider: "fireworks",
4375
5265
  baseUrl: "https://api.fireworks.ai/inference",
4376
5266
  reasoning: true,
4377
5267
  input: ["text"],
4378
5268
  cost: {
4379
- input: 2.1,
4380
- output: 6.6,
4381
- cacheRead: 0.21,
5269
+ input: 0.3,
5270
+ output: 1.2,
5271
+ cacheRead: 0.06,
4382
5272
  cacheWrite: 0,
4383
5273
  },
4384
- contextWindow: 1048575,
4385
- maxTokens: 131072,
5274
+ contextWindow: 512000,
5275
+ maxTokens: 512000,
4386
5276
  },
4387
- "accounts/fireworks/routers/kimi-k3-fast": {
4388
- id: "accounts/fireworks/routers/kimi-k3-fast",
4389
- name: "Kimi K3 Fast",
5277
+ "accounts/fireworks/routers/qwen-max-latest": {
5278
+ id: "accounts/fireworks/routers/qwen-max-latest",
5279
+ name: "Qwen Max Latest (Qwen3.8 Max)",
4390
5280
  api: "anthropic-messages",
4391
5281
  provider: "fireworks",
4392
5282
  baseUrl: "https://api.fireworks.ai/inference",
4393
5283
  reasoning: true,
4394
5284
  input: ["text", "image"],
4395
5285
  cost: {
4396
- input: 4.5,
4397
- output: 22.5,
4398
- cacheRead: 0.45,
5286
+ input: 2,
5287
+ output: 6,
5288
+ cacheRead: 0.25,
4399
5289
  cacheWrite: 0,
4400
5290
  },
4401
- contextWindow: 1048576,
5291
+ contextWindow: 262144,
4402
5292
  maxTokens: 131072,
4403
5293
  },
4404
5294
  },
@@ -6396,6 +7286,78 @@ export const MODELS = {
6396
7286
  contextWindow: 1000000,
6397
7287
  maxTokens: 384000,
6398
7288
  },
7289
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
7290
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
7291
+ name: "DeepSeek V4.1 Flash",
7292
+ api: "openai-completions",
7293
+ provider: "huggingface",
7294
+ baseUrl: "https://router.huggingface.co/v1",
7295
+ compat: { "supportsDeveloperRole": false },
7296
+ reasoning: true,
7297
+ input: ["text", "image"],
7298
+ cost: {
7299
+ input: 0.3,
7300
+ output: 1.2,
7301
+ cacheRead: 0,
7302
+ cacheWrite: 0,
7303
+ },
7304
+ contextWindow: 1048576,
7305
+ maxTokens: 384000,
7306
+ },
7307
+ "google/gemma-3-12b-it": {
7308
+ id: "google/gemma-3-12b-it",
7309
+ name: "Gemma 3 12B IT",
7310
+ api: "openai-completions",
7311
+ provider: "huggingface",
7312
+ baseUrl: "https://router.huggingface.co/v1",
7313
+ compat: { "supportsDeveloperRole": false },
7314
+ reasoning: false,
7315
+ input: ["text", "image"],
7316
+ cost: {
7317
+ input: 0.05,
7318
+ output: 0.15,
7319
+ cacheRead: 0,
7320
+ cacheWrite: 0,
7321
+ },
7322
+ contextWindow: 131072,
7323
+ maxTokens: 131072,
7324
+ },
7325
+ "google/gemma-3-27b-it": {
7326
+ id: "google/gemma-3-27b-it",
7327
+ name: "Gemma 3 27B IT",
7328
+ api: "openai-completions",
7329
+ provider: "huggingface",
7330
+ baseUrl: "https://router.huggingface.co/v1",
7331
+ compat: { "supportsDeveloperRole": false },
7332
+ reasoning: false,
7333
+ input: ["text", "image"],
7334
+ cost: {
7335
+ input: 0.08,
7336
+ output: 0.16,
7337
+ cacheRead: 0,
7338
+ cacheWrite: 0,
7339
+ },
7340
+ contextWindow: 131072,
7341
+ maxTokens: 131072,
7342
+ },
7343
+ "google/gemma-3-4b-it": {
7344
+ id: "google/gemma-3-4b-it",
7345
+ name: "Gemma 3 4B IT",
7346
+ api: "openai-completions",
7347
+ provider: "huggingface",
7348
+ baseUrl: "https://router.huggingface.co/v1",
7349
+ compat: { "supportsDeveloperRole": false },
7350
+ reasoning: false,
7351
+ input: ["text", "image"],
7352
+ cost: {
7353
+ input: 0.05,
7354
+ output: 0.1,
7355
+ cacheRead: 0,
7356
+ cacheWrite: 0,
7357
+ },
7358
+ contextWindow: 131072,
7359
+ maxTokens: 131072,
7360
+ },
6399
7361
  "google/gemma-4-26B-A4B-it": {
6400
7362
  id: "google/gemma-4-26B-A4B-it",
6401
7363
  name: "Gemma 4 26B A4B IT",
@@ -6976,7 +7938,7 @@ export const MODELS = {
6976
7938
  },
6977
7939
  "kimi-for-coding": {
6978
7940
  id: "kimi-for-coding",
6979
- name: "Kimi K2.7 Code",
7941
+ name: "kimi-for-coding",
6980
7942
  api: "anthropic-messages",
6981
7943
  provider: "kimi-coding",
6982
7944
  baseUrl: "https://api.kimi.com/coding",
@@ -6989,7 +7951,7 @@ export const MODELS = {
6989
7951
  cacheRead: 0,
6990
7952
  cacheWrite: 0,
6991
7953
  },
6992
- contextWindow: 262144,
7954
+ contextWindow: 1048576,
6993
7955
  maxTokens: 32768,
6994
7956
  },
6995
7957
  "kimi-for-coding-highspeed": {
@@ -7609,169 +8571,61 @@ export const MODELS = {
7609
8571
  cacheWrite: 0,
7610
8572
  },
7611
8573
  contextWindow: 128000,
7612
- maxTokens: 128000,
7613
- },
7614
- "pixtral-large-latest": {
7615
- id: "pixtral-large-latest",
7616
- name: "Pixtral Large (latest)",
7617
- api: "mistral-conversations",
7618
- provider: "mistral",
7619
- baseUrl: "https://api.mistral.ai",
7620
- reasoning: false,
7621
- input: ["text", "image"],
7622
- cost: {
7623
- input: 2,
7624
- output: 6,
7625
- cacheRead: 0,
7626
- cacheWrite: 0,
7627
- },
7628
- contextWindow: 128000,
7629
- maxTokens: 128000,
7630
- },
7631
- "voxtral-small-latest": {
7632
- id: "voxtral-small-latest",
7633
- name: "Voxtral Small (latest)",
7634
- api: "mistral-conversations",
7635
- provider: "mistral",
7636
- baseUrl: "https://api.mistral.ai",
7637
- reasoning: false,
7638
- input: ["text"],
7639
- cost: {
7640
- input: 0.1,
7641
- output: 0.3,
7642
- cacheRead: 0,
7643
- cacheWrite: 0,
7644
- },
7645
- contextWindow: 32000,
7646
- maxTokens: 32000,
7647
- },
7648
- "zai-glm-5-2": {
7649
- id: "zai-glm-5-2",
7650
- name: "GLM-5.2",
7651
- api: "mistral-conversations",
7652
- provider: "mistral",
7653
- baseUrl: "https://api.mistral.ai",
7654
- reasoning: true,
7655
- input: ["text"],
7656
- cost: {
7657
- input: 1.4,
7658
- output: 4.4,
7659
- cacheRead: 0.14,
7660
- cacheWrite: 0,
7661
- },
7662
- contextWindow: 1000000,
7663
- maxTokens: 131072,
7664
- },
7665
- },
7666
- "moonshotai": {
7667
- "kimi-k2-0711-preview": {
7668
- id: "kimi-k2-0711-preview",
7669
- name: "Kimi K2 0711",
7670
- api: "openai-completions",
7671
- provider: "moonshotai",
7672
- baseUrl: "https://api.moonshot.ai/v1",
7673
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7674
- reasoning: false,
7675
- input: ["text"],
7676
- cost: {
7677
- input: 0.6,
7678
- output: 2.5,
7679
- cacheRead: 0.15,
7680
- cacheWrite: 0,
7681
- },
7682
- contextWindow: 131072,
7683
- maxTokens: 16384,
7684
- },
7685
- "kimi-k2-0905-preview": {
7686
- id: "kimi-k2-0905-preview",
7687
- name: "Kimi K2 0905",
7688
- api: "openai-completions",
7689
- provider: "moonshotai",
7690
- baseUrl: "https://api.moonshot.ai/v1",
7691
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7692
- reasoning: false,
7693
- input: ["text"],
7694
- cost: {
7695
- input: 0.6,
7696
- output: 2.5,
7697
- cacheRead: 0.15,
7698
- cacheWrite: 0,
7699
- },
7700
- contextWindow: 262144,
7701
- maxTokens: 262144,
7702
- },
7703
- "kimi-k2-thinking": {
7704
- id: "kimi-k2-thinking",
7705
- name: "Kimi K2 Thinking",
7706
- api: "openai-completions",
7707
- provider: "moonshotai",
7708
- baseUrl: "https://api.moonshot.ai/v1",
7709
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7710
- reasoning: true,
7711
- input: ["text"],
7712
- cost: {
7713
- input: 0.6,
7714
- output: 2.5,
7715
- cacheRead: 0.15,
7716
- cacheWrite: 0,
7717
- },
7718
- contextWindow: 262144,
7719
- maxTokens: 262144,
7720
- },
7721
- "kimi-k2-thinking-turbo": {
7722
- id: "kimi-k2-thinking-turbo",
7723
- name: "Kimi K2 Thinking Turbo",
7724
- api: "openai-completions",
7725
- provider: "moonshotai",
7726
- baseUrl: "https://api.moonshot.ai/v1",
7727
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7728
- reasoning: true,
7729
- input: ["text"],
8574
+ maxTokens: 128000,
8575
+ },
8576
+ "pixtral-large-latest": {
8577
+ id: "pixtral-large-latest",
8578
+ name: "Pixtral Large (latest)",
8579
+ api: "mistral-conversations",
8580
+ provider: "mistral",
8581
+ baseUrl: "https://api.mistral.ai",
8582
+ reasoning: false,
8583
+ input: ["text", "image"],
7730
8584
  cost: {
7731
- input: 1.15,
7732
- output: 8,
7733
- cacheRead: 0.15,
8585
+ input: 2,
8586
+ output: 6,
8587
+ cacheRead: 0,
7734
8588
  cacheWrite: 0,
7735
8589
  },
7736
- contextWindow: 262144,
7737
- maxTokens: 262144,
8590
+ contextWindow: 128000,
8591
+ maxTokens: 128000,
7738
8592
  },
7739
- "kimi-k2-turbo-preview": {
7740
- id: "kimi-k2-turbo-preview",
7741
- name: "Kimi K2 Turbo",
7742
- api: "openai-completions",
7743
- provider: "moonshotai",
7744
- baseUrl: "https://api.moonshot.ai/v1",
7745
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
8593
+ "voxtral-small-latest": {
8594
+ id: "voxtral-small-latest",
8595
+ name: "Voxtral Small (latest)",
8596
+ api: "mistral-conversations",
8597
+ provider: "mistral",
8598
+ baseUrl: "https://api.mistral.ai",
7746
8599
  reasoning: false,
7747
8600
  input: ["text"],
7748
8601
  cost: {
7749
- input: 2.4,
7750
- output: 10,
7751
- cacheRead: 0.6,
8602
+ input: 0.1,
8603
+ output: 0.3,
8604
+ cacheRead: 0,
7752
8605
  cacheWrite: 0,
7753
8606
  },
7754
- contextWindow: 262144,
7755
- maxTokens: 262144,
8607
+ contextWindow: 32000,
8608
+ maxTokens: 32000,
7756
8609
  },
7757
- "kimi-k2.5": {
7758
- id: "kimi-k2.5",
7759
- name: "Kimi K2.5",
7760
- api: "openai-completions",
7761
- provider: "moonshotai",
7762
- baseUrl: "https://api.moonshot.ai/v1",
7763
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
8610
+ "zai-glm-5-2": {
8611
+ id: "zai-glm-5-2",
8612
+ name: "GLM-5.2",
8613
+ api: "mistral-conversations",
8614
+ provider: "mistral",
8615
+ baseUrl: "https://api.mistral.ai",
7764
8616
  reasoning: true,
7765
- input: ["text", "image"],
8617
+ input: ["text"],
7766
8618
  cost: {
7767
- input: 0.6,
7768
- output: 3,
7769
- cacheRead: 0.1,
8619
+ input: 1.4,
8620
+ output: 4.4,
8621
+ cacheRead: 0.14,
7770
8622
  cacheWrite: 0,
7771
8623
  },
7772
- contextWindow: 262144,
7773
- maxTokens: 262144,
8624
+ contextWindow: 1000000,
8625
+ maxTokens: 131072,
7774
8626
  },
8627
+ },
8628
+ "moonshotai": {
7775
8629
  "kimi-k2.6": {
7776
8630
  id: "kimi-k2.6",
7777
8631
  name: "Kimi K2.6",
@@ -7846,114 +8700,6 @@ export const MODELS = {
7846
8700
  },
7847
8701
  },
7848
8702
  "moonshotai-cn": {
7849
- "kimi-k2-0711-preview": {
7850
- id: "kimi-k2-0711-preview",
7851
- name: "Kimi K2 0711",
7852
- api: "openai-completions",
7853
- provider: "moonshotai-cn",
7854
- baseUrl: "https://api.moonshot.cn/v1",
7855
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7856
- reasoning: false,
7857
- input: ["text"],
7858
- cost: {
7859
- input: 0.6,
7860
- output: 2.5,
7861
- cacheRead: 0.15,
7862
- cacheWrite: 0,
7863
- },
7864
- contextWindow: 131072,
7865
- maxTokens: 16384,
7866
- },
7867
- "kimi-k2-0905-preview": {
7868
- id: "kimi-k2-0905-preview",
7869
- name: "Kimi K2 0905",
7870
- api: "openai-completions",
7871
- provider: "moonshotai-cn",
7872
- baseUrl: "https://api.moonshot.cn/v1",
7873
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7874
- reasoning: false,
7875
- input: ["text"],
7876
- cost: {
7877
- input: 0.6,
7878
- output: 2.5,
7879
- cacheRead: 0.15,
7880
- cacheWrite: 0,
7881
- },
7882
- contextWindow: 262144,
7883
- maxTokens: 262144,
7884
- },
7885
- "kimi-k2-thinking": {
7886
- id: "kimi-k2-thinking",
7887
- name: "Kimi K2 Thinking",
7888
- api: "openai-completions",
7889
- provider: "moonshotai-cn",
7890
- baseUrl: "https://api.moonshot.cn/v1",
7891
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7892
- reasoning: true,
7893
- input: ["text"],
7894
- cost: {
7895
- input: 0.6,
7896
- output: 2.5,
7897
- cacheRead: 0.15,
7898
- cacheWrite: 0,
7899
- },
7900
- contextWindow: 262144,
7901
- maxTokens: 262144,
7902
- },
7903
- "kimi-k2-thinking-turbo": {
7904
- id: "kimi-k2-thinking-turbo",
7905
- name: "Kimi K2 Thinking Turbo",
7906
- api: "openai-completions",
7907
- provider: "moonshotai-cn",
7908
- baseUrl: "https://api.moonshot.cn/v1",
7909
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7910
- reasoning: true,
7911
- input: ["text"],
7912
- cost: {
7913
- input: 1.15,
7914
- output: 8,
7915
- cacheRead: 0.15,
7916
- cacheWrite: 0,
7917
- },
7918
- contextWindow: 262144,
7919
- maxTokens: 262144,
7920
- },
7921
- "kimi-k2-turbo-preview": {
7922
- id: "kimi-k2-turbo-preview",
7923
- name: "Kimi K2 Turbo",
7924
- api: "openai-completions",
7925
- provider: "moonshotai-cn",
7926
- baseUrl: "https://api.moonshot.cn/v1",
7927
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7928
- reasoning: false,
7929
- input: ["text"],
7930
- cost: {
7931
- input: 2.4,
7932
- output: 10,
7933
- cacheRead: 0.6,
7934
- cacheWrite: 0,
7935
- },
7936
- contextWindow: 262144,
7937
- maxTokens: 262144,
7938
- },
7939
- "kimi-k2.5": {
7940
- id: "kimi-k2.5",
7941
- name: "Kimi K2.5",
7942
- api: "openai-completions",
7943
- provider: "moonshotai-cn",
7944
- baseUrl: "https://api.moonshot.cn/v1",
7945
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7946
- reasoning: true,
7947
- input: ["text", "image"],
7948
- cost: {
7949
- input: 0.6,
7950
- output: 3,
7951
- cacheRead: 0.1,
7952
- cacheWrite: 0,
7953
- },
7954
- contextWindow: 262144,
7955
- maxTokens: 262144,
7956
- },
7957
8703
  "kimi-k2.6": {
7958
8704
  id: "kimi-k2.6",
7959
8705
  name: "Kimi K2.6",
@@ -10086,9 +10832,9 @@ export const MODELS = {
10086
10832
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10087
10833
  input: ["text"],
10088
10834
  cost: {
10089
- input: 0.22,
10090
- output: 0.66,
10091
- cacheRead: 0.007,
10835
+ input: 0.15,
10836
+ output: 0.6,
10837
+ cacheRead: 0.003,
10092
10838
  cacheWrite: 0,
10093
10839
  },
10094
10840
  contextWindow: 1000000,
@@ -10105,9 +10851,9 @@ export const MODELS = {
10105
10851
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10106
10852
  input: ["text", "image"],
10107
10853
  cost: {
10108
- input: 0.22,
10109
- output: 0.66,
10110
- cacheRead: 0.007,
10854
+ input: 0.15,
10855
+ output: 0.6,
10856
+ cacheRead: 0.003,
10111
10857
  cacheWrite: 0,
10112
10858
  },
10113
10859
  contextWindow: 1000000,
@@ -10132,6 +10878,25 @@ export const MODELS = {
10132
10878
  contextWindow: 1000000,
10133
10879
  maxTokens: 384000,
10134
10880
  },
10881
+ "deepseek-v4.1-flash": {
10882
+ id: "deepseek-v4.1-flash",
10883
+ name: "DeepSeek V4.1 Flash",
10884
+ api: "openai-completions",
10885
+ provider: "opencode-go",
10886
+ baseUrl: "https://opencode.ai/zen/go/v1",
10887
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
10888
+ reasoning: true,
10889
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10890
+ input: ["text", "image"],
10891
+ cost: {
10892
+ input: 0.15,
10893
+ output: 0.6,
10894
+ cacheRead: 0.003,
10895
+ cacheWrite: 0,
10896
+ },
10897
+ contextWindow: 1000000,
10898
+ maxTokens: 384000,
10899
+ },
10135
10900
  "glm-5.1": {
10136
10901
  id: "glm-5.1",
10137
10902
  name: "GLM-5.1",
@@ -10185,16 +10950,16 @@ export const MODELS = {
10185
10950
  },
10186
10951
  "glm-5.3-flash": {
10187
10952
  id: "glm-5.3-flash",
10188
- name: "GLM-5.3-Flash (2x usage)",
10953
+ name: "GLM-5.3-Flash",
10189
10954
  api: "openai-completions",
10190
10955
  provider: "opencode-go",
10191
10956
  baseUrl: "https://opencode.ai/zen/go/v1",
10192
10957
  reasoning: true,
10193
10958
  input: ["text", "image"],
10194
10959
  cost: {
10195
- input: 0.075,
10196
- output: 0.25,
10197
- cacheRead: 0.015,
10960
+ input: 0.15,
10961
+ output: 0.5,
10962
+ cacheRead: 0.03,
10198
10963
  cacheWrite: 0,
10199
10964
  },
10200
10965
  contextWindow: 1000000,
@@ -10439,23 +11204,6 @@ export const MODELS = {
10439
11204
  contextWindow: 1048576,
10440
11205
  maxTokens: 131072,
10441
11206
  },
10442
- "omen-alpha": {
10443
- id: "omen-alpha",
10444
- name: "Omen Alpha",
10445
- api: "openai-completions",
10446
- provider: "opencode-go",
10447
- baseUrl: "https://opencode.ai/zen/go/v1",
10448
- reasoning: true,
10449
- input: ["text", "image"],
10450
- cost: {
10451
- input: 0.2,
10452
- output: 0.66,
10453
- cacheRead: 0.04,
10454
- cacheWrite: 0,
10455
- },
10456
- contextWindow: 500000,
10457
- maxTokens: 128000,
10458
- },
10459
11207
  "qwen3.6-plus": {
10460
11208
  id: "qwen3.6-plus",
10461
11209
  name: "Qwen3.6 Plus",
@@ -10998,7 +11746,7 @@ export const MODELS = {
10998
11746
  },
10999
11747
  "anthropic/claude-opus-5": {
11000
11748
  id: "anthropic/claude-opus-5",
11001
- name: "Claude Opus 5",
11749
+ name: "Anthropic: Claude Opus 5",
11002
11750
  api: "openai-completions",
11003
11751
  provider: "openrouter",
11004
11752
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11015,7 +11763,7 @@ export const MODELS = {
11015
11763
  },
11016
11764
  "anthropic/claude-opus-5:batch": {
11017
11765
  id: "anthropic/claude-opus-5:batch",
11018
- name: "Claude Opus 5 (batch)",
11766
+ name: "Anthropic: Claude Opus 5 (batch)",
11019
11767
  api: "openai-completions",
11020
11768
  provider: "openrouter",
11021
11769
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11345,13 +12093,13 @@ export const MODELS = {
11345
12093
  reasoning: false,
11346
12094
  input: ["text"],
11347
12095
  cost: {
11348
- input: 0.32,
11349
- output: 0.8899999999999999,
12096
+ input: 0.2574,
12097
+ output: 1.0287,
11350
12098
  cacheRead: 0,
11351
12099
  cacheWrite: 0,
11352
12100
  },
11353
12101
  contextWindow: 163840,
11354
- maxTokens: 16384,
12102
+ maxTokens: 16000,
11355
12103
  },
11356
12104
  "deepseek/deepseek-chat-v3-0324": {
11357
12105
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11379,13 +12127,13 @@ export const MODELS = {
11379
12127
  reasoning: true,
11380
12128
  input: ["text"],
11381
12129
  cost: {
11382
- input: 0.55,
11383
- output: 1.6500000000000001,
11384
- cacheRead: 0.55,
12130
+ input: 0.25,
12131
+ output: 0.95,
12132
+ cacheRead: 0.13,
11385
12133
  cacheWrite: 0,
11386
12134
  },
11387
12135
  contextWindow: 163840,
11388
- maxTokens: 144900,
12136
+ maxTokens: 32768,
11389
12137
  },
11390
12138
  "deepseek/deepseek-r1": {
11391
12139
  id: "deepseek/deepseek-r1",
@@ -11483,9 +12231,9 @@ export const MODELS = {
11483
12231
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11484
12232
  input: ["text"],
11485
12233
  cost: {
11486
- input: 0.08358,
11487
- output: 0.16716,
11488
- cacheRead: 0.016715999999999998,
12234
+ input: 0.088606,
12235
+ output: 0.177212,
12236
+ cacheRead: 0.017721200000000003,
11489
12237
  cacheWrite: 0,
11490
12238
  },
11491
12239
  contextWindow: 1048576,
@@ -11502,9 +12250,9 @@ export const MODELS = {
11502
12250
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11503
12251
  input: ["text"],
11504
12252
  cost: {
11505
- input: 0.065,
11506
- output: 0.18,
11507
- cacheRead: 0.016,
12253
+ input: 0.06,
12254
+ output: 0.12,
12255
+ cacheRead: 0.012,
11508
12256
  cacheWrite: 0,
11509
12257
  },
11510
12258
  contextWindow: 1310720,
@@ -11521,9 +12269,9 @@ export const MODELS = {
11521
12269
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11522
12270
  input: ["text"],
11523
12271
  cost: {
11524
- input: 0.14,
11525
- output: 0.28,
11526
- cacheRead: 0.03,
12272
+ input: 0.11,
12273
+ output: 0.33,
12274
+ cacheRead: 0.0035,
11527
12275
  cacheWrite: 0,
11528
12276
  },
11529
12277
  contextWindow: 1048576,
@@ -11546,7 +12294,26 @@ export const MODELS = {
11546
12294
  cacheWrite: 0,
11547
12295
  },
11548
12296
  contextWindow: 1048576,
11549
- maxTokens: 384000,
12297
+ maxTokens: 943718,
12298
+ },
12299
+ "deepseek/deepseek-v4-flash-vision-exp:batch": {
12300
+ id: "deepseek/deepseek-v4-flash-vision-exp:batch",
12301
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp (batch)",
12302
+ api: "openai-completions",
12303
+ provider: "openrouter",
12304
+ baseUrl: "https://openrouter.ai/api/v1",
12305
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
12306
+ reasoning: true,
12307
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12308
+ input: ["text", "image"],
12309
+ cost: {
12310
+ input: 0.11,
12311
+ output: 0.33,
12312
+ cacheRead: 0.0035,
12313
+ cacheWrite: 0,
12314
+ },
12315
+ contextWindow: 1048576,
12316
+ maxTokens: 943718,
11550
12317
  },
11551
12318
  "deepseek/deepseek-v4-pro": {
11552
12319
  id: "deepseek/deepseek-v4-pro",
@@ -11559,13 +12326,13 @@ export const MODELS = {
11559
12326
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11560
12327
  input: ["text"],
11561
12328
  cost: {
11562
- input: 0.8193659999999999,
11563
- output: 1.6387319999999999,
11564
- cacheRead: 0.0682805,
12329
+ input: 1.5999999999999999,
12330
+ output: 3.1999999999999997,
12331
+ cacheRead: 0.135,
11565
12332
  cacheWrite: 0,
11566
12333
  },
11567
12334
  contextWindow: 1048576,
11568
- maxTokens: 384000,
12335
+ maxTokens: 393216,
11569
12336
  },
11570
12337
  "deepseek/deepseek-v4-pro-0813": {
11571
12338
  id: "deepseek/deepseek-v4-pro-0813",
@@ -11578,32 +12345,51 @@ export const MODELS = {
11578
12345
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11579
12346
  input: ["text"],
11580
12347
  cost: {
11581
- input: 0.57948,
11582
- output: 1.73844,
11583
- cacheRead: 0.019316,
12348
+ input: 0.9833999999999999,
12349
+ output: 2.9502,
12350
+ cacheRead: 0.03278,
12351
+ cacheWrite: 0,
12352
+ },
12353
+ contextWindow: 1048576,
12354
+ maxTokens: 384000,
12355
+ },
12356
+ "deepseek/deepseek-v4-pro-0813:batch": {
12357
+ id: "deepseek/deepseek-v4-pro-0813:batch",
12358
+ name: "DeepSeek: DeepSeek V4 Pro 0813 (batch)",
12359
+ api: "openai-completions",
12360
+ provider: "openrouter",
12361
+ baseUrl: "https://openrouter.ai/api/v1",
12362
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
12363
+ reasoning: true,
12364
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12365
+ input: ["text"],
12366
+ cost: {
12367
+ input: 0.66,
12368
+ output: 1.9800000000000002,
12369
+ cacheRead: 0.022,
11584
12370
  cacheWrite: 0,
11585
12371
  },
11586
12372
  contextWindow: 1048576,
11587
- maxTokens: 384000,
12373
+ maxTokens: 943718,
11588
12374
  },
11589
- "deepseek/deepseek-v4-pro-0813:batch": {
11590
- id: "deepseek/deepseek-v4-pro-0813:batch",
11591
- name: "DeepSeek: DeepSeek V4 Pro 0813 (batch)",
12375
+ "deepseek/deepseek-v4.1-flash": {
12376
+ id: "deepseek/deepseek-v4.1-flash",
12377
+ name: "DeepSeek: DeepSeek V4.1 Flash",
11592
12378
  api: "openai-completions",
11593
12379
  provider: "openrouter",
11594
12380
  baseUrl: "https://openrouter.ai/api/v1",
11595
12381
  compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
11596
12382
  reasoning: true,
11597
12383
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11598
- input: ["text"],
12384
+ input: ["text", "image"],
11599
12385
  cost: {
11600
- input: 1.32,
11601
- output: 3.9600000000000004,
11602
- cacheRead: 0.13,
12386
+ input: 0.3,
12387
+ output: 1.2,
12388
+ cacheRead: 0.006,
11603
12389
  cacheWrite: 0,
11604
12390
  },
11605
12391
  contextWindow: 1048576,
11606
- maxTokens: 943718,
12392
+ maxTokens: 384000,
11607
12393
  },
11608
12394
  "dots-studio/dots-3-note-preview:free": {
11609
12395
  id: "dots-studio/dots-3-note-preview:free",
@@ -11724,23 +12510,6 @@ export const MODELS = {
11724
12510
  contextWindow: 1048576,
11725
12511
  maxTokens: 65536,
11726
12512
  },
11727
- "google/gemini-2.5-pro-preview-05-06": {
11728
- id: "google/gemini-2.5-pro-preview-05-06",
11729
- name: "Google: Gemini 2.5 Pro Preview 05-06",
11730
- api: "openai-completions",
11731
- provider: "openrouter",
11732
- baseUrl: "https://openrouter.ai/api/v1",
11733
- reasoning: true,
11734
- input: ["text", "image"],
11735
- cost: {
11736
- input: 1.25,
11737
- output: 10,
11738
- cacheRead: 0.125,
11739
- cacheWrite: 0.375,
11740
- },
11741
- contextWindow: 1048576,
11742
- maxTokens: 65535,
11743
- },
11744
12513
  "google/gemini-2.5-pro:batch": {
11745
12514
  id: "google/gemini-2.5-pro:batch",
11746
12515
  name: "Google: Gemini 2.5 Pro (batch)",
@@ -12124,13 +12893,13 @@ export const MODELS = {
12124
12893
  reasoning: true,
12125
12894
  input: ["text", "image"],
12126
12895
  cost: {
12127
- input: 0.07,
12128
- output: 0.33999999999999997,
12129
- cacheRead: 0,
12896
+ input: 0.09,
12897
+ output: 0.3,
12898
+ cacheRead: 0.049999999999999996,
12130
12899
  cacheWrite: 0,
12131
12900
  },
12132
12901
  contextWindow: 262144,
12133
- maxTokens: 16384,
12902
+ maxTokens: 235929,
12134
12903
  },
12135
12904
  "google/gemma-4-26b-a4b-it:free": {
12136
12905
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12166,23 +12935,6 @@ export const MODELS = {
12166
12935
  contextWindow: 262144,
12167
12936
  maxTokens: 16384,
12168
12937
  },
12169
- "google/gemma-4-31b-it:batch": {
12170
- id: "google/gemma-4-31b-it:batch",
12171
- name: "Google: Gemma 4 31B (batch)",
12172
- api: "openai-completions",
12173
- provider: "openrouter",
12174
- baseUrl: "https://openrouter.ai/api/v1",
12175
- reasoning: true,
12176
- input: ["text", "image"],
12177
- cost: {
12178
- input: 0.39,
12179
- output: 0.9700000000000001,
12180
- cacheRead: 0,
12181
- cacheWrite: 0,
12182
- },
12183
- contextWindow: 262144,
12184
- maxTokens: 235929,
12185
- },
12186
12938
  "google/gemma-4-31b-it:free": {
12187
12939
  id: "google/gemma-4-31b-it:free",
12188
12940
  name: "Google: Gemma 4 31B (free)",
@@ -12209,9 +12961,9 @@ export const MODELS = {
12209
12961
  reasoning: true,
12210
12962
  input: ["text"],
12211
12963
  cost: {
12212
- input: 0.09999999999999999,
12213
- output: 0.15,
12214
- cacheRead: 0.049999999999999996,
12964
+ input: 0.06,
12965
+ output: 0.25,
12966
+ cacheRead: 0.015,
12215
12967
  cacheWrite: 0,
12216
12968
  },
12217
12969
  contextWindow: 131072,
@@ -12234,9 +12986,9 @@ export const MODELS = {
12234
12986
  contextWindow: 128000,
12235
12987
  maxTokens: 50000,
12236
12988
  },
12237
- "inception/mercury-2.5-preview": {
12238
- id: "inception/mercury-2.5-preview",
12239
- name: "Inception: Mercury 2.5 Preview",
12989
+ "inception/mercury-2.5": {
12990
+ id: "inception/mercury-2.5",
12991
+ name: "Inception: Mercury 2.5",
12240
12992
  api: "openai-completions",
12241
12993
  provider: "openrouter",
12242
12994
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12319,6 +13071,40 @@ export const MODELS = {
12319
13071
  contextWindow: 262144,
12320
13072
  maxTokens: 32768,
12321
13073
  },
13074
+ "inclusionai/ling-3.0-flash-vl": {
13075
+ id: "inclusionai/ling-3.0-flash-vl",
13076
+ name: "inclusionAI: Ling 3.0 Flash VL",
13077
+ api: "openai-completions",
13078
+ provider: "openrouter",
13079
+ baseUrl: "https://openrouter.ai/api/v1",
13080
+ reasoning: true,
13081
+ input: ["text", "image"],
13082
+ cost: {
13083
+ input: 0.06,
13084
+ output: 0.18,
13085
+ cacheRead: 0.012,
13086
+ cacheWrite: 0,
13087
+ },
13088
+ contextWindow: 131072,
13089
+ maxTokens: 32768,
13090
+ },
13091
+ "inclusionai/ling-3.0-flash-vl:free": {
13092
+ id: "inclusionai/ling-3.0-flash-vl:free",
13093
+ name: "inclusionAI: Ling 3.0 Flash VL (free)",
13094
+ api: "openai-completions",
13095
+ provider: "openrouter",
13096
+ baseUrl: "https://openrouter.ai/api/v1",
13097
+ reasoning: true,
13098
+ input: ["text", "image"],
13099
+ cost: {
13100
+ input: 0,
13101
+ output: 0,
13102
+ cacheRead: 0,
13103
+ cacheWrite: 0,
13104
+ },
13105
+ contextWindow: 262144,
13106
+ maxTokens: 32768,
13107
+ },
12322
13108
  "kwaipilot/kat-coder-pro-v2": {
12323
13109
  id: "kwaipilot/kat-coder-pro-v2",
12324
13110
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -12447,13 +13233,13 @@ export const MODELS = {
12447
13233
  reasoning: false,
12448
13234
  input: ["text", "image"],
12449
13235
  cost: {
12450
- input: 0.19999999999999998,
12451
- output: 0.696,
13236
+ input: 0.1875,
13237
+ output: 0.6525,
12452
13238
  cacheRead: 0,
12453
13239
  cacheWrite: 0,
12454
13240
  },
12455
13241
  contextWindow: 1048576,
12456
- maxTokens: 115200,
13242
+ maxTokens: 16384,
12457
13243
  },
12458
13244
  "meta-llama/llama-4-scout": {
12459
13245
  id: "meta-llama/llama-4-scout",
@@ -12481,8 +13267,8 @@ export const MODELS = {
12481
13267
  reasoning: true,
12482
13268
  input: ["text", "image"],
12483
13269
  cost: {
12484
- input: 0.3,
12485
- output: 1.1,
13270
+ input: 0.35,
13271
+ output: 1.5,
12486
13272
  cacheRead: 0.04,
12487
13273
  cacheWrite: 0,
12488
13274
  },
@@ -12498,9 +13284,9 @@ export const MODELS = {
12498
13284
  reasoning: true,
12499
13285
  input: ["text", "image"],
12500
13286
  cost: {
12501
- input: 0.35,
12502
- output: 1.5,
12503
- cacheRead: 0.04,
13287
+ input: 0.175,
13288
+ output: 0.75,
13289
+ cacheRead: 0.02,
12504
13290
  cacheWrite: 0,
12505
13291
  },
12506
13292
  contextWindow: 131072,
@@ -12600,7 +13386,7 @@ export const MODELS = {
12600
13386
  reasoning: true,
12601
13387
  input: ["text"],
12602
13388
  cost: {
12603
- input: 0.55,
13389
+ input: 0.39999999999999997,
12604
13390
  output: 2.2,
12605
13391
  cacheRead: 0,
12606
13392
  cacheWrite: 0,
@@ -12676,23 +13462,6 @@ export const MODELS = {
12676
13462
  contextWindow: 204800,
12677
13463
  maxTokens: 131072,
12678
13464
  },
12679
- "minimax/minimax-m2.7:free": {
12680
- id: "minimax/minimax-m2.7:free",
12681
- name: "MiniMax: MiniMax M2.7 (free)",
12682
- api: "openai-completions",
12683
- provider: "openrouter",
12684
- baseUrl: "https://openrouter.ai/api/v1",
12685
- reasoning: true,
12686
- input: ["text"],
12687
- cost: {
12688
- input: 0,
12689
- output: 0,
12690
- cacheRead: 0,
12691
- cacheWrite: 0,
12692
- },
12693
- contextWindow: 196608,
12694
- maxTokens: 176947,
12695
- },
12696
13465
  "minimax/minimax-m3": {
12697
13466
  id: "minimax/minimax-m3",
12698
13467
  name: "MiniMax: MiniMax M3",
@@ -12727,35 +13496,35 @@ export const MODELS = {
12727
13496
  contextWindow: 524288,
12728
13497
  maxTokens: 471859,
12729
13498
  },
12730
- "minimax/minimax-m3:free": {
12731
- id: "minimax/minimax-m3:free",
12732
- name: "MiniMax: MiniMax M3 (free)",
13499
+ "mistralai/codestral-2508": {
13500
+ id: "mistralai/codestral-2508",
13501
+ name: "Mistral: Codestral 2508",
12733
13502
  api: "openai-completions",
12734
13503
  provider: "openrouter",
12735
13504
  baseUrl: "https://openrouter.ai/api/v1",
12736
- reasoning: true,
12737
- input: ["text", "image"],
13505
+ reasoning: false,
13506
+ input: ["text"],
12738
13507
  cost: {
12739
- input: 0,
12740
- output: 0,
12741
- cacheRead: 0,
13508
+ input: 0.3,
13509
+ output: 0.8999999999999999,
13510
+ cacheRead: 0.03,
12742
13511
  cacheWrite: 0,
12743
13512
  },
12744
- contextWindow: 1048576,
12745
- maxTokens: 943718,
13513
+ contextWindow: 256000,
13514
+ maxTokens: 204800,
12746
13515
  },
12747
- "mistralai/codestral-2508": {
12748
- id: "mistralai/codestral-2508",
12749
- name: "Mistral: Codestral 2508",
13516
+ "mistralai/codestral-2508:batch": {
13517
+ id: "mistralai/codestral-2508:batch",
13518
+ name: "Mistral: Codestral 2508 (batch)",
12750
13519
  api: "openai-completions",
12751
13520
  provider: "openrouter",
12752
13521
  baseUrl: "https://openrouter.ai/api/v1",
12753
13522
  reasoning: false,
12754
13523
  input: ["text"],
12755
13524
  cost: {
12756
- input: 0.3,
12757
- output: 0.8999999999999999,
12758
- cacheRead: 0.03,
13525
+ input: 0.15,
13526
+ output: 0.44999999999999996,
13527
+ cacheRead: 0.015,
12759
13528
  cacheWrite: 0,
12760
13529
  },
12761
13530
  contextWindow: 256000,
@@ -12829,6 +13598,23 @@ export const MODELS = {
12829
13598
  contextWindow: 262144,
12830
13599
  maxTokens: 209715,
12831
13600
  },
13601
+ "mistralai/ministral-8b-2512:batch": {
13602
+ id: "mistralai/ministral-8b-2512:batch",
13603
+ name: "Mistral: Ministral 3 8B 2512 (batch)",
13604
+ api: "openai-completions",
13605
+ provider: "openrouter",
13606
+ baseUrl: "https://openrouter.ai/api/v1",
13607
+ reasoning: false,
13608
+ input: ["text", "image"],
13609
+ cost: {
13610
+ input: 0.075,
13611
+ output: 0.075,
13612
+ cacheRead: 0.0075,
13613
+ cacheWrite: 0,
13614
+ },
13615
+ contextWindow: 262144,
13616
+ maxTokens: 209715,
13617
+ },
12832
13618
  "mistralai/mistral-large": {
12833
13619
  id: "mistralai/mistral-large",
12834
13620
  name: "Mistral Large",
@@ -12880,6 +13666,23 @@ export const MODELS = {
12880
13666
  contextWindow: 262144,
12881
13667
  maxTokens: 209715,
12882
13668
  },
13669
+ "mistralai/mistral-large-2512:batch": {
13670
+ id: "mistralai/mistral-large-2512:batch",
13671
+ name: "Mistral: Mistral Large 3 2512 (batch)",
13672
+ api: "openai-completions",
13673
+ provider: "openrouter",
13674
+ baseUrl: "https://openrouter.ai/api/v1",
13675
+ reasoning: false,
13676
+ input: ["text", "image"],
13677
+ cost: {
13678
+ input: 0.25,
13679
+ output: 0.75,
13680
+ cacheRead: 0.024999999999999998,
13681
+ cacheWrite: 0,
13682
+ },
13683
+ contextWindow: 262144,
13684
+ maxTokens: 209715,
13685
+ },
12883
13686
  "mistralai/mistral-medium-3": {
12884
13687
  id: "mistralai/mistral-medium-3",
12885
13688
  name: "Mistral: Mistral Medium 3",
@@ -12928,8 +13731,8 @@ export const MODELS = {
12928
13731
  cacheRead: 0,
12929
13732
  cacheWrite: 0,
12930
13733
  },
12931
- contextWindow: 32768,
12932
- maxTokens: 26214,
13734
+ contextWindow: 262144,
13735
+ maxTokens: 209715,
12933
13736
  },
12934
13737
  "mistralai/mistral-medium-3.1": {
12935
13738
  id: "mistralai/mistral-medium-3.1",
@@ -12948,6 +13751,23 @@ export const MODELS = {
12948
13751
  contextWindow: 131072,
12949
13752
  maxTokens: 104857,
12950
13753
  },
13754
+ "mistralai/mistral-medium-3.1:batch": {
13755
+ id: "mistralai/mistral-medium-3.1:batch",
13756
+ name: "Mistral: Mistral Medium 3.1 (batch)",
13757
+ api: "openai-completions",
13758
+ provider: "openrouter",
13759
+ baseUrl: "https://openrouter.ai/api/v1",
13760
+ reasoning: false,
13761
+ input: ["text", "image"],
13762
+ cost: {
13763
+ input: 0.19999999999999998,
13764
+ output: 1,
13765
+ cacheRead: 0.02,
13766
+ cacheWrite: 0,
13767
+ },
13768
+ contextWindow: 131072,
13769
+ maxTokens: 104857,
13770
+ },
12951
13771
  "mistralai/mistral-nemo": {
12952
13772
  id: "mistralai/mistral-nemo",
12953
13773
  name: "Mistral: Mistral Nemo",
@@ -12999,6 +13819,23 @@ export const MODELS = {
12999
13819
  contextWindow: 262144,
13000
13820
  maxTokens: 209715,
13001
13821
  },
13822
+ "mistralai/mistral-small-2603:batch": {
13823
+ id: "mistralai/mistral-small-2603:batch",
13824
+ name: "Mistral: Mistral Small 4 (batch)",
13825
+ api: "openai-completions",
13826
+ provider: "openrouter",
13827
+ baseUrl: "https://openrouter.ai/api/v1",
13828
+ reasoning: true,
13829
+ input: ["text", "image"],
13830
+ cost: {
13831
+ input: 0.075,
13832
+ output: 0.3,
13833
+ cacheRead: 0.0075,
13834
+ cacheWrite: 0,
13835
+ },
13836
+ contextWindow: 262144,
13837
+ maxTokens: 209715,
13838
+ },
13002
13839
  "mistralai/mistral-small-3.2-24b-instruct": {
13003
13840
  id: "mistralai/mistral-small-3.2-24b-instruct",
13004
13841
  name: "Mistral: Mistral Small 3.2 24B",
@@ -13008,12 +13845,12 @@ export const MODELS = {
13008
13845
  reasoning: false,
13009
13846
  input: ["text", "image"],
13010
13847
  cost: {
13011
- input: 0.075,
13012
- output: 0.19999999999999998,
13848
+ input: 0.09375,
13849
+ output: 0.25,
13013
13850
  cacheRead: 0,
13014
13851
  cacheWrite: 0,
13015
13852
  },
13016
- contextWindow: 131072,
13853
+ contextWindow: 256000,
13017
13854
  maxTokens: 16384,
13018
13855
  },
13019
13856
  "mistralai/mixtral-8x22b-instruct": {
@@ -13065,7 +13902,7 @@ export const MODELS = {
13065
13902
  cacheWrite: 0,
13066
13903
  },
13067
13904
  contextWindow: 131072,
13068
- maxTokens: 100352,
13905
+ maxTokens: 98304,
13069
13906
  },
13070
13907
  "moonshotai/kimi-k2-0905": {
13071
13908
  id: "moonshotai/kimi-k2-0905",
@@ -13082,7 +13919,7 @@ export const MODELS = {
13082
13919
  cacheWrite: 0,
13083
13920
  },
13084
13921
  contextWindow: 262144,
13085
- maxTokens: 100352,
13922
+ maxTokens: 98304,
13086
13923
  },
13087
13924
  "moonshotai/kimi-k2-thinking": {
13088
13925
  id: "moonshotai/kimi-k2-thinking",
@@ -13099,7 +13936,7 @@ export const MODELS = {
13099
13936
  cacheWrite: 0,
13100
13937
  },
13101
13938
  contextWindow: 262144,
13102
- maxTokens: 100352,
13939
+ maxTokens: 98304,
13103
13940
  },
13104
13941
  "moonshotai/kimi-k2.5": {
13105
13942
  id: "moonshotai/kimi-k2.5",
@@ -13144,8 +13981,8 @@ export const MODELS = {
13144
13981
  reasoning: true,
13145
13982
  input: ["text", "image"],
13146
13983
  cost: {
13147
- input: 0.66,
13148
- output: 3.4,
13984
+ input: 0.7062,
13985
+ output: 3.21,
13149
13986
  cacheRead: 0.18,
13150
13987
  cacheWrite: 0,
13151
13988
  },
@@ -13161,9 +13998,9 @@ export const MODELS = {
13161
13998
  reasoning: true,
13162
13999
  input: ["text", "image"],
13163
14000
  cost: {
13164
- input: 3,
13165
- output: 15,
13166
- cacheRead: 0.3,
14001
+ input: 2.6481380629999998,
14002
+ output: 13.282724250000001,
14003
+ cacheRead: 0.30264435,
13167
14004
  cacheWrite: 0,
13168
14005
  },
13169
14006
  contextWindow: 1048576,
@@ -13186,35 +14023,35 @@ export const MODELS = {
13186
14023
  contextWindow: 1048576,
13187
14024
  maxTokens: 943718,
13188
14025
  },
13189
- "nex-agi/nex-n2-mini": {
13190
- id: "nex-agi/nex-n2-mini",
13191
- name: "Nex AGI: Nex-N2-Mini",
14026
+ "nex-agi/nex-n2.5-mini:free": {
14027
+ id: "nex-agi/nex-n2.5-mini:free",
14028
+ name: "Nex AGI: Nex-N2.5-Mini (free)",
13192
14029
  api: "openai-completions",
13193
14030
  provider: "openrouter",
13194
14031
  baseUrl: "https://openrouter.ai/api/v1",
13195
14032
  reasoning: true,
13196
14033
  input: ["text", "image"],
13197
14034
  cost: {
13198
- input: 0.024999999999999998,
13199
- output: 0.09999999999999999,
13200
- cacheRead: 0.0025,
14035
+ input: 0,
14036
+ output: 0,
14037
+ cacheRead: 0,
13201
14038
  cacheWrite: 0,
13202
14039
  },
13203
14040
  contextWindow: 262144,
13204
14041
  maxTokens: 235929,
13205
14042
  },
13206
- "nex-agi/nex-n2-pro": {
13207
- id: "nex-agi/nex-n2-pro",
13208
- name: "Nex AGI: Nex-N2-Pro",
14043
+ "nex-agi/nex-n2.5-pro:free": {
14044
+ id: "nex-agi/nex-n2.5-pro:free",
14045
+ name: "Nex AGI: Nex-N2.5-Pro (free)",
13209
14046
  api: "openai-completions",
13210
14047
  provider: "openrouter",
13211
14048
  baseUrl: "https://openrouter.ai/api/v1",
13212
14049
  reasoning: true,
13213
14050
  input: ["text", "image"],
13214
14051
  cost: {
13215
- input: 0.25,
13216
- output: 1,
13217
- cacheRead: 0.024999999999999998,
14052
+ input: 0,
14053
+ output: 0,
14054
+ cacheRead: 0,
13218
14055
  cacheWrite: 0,
13219
14056
  },
13220
14057
  contextWindow: 262144,
@@ -13263,13 +14100,13 @@ export const MODELS = {
13263
14100
  reasoning: true,
13264
14101
  input: ["text"],
13265
14102
  cost: {
13266
- input: 0.08499999999999999,
13267
- output: 0.39999999999999997,
14103
+ input: 0.08,
14104
+ output: 0.44999999999999996,
13268
14105
  cacheRead: 0,
13269
14106
  cacheWrite: 0,
13270
14107
  },
13271
- contextWindow: 1000000,
13272
- maxTokens: 16384,
14108
+ contextWindow: 262144,
14109
+ maxTokens: 235929,
13273
14110
  },
13274
14111
  "nvidia/nemotron-3-super-120b-a12b:free": {
13275
14112
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -14982,13 +15819,13 @@ export const MODELS = {
14982
15819
  reasoning: false,
14983
15820
  input: ["text"],
14984
15821
  cost: {
14985
- input: 0.09,
14986
- output: 0.55,
14987
- cacheRead: 0,
15822
+ input: 0.0875,
15823
+ output: 0.35,
15824
+ cacheRead: 0.0175,
14988
15825
  cacheWrite: 0,
14989
15826
  },
14990
15827
  contextWindow: 262144,
14991
- maxTokens: 16384,
15828
+ maxTokens: 235929,
14992
15829
  },
14993
15830
  "qwen/qwen3-235b-a22b-thinking-2507": {
14994
15831
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -15220,13 +16057,13 @@ export const MODELS = {
15220
16057
  reasoning: false,
15221
16058
  input: ["text"],
15222
16059
  cost: {
15223
- input: 0.09999999999999999,
16060
+ input: 0.09,
15224
16061
  output: 1.1,
15225
- cacheRead: 0.07,
16062
+ cacheRead: 0,
15226
16063
  cacheWrite: 0,
15227
16064
  },
15228
16065
  contextWindow: 262144,
15229
- maxTokens: 235929,
16066
+ maxTokens: 16384,
15230
16067
  },
15231
16068
  "qwen/qwen3-next-80b-a3b-thinking": {
15232
16069
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -15288,13 +16125,13 @@ export const MODELS = {
15288
16125
  reasoning: false,
15289
16126
  input: ["text", "image"],
15290
16127
  cost: {
15291
- input: 0.15,
15292
- output: 0.6,
16128
+ input: 0.13,
16129
+ output: 0.52,
15293
16130
  cacheRead: 0,
15294
16131
  cacheWrite: 0,
15295
16132
  },
15296
16133
  contextWindow: 262144,
15297
- maxTokens: 16384,
16134
+ maxTokens: 32768,
15298
16135
  },
15299
16136
  "qwen/qwen3-vl-30b-a3b-thinking": {
15300
16137
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -15373,13 +16210,13 @@ export const MODELS = {
15373
16210
  reasoning: true,
15374
16211
  input: ["text", "image"],
15375
16212
  cost: {
15376
- input: 0.29,
15377
- output: 2.4,
16213
+ input: 0.26,
16214
+ output: 2.08,
15378
16215
  cacheRead: 0,
15379
16216
  cacheWrite: 0,
15380
16217
  },
15381
16218
  contextWindow: 262144,
15382
- maxTokens: 81920,
16219
+ maxTokens: 65536,
15383
16220
  },
15384
16221
  "qwen/qwen3.5-27b": {
15385
16222
  id: "qwen/qwen3.5-27b",
@@ -15407,13 +16244,13 @@ export const MODELS = {
15407
16244
  reasoning: true,
15408
16245
  input: ["text", "image"],
15409
16246
  cost: {
15410
- input: 0.3125,
15411
- output: 1.25,
15412
- cacheRead: 0.15625,
16247
+ input: 0.1625,
16248
+ output: 1.3,
16249
+ cacheRead: 0,
15413
16250
  cacheWrite: 0,
15414
16251
  },
15415
16252
  contextWindow: 262144,
15416
- maxTokens: 16384,
16253
+ maxTokens: 65536,
15417
16254
  },
15418
16255
  "qwen/qwen3.5-397b-a17b": {
15419
16256
  id: "qwen/qwen3.5-397b-a17b",
@@ -15668,7 +16505,7 @@ export const MODELS = {
15668
16505
  cacheWrite: 0,
15669
16506
  },
15670
16507
  contextWindow: 1048576,
15671
- maxTokens: 262144,
16508
+ maxTokens: 131072,
15672
16509
  },
15673
16510
  "qwen/qwen3.8-2.4t-a95b:batch": {
15674
16511
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15696,9 +16533,9 @@ export const MODELS = {
15696
16533
  reasoning: true,
15697
16534
  input: ["text", "image"],
15698
16535
  cost: {
15699
- input: 0.42,
15700
- output: 3,
15701
- cacheRead: 0.08499999999999999,
16536
+ input: 0.21400000000000002,
16537
+ output: 2.5500000000000003,
16538
+ cacheRead: 0.15,
15702
16539
  cacheWrite: 0,
15703
16540
  },
15704
16541
  contextWindow: 1000000,
@@ -15772,6 +16609,23 @@ export const MODELS = {
15772
16609
  contextWindow: 256000,
15773
16610
  maxTokens: 128000,
15774
16611
  },
16612
+ "sakana/fugu-max": {
16613
+ id: "sakana/fugu-max",
16614
+ name: "Sakana: Fugu Max",
16615
+ api: "openai-completions",
16616
+ provider: "openrouter",
16617
+ baseUrl: "https://openrouter.ai/api/v1",
16618
+ reasoning: true,
16619
+ input: ["text", "image"],
16620
+ cost: {
16621
+ input: 2,
16622
+ output: 6,
16623
+ cacheRead: 0.25,
16624
+ cacheWrite: 0,
16625
+ },
16626
+ contextWindow: 1000000,
16627
+ maxTokens: 128000,
16628
+ },
15775
16629
  "sakana/fugu-ultra": {
15776
16630
  id: "sakana/fugu-ultra",
15777
16631
  name: "Sakana: Fugu Ultra",
@@ -15789,6 +16643,23 @@ export const MODELS = {
15789
16643
  contextWindow: 1000000,
15790
16644
  maxTokens: 128000,
15791
16645
  },
16646
+ "sakana/fugu-ultra-v2": {
16647
+ id: "sakana/fugu-ultra-v2",
16648
+ name: "Sakana: Fugu Ultra v2",
16649
+ api: "openai-completions",
16650
+ provider: "openrouter",
16651
+ baseUrl: "https://openrouter.ai/api/v1",
16652
+ reasoning: true,
16653
+ input: ["text", "image"],
16654
+ cost: {
16655
+ input: 5,
16656
+ output: 30,
16657
+ cacheRead: 0.5,
16658
+ cacheWrite: 0,
16659
+ },
16660
+ contextWindow: 1000000,
16661
+ maxTokens: 128000,
16662
+ },
15792
16663
  "sakana/sakana-namazu": {
15793
16664
  id: "sakana/sakana-namazu",
15794
16665
  name: "Sakana: Sakana Namazu",
@@ -15866,9 +16737,9 @@ export const MODELS = {
15866
16737
  reasoning: true,
15867
16738
  input: ["text"],
15868
16739
  cost: {
15869
- input: 0.0825,
15870
- output: 0.33,
15871
- cacheRead: 0.020625,
16740
+ input: 0.13199999999999998,
16741
+ output: 0.5279999999999999,
16742
+ cacheRead: 0.032999999999999995,
15872
16743
  cacheWrite: 0,
15873
16744
  },
15874
16745
  contextWindow: 262144,
@@ -15906,7 +16777,7 @@ export const MODELS = {
15906
16777
  cacheWrite: 0,
15907
16778
  },
15908
16779
  contextWindow: 1024000,
15909
- maxTokens: 26214,
16780
+ maxTokens: 819200,
15910
16781
  },
15911
16782
  "thinkingmachines/inkling": {
15912
16783
  id: "thinkingmachines/inkling",
@@ -15942,23 +16813,6 @@ export const MODELS = {
15942
16813
  contextWindow: 1048576,
15943
16814
  maxTokens: 262144,
15944
16815
  },
15945
- "thinkingmachines/inkling-small:batch": {
15946
- id: "thinkingmachines/inkling-small:batch",
15947
- name: "Thinking Machines: Inkling Small (batch)",
15948
- api: "openai-completions",
15949
- provider: "openrouter",
15950
- baseUrl: "https://openrouter.ai/api/v1",
15951
- reasoning: true,
15952
- input: ["text", "image"],
15953
- cost: {
15954
- input: 0.5,
15955
- output: 1.2,
15956
- cacheRead: 0.09999999999999999,
15957
- cacheWrite: 0,
15958
- },
15959
- contextWindow: 524288,
15960
- maxTokens: 471859,
15961
- },
15962
16816
  "thinkingmachines/inkling-small:free": {
15963
16817
  id: "thinkingmachines/inkling-small:free",
15964
16818
  name: "Thinking Machines: Inkling Small (free)",
@@ -16036,9 +16890,9 @@ export const MODELS = {
16036
16890
  reasoning: true,
16037
16891
  input: ["text"],
16038
16892
  cost: {
16039
- input: 0.03,
16040
- output: 0.12,
16041
- cacheRead: 0.006,
16893
+ input: 0.09,
16894
+ output: 0.36,
16895
+ cacheRead: 0.018,
16042
16896
  cacheWrite: 0,
16043
16897
  },
16044
16898
  contextWindow: 524288,
@@ -16240,13 +17094,13 @@ export const MODELS = {
16240
17094
  reasoning: true,
16241
17095
  input: ["text"],
16242
17096
  cost: {
16243
- input: 0.55,
16244
- output: 2.2,
16245
- cacheRead: 0.11,
17097
+ input: 0.43,
17098
+ output: 1.75,
17099
+ cacheRead: 0.08,
16246
17100
  cacheWrite: 0,
16247
17101
  },
16248
17102
  contextWindow: 204800,
16249
- maxTokens: 131072,
17103
+ maxTokens: 16384,
16250
17104
  },
16251
17105
  "z-ai/glm-4.6v": {
16252
17106
  id: "z-ai/glm-4.6v",
@@ -16291,13 +17145,13 @@ export const MODELS = {
16291
17145
  reasoning: true,
16292
17146
  input: ["text"],
16293
17147
  cost: {
16294
- input: 0.06,
17148
+ input: 0.060500000000000005,
16295
17149
  output: 0.39999999999999997,
16296
- cacheRead: 0.01,
17150
+ cacheRead: 0,
16297
17151
  cacheWrite: 0,
16298
17152
  },
16299
- contextWindow: 202752,
16300
- maxTokens: 16384,
17153
+ contextWindow: 200000,
17154
+ maxTokens: 117964,
16301
17155
  },
16302
17156
  "z-ai/glm-5": {
16303
17157
  id: "z-ai/glm-5",
@@ -16359,30 +17213,30 @@ export const MODELS = {
16359
17213
  reasoning: true,
16360
17214
  input: ["text"],
16361
17215
  cost: {
16362
- input: 0.966,
16363
- output: 3.036,
16364
- cacheRead: 0.1932,
17216
+ input: 1.4,
17217
+ output: 4.4,
17218
+ cacheRead: 0.14,
16365
17219
  cacheWrite: 0,
16366
17220
  },
16367
17221
  contextWindow: 1048576,
16368
- maxTokens: 131072,
17222
+ maxTokens: 128000,
16369
17223
  },
16370
- "z-ai/glm-5.2:free": {
16371
- id: "z-ai/glm-5.2:free",
16372
- name: "Z.ai: GLM 5.2 (free)",
17224
+ "z-ai/glm-5.2:batch": {
17225
+ id: "z-ai/glm-5.2:batch",
17226
+ name: "Z.ai: GLM 5.2 (batch)",
16373
17227
  api: "openai-completions",
16374
17228
  provider: "openrouter",
16375
17229
  baseUrl: "https://openrouter.ai/api/v1",
16376
17230
  reasoning: true,
16377
17231
  input: ["text"],
16378
17232
  cost: {
16379
- input: 0,
16380
- output: 0,
16381
- cacheRead: 0,
17233
+ input: 0.7,
17234
+ output: 2.2,
17235
+ cacheRead: 0.07,
16382
17236
  cacheWrite: 0,
16383
17237
  },
16384
- contextWindow: 256000,
16385
- maxTokens: 230400,
17238
+ contextWindow: 1048576,
17239
+ maxTokens: 943718,
16386
17240
  },
16387
17241
  "z-ai/glm-5.3": {
16388
17242
  id: "z-ai/glm-5.3",
@@ -16395,11 +17249,11 @@ export const MODELS = {
16395
17249
  cost: {
16396
17250
  input: 1.4,
16397
17251
  output: 4.4,
16398
- cacheRead: 0.14,
17252
+ cacheRead: 0.26,
16399
17253
  cacheWrite: 0,
16400
17254
  },
16401
17255
  contextWindow: 1310720,
16402
- maxTokens: 262144,
17256
+ maxTokens: 943717,
16403
17257
  },
16404
17258
  "z-ai/glm-5.3-flash": {
16405
17259
  id: "z-ai/glm-5.3-flash",
@@ -16410,9 +17264,9 @@ export const MODELS = {
16410
17264
  reasoning: true,
16411
17265
  input: ["text", "image"],
16412
17266
  cost: {
16413
- input: 0.075,
16414
- output: 0.25,
16415
- cacheRead: 0.015,
17267
+ input: 0.09999999999999999,
17268
+ output: 0.3333,
17269
+ cacheRead: 0.02,
16416
17270
  cacheWrite: 0,
16417
17271
  },
16418
17272
  contextWindow: 1310720,
@@ -16427,13 +17281,30 @@ export const MODELS = {
16427
17281
  reasoning: true,
16428
17282
  input: ["text", "image"],
16429
17283
  cost: {
16430
- input: 0.15,
16431
- output: 0.5,
16432
- cacheRead: 0.03,
17284
+ input: 0.075,
17285
+ output: 0.25,
17286
+ cacheRead: 0.015,
16433
17287
  cacheWrite: 0,
16434
17288
  },
16435
- contextWindow: 1048575,
16436
- maxTokens: 943717,
17289
+ contextWindow: 1048576,
17290
+ maxTokens: 943718,
17291
+ },
17292
+ "z-ai/glm-5.3:batch": {
17293
+ id: "z-ai/glm-5.3:batch",
17294
+ name: "Z.ai: GLM 5.3 (batch)",
17295
+ api: "openai-completions",
17296
+ provider: "openrouter",
17297
+ baseUrl: "https://openrouter.ai/api/v1",
17298
+ reasoning: true,
17299
+ input: ["text"],
17300
+ cost: {
17301
+ input: 0.7,
17302
+ output: 2.2,
17303
+ cacheRead: 0.13,
17304
+ cacheWrite: 0,
17305
+ },
17306
+ contextWindow: 1048576,
17307
+ maxTokens: 943718,
16437
17308
  },
16438
17309
  "z-ai/glm-5v-turbo": {
16439
17310
  id: "z-ai/glm-5v-turbo",
@@ -16472,7 +17343,7 @@ export const MODELS = {
16472
17343
  },
16473
17344
  "~anthropic/claude-haiku-latest": {
16474
17345
  id: "~anthropic/claude-haiku-latest",
16475
- name: "Anthropic Claude Haiku Latest",
17346
+ name: "Anthropic: Claude Haiku Latest",
16476
17347
  api: "openai-completions",
16477
17348
  provider: "openrouter",
16478
17349
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16506,7 +17377,7 @@ export const MODELS = {
16506
17377
  },
16507
17378
  "~anthropic/claude-sonnet-latest": {
16508
17379
  id: "~anthropic/claude-sonnet-latest",
16509
- name: "Anthropic Claude Sonnet Latest",
17380
+ name: "Anthropic: Claude Sonnet Latest",
16510
17381
  api: "openai-completions",
16511
17382
  provider: "openrouter",
16512
17383
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16521,9 +17392,43 @@ export const MODELS = {
16521
17392
  contextWindow: 1000000,
16522
17393
  maxTokens: 128000,
16523
17394
  },
17395
+ "~deepseek/deepseek-flash-latest": {
17396
+ id: "~deepseek/deepseek-flash-latest",
17397
+ name: "DeepSeek: DeepSeek Flash Latest",
17398
+ api: "openai-completions",
17399
+ provider: "openrouter",
17400
+ baseUrl: "https://openrouter.ai/api/v1",
17401
+ reasoning: true,
17402
+ input: ["text", "image"],
17403
+ cost: {
17404
+ input: 0.15,
17405
+ output: 0.6,
17406
+ cacheRead: 0.015,
17407
+ cacheWrite: 0,
17408
+ },
17409
+ contextWindow: 1048576,
17410
+ maxTokens: 943718,
17411
+ },
17412
+ "~deepseek/deepseek-pro-latest": {
17413
+ id: "~deepseek/deepseek-pro-latest",
17414
+ name: "DeepSeek: DeepSeek Pro Latest",
17415
+ api: "openai-completions",
17416
+ provider: "openrouter",
17417
+ baseUrl: "https://openrouter.ai/api/v1",
17418
+ reasoning: true,
17419
+ input: ["text"],
17420
+ cost: {
17421
+ input: 0.96,
17422
+ output: 2.88,
17423
+ cacheRead: 0.088,
17424
+ cacheWrite: 0,
17425
+ },
17426
+ contextWindow: 1048576,
17427
+ maxTokens: 943718,
17428
+ },
16524
17429
  "~deepseek/deepseek-v4-flash-latest": {
16525
17430
  id: "~deepseek/deepseek-v4-flash-latest",
16526
- name: "DeepSeek V4 Flash Latest",
17431
+ name: "DeepSeek: DeepSeek V4 Flash Latest",
16527
17432
  api: "openai-completions",
16528
17433
  provider: "openrouter",
16529
17434
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16532,17 +17437,17 @@ export const MODELS = {
16532
17437
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
16533
17438
  input: ["text"],
16534
17439
  cost: {
16535
- input: 0.049980000000000004,
16536
- output: 0.09996000000000001,
16537
- cacheRead: 0.009996000000000001,
17440
+ input: 0.04,
17441
+ output: 0.09999999999999999,
17442
+ cacheRead: 0.01,
16538
17443
  cacheWrite: 0,
16539
17444
  },
16540
17445
  contextWindow: 1310720,
16541
- maxTokens: 131072,
17446
+ maxTokens: 393216,
16542
17447
  },
16543
17448
  "~google/gemini-flash-latest": {
16544
17449
  id: "~google/gemini-flash-latest",
16545
- name: "Google Gemini Flash Latest",
17450
+ name: "Google: Gemini Flash Latest",
16546
17451
  api: "openai-completions",
16547
17452
  provider: "openrouter",
16548
17453
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16559,7 +17464,7 @@ export const MODELS = {
16559
17464
  },
16560
17465
  "~google/gemini-pro-latest": {
16561
17466
  id: "~google/gemini-pro-latest",
16562
- name: "Google Gemini Pro Latest",
17467
+ name: "Google: Gemini Pro Latest",
16563
17468
  api: "openai-completions",
16564
17469
  provider: "openrouter",
16565
17470
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16576,41 +17481,58 @@ export const MODELS = {
16576
17481
  },
16577
17482
  "~moonshotai/kimi-latest": {
16578
17483
  id: "~moonshotai/kimi-latest",
16579
- name: "MoonshotAI Kimi Latest",
17484
+ name: "MoonshotAI: Kimi Latest",
16580
17485
  api: "openai-completions",
16581
17486
  provider: "openrouter",
16582
17487
  baseUrl: "https://openrouter.ai/api/v1",
16583
17488
  reasoning: true,
16584
17489
  input: ["text", "image"],
16585
17490
  cost: {
16586
- input: 2.5500000000000003,
16587
- output: 12.75,
16588
- cacheRead: 0.25599998999999996,
17491
+ input: 2,
17492
+ output: 11.2,
17493
+ cacheRead: 0.232,
16589
17494
  cacheWrite: 0,
16590
17495
  },
16591
17496
  contextWindow: 1048576,
16592
17497
  maxTokens: 943718,
16593
17498
  },
16594
- "~openai/gpt-latest": {
16595
- id: "~openai/gpt-latest",
16596
- name: "OpenAI GPT Latest",
17499
+ "~openai/gpt-astra-latest": {
17500
+ id: "~openai/gpt-astra-latest",
17501
+ name: "OpenAI: GPT Astra Latest",
16597
17502
  api: "openai-completions",
16598
17503
  provider: "openrouter",
16599
17504
  baseUrl: "https://openrouter.ai/api/v1",
16600
17505
  reasoning: true,
16601
17506
  input: ["text", "image"],
16602
17507
  cost: {
16603
- input: 2,
16604
- output: 10,
16605
- cacheRead: 0.19999999999999998,
16606
- cacheWrite: 2.5,
17508
+ input: 10,
17509
+ output: 50,
17510
+ cacheRead: 1,
17511
+ cacheWrite: 12.5,
17512
+ },
17513
+ contextWindow: 1050000,
17514
+ maxTokens: 128000,
17515
+ },
17516
+ "~openai/gpt-luna-latest": {
17517
+ id: "~openai/gpt-luna-latest",
17518
+ name: "OpenAI: GPT Luna Latest",
17519
+ api: "openai-completions",
17520
+ provider: "openrouter",
17521
+ baseUrl: "https://openrouter.ai/api/v1",
17522
+ reasoning: true,
17523
+ input: ["text", "image"],
17524
+ cost: {
17525
+ input: 0.19999999999999998,
17526
+ output: 1.2,
17527
+ cacheRead: 0.02,
17528
+ cacheWrite: 0.25,
16607
17529
  },
16608
17530
  contextWindow: 1050000,
16609
17531
  maxTokens: 128000,
16610
17532
  },
16611
17533
  "~openai/gpt-mini-latest": {
16612
17534
  id: "~openai/gpt-mini-latest",
16613
- name: "OpenAI GPT Mini Latest",
17535
+ name: "OpenAI: GPT Mini Latest",
16614
17536
  api: "openai-completions",
16615
17537
  provider: "openrouter",
16616
17538
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16625,6 +17547,40 @@ export const MODELS = {
16625
17547
  contextWindow: 400000,
16626
17548
  maxTokens: 128000,
16627
17549
  },
17550
+ "~openai/gpt-sol-latest": {
17551
+ id: "~openai/gpt-sol-latest",
17552
+ name: "OpenAI: GPT Sol Latest",
17553
+ api: "openai-completions",
17554
+ provider: "openrouter",
17555
+ baseUrl: "https://openrouter.ai/api/v1",
17556
+ reasoning: true,
17557
+ input: ["text", "image"],
17558
+ cost: {
17559
+ input: 2,
17560
+ output: 10,
17561
+ cacheRead: 0.19999999999999998,
17562
+ cacheWrite: 2.5,
17563
+ },
17564
+ contextWindow: 1050000,
17565
+ maxTokens: 128000,
17566
+ },
17567
+ "~openai/gpt-terra-latest": {
17568
+ id: "~openai/gpt-terra-latest",
17569
+ name: "OpenAI: GPT Terra Latest",
17570
+ api: "openai-completions",
17571
+ provider: "openrouter",
17572
+ baseUrl: "https://openrouter.ai/api/v1",
17573
+ reasoning: true,
17574
+ input: ["text", "image"],
17575
+ cost: {
17576
+ input: 2,
17577
+ output: 12,
17578
+ cacheRead: 0.19999999999999998,
17579
+ cacheWrite: 2.5,
17580
+ },
17581
+ contextWindow: 1050000,
17582
+ maxTokens: 128000,
17583
+ },
16628
17584
  "~x-ai/grok-latest": {
16629
17585
  id: "~x-ai/grok-latest",
16630
17586
  name: "xAI: Grok Latest",
@@ -16657,7 +17613,7 @@ export const MODELS = {
16657
17613
  cacheWrite: 0,
16658
17614
  },
16659
17615
  contextWindow: 1310720,
16660
- maxTokens: 943718,
17616
+ maxTokens: 131072,
16661
17617
  },
16662
17618
  "~z-ai/glm-latest": {
16663
17619
  id: "~z-ai/glm-latest",
@@ -16668,9 +17624,9 @@ export const MODELS = {
16668
17624
  reasoning: true,
16669
17625
  input: ["text"],
16670
17626
  cost: {
16671
- input: 1.15,
16672
- output: 3.5,
16673
- cacheRead: 0.09999999999999999,
17627
+ input: 0.8775,
17628
+ output: 2.9699999999999998,
17629
+ cacheRead: 0.1755,
16674
17630
  cacheWrite: 0,
16675
17631
  },
16676
17632
  contextWindow: 1310720,
@@ -16847,6 +17803,25 @@ export const MODELS = {
16847
17803
  contextWindow: 1048576,
16848
17804
  maxTokens: 384000,
16849
17805
  },
17806
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
17807
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
17808
+ name: "DeepSeek V4.1 Flash",
17809
+ api: "openai-completions",
17810
+ provider: "together",
17811
+ baseUrl: "https://api.together.ai/v1",
17812
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
17813
+ reasoning: true,
17814
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
17815
+ input: ["text", "image"],
17816
+ cost: {
17817
+ input: 0.3,
17818
+ output: 1.2,
17819
+ cacheRead: 0.006,
17820
+ cacheWrite: 0,
17821
+ },
17822
+ contextWindow: 1048576,
17823
+ maxTokens: 384000,
17824
+ },
16850
17825
  "google/gemma-4-31B-it": {
16851
17826
  id: "google/gemma-4-31B-it",
16852
17827
  name: "Gemma 4 31B Instruct",
@@ -17155,7 +18130,7 @@ export const MODELS = {
17155
18130
  cost: {
17156
18131
  input: 1.3,
17157
18132
  output: 7.8,
17158
- cacheRead: 0.26,
18133
+ cacheRead: 0.13,
17159
18134
  cacheWrite: 1.625,
17160
18135
  },
17161
18136
  contextWindow: 240000,
@@ -17393,7 +18368,7 @@ export const MODELS = {
17393
18368
  cost: {
17394
18369
  input: 0.09999999999999999,
17395
18370
  output: 0.39999999999999997,
17396
- cacheRead: 0.001,
18371
+ cacheRead: 0.01,
17397
18372
  cacheWrite: 0.125,
17398
18373
  },
17399
18374
  contextWindow: 1000000,
@@ -17409,7 +18384,7 @@ export const MODELS = {
17409
18384
  input: ["text", "image"],
17410
18385
  cost: {
17411
18386
  input: 0.39999999999999997,
17412
- output: 2.4,
18387
+ output: 2.5,
17413
18388
  cacheRead: 0.04,
17414
18389
  cacheWrite: 0.5,
17415
18390
  },
@@ -17444,7 +18419,7 @@ export const MODELS = {
17444
18419
  cost: {
17445
18420
  input: 0.5,
17446
18421
  output: 3,
17447
- cacheRead: 0.09999999999999999,
18422
+ cacheRead: 0.049999999999999996,
17448
18423
  cacheWrite: 0.625,
17449
18424
  },
17450
18425
  contextWindow: 1000000,
@@ -17544,7 +18519,7 @@ export const MODELS = {
17544
18519
  reasoning: true,
17545
18520
  input: ["text", "image"],
17546
18521
  cost: {
17547
- input: 0.16,
18522
+ input: 0.15,
17548
18523
  output: 0.47,
17549
18524
  cacheRead: 0.016,
17550
18525
  cacheWrite: 0.19999999999999998,
@@ -17552,23 +18527,6 @@ export const MODELS = {
17552
18527
  contextWindow: 991000,
17553
18528
  maxTokens: 128000,
17554
18529
  },
17555
- "alibaba/qwen3.8-flash-next": {
17556
- id: "alibaba/qwen3.8-flash-next",
17557
- name: "Qwen 3.8 Flash Next",
17558
- api: "anthropic-messages",
17559
- provider: "vercel-ai-gateway",
17560
- baseUrl: "https://ai-gateway.vercel.sh",
17561
- reasoning: true,
17562
- input: ["text", "image"],
17563
- cost: {
17564
- input: 0.12,
17565
- output: 0.39999999999999997,
17566
- cacheRead: 0.01,
17567
- cacheWrite: 0,
17568
- },
17569
- contextWindow: 1048576,
17570
- maxTokens: 1048576,
17571
- },
17572
18530
  "alibaba/qwen3.8-max": {
17573
18531
  id: "alibaba/qwen3.8-max",
17574
18532
  name: "Qwen 3.8 Max",
@@ -18000,6 +18958,23 @@ export const MODELS = {
18000
18958
  contextWindow: 256000,
18001
18959
  maxTokens: 64000,
18002
18960
  },
18961
+ "bytedance/seed-2.1-turbo": {
18962
+ id: "bytedance/seed-2.1-turbo",
18963
+ name: "Seed 2.1 Turbo",
18964
+ api: "anthropic-messages",
18965
+ provider: "vercel-ai-gateway",
18966
+ baseUrl: "https://ai-gateway.vercel.sh",
18967
+ reasoning: true,
18968
+ input: ["text", "image"],
18969
+ cost: {
18970
+ input: 0.5,
18971
+ output: 2.5,
18972
+ cacheRead: 0.09999999999999999,
18973
+ cacheWrite: 0,
18974
+ },
18975
+ contextWindow: 262144,
18976
+ maxTokens: 262144,
18977
+ },
18003
18978
  "cohere/command-a": {
18004
18979
  id: "cohere/command-a",
18005
18980
  name: "Command A",
@@ -18077,9 +19052,9 @@ export const MODELS = {
18077
19052
  reasoning: false,
18078
19053
  input: ["text"],
18079
19054
  cost: {
18080
- input: 0.28,
18081
- output: 0.42,
18082
- cacheRead: 0.028,
19055
+ input: 0.62,
19056
+ output: 1.85,
19057
+ cacheRead: 0,
18083
19058
  cacheWrite: 0,
18084
19059
  },
18085
19060
  contextWindow: 128000,
@@ -18187,6 +19162,23 @@ export const MODELS = {
18187
19162
  contextWindow: 1000000,
18188
19163
  maxTokens: 384000,
18189
19164
  },
19165
+ "deepseek/deepseek-v4.1-flash": {
19166
+ id: "deepseek/deepseek-v4.1-flash",
19167
+ name: "DeepSeek V4.1 Flash",
19168
+ api: "anthropic-messages",
19169
+ provider: "vercel-ai-gateway",
19170
+ baseUrl: "https://ai-gateway.vercel.sh",
19171
+ reasoning: true,
19172
+ input: ["text", "image"],
19173
+ cost: {
19174
+ input: 0.15,
19175
+ output: 0.6,
19176
+ cacheRead: 0.015,
19177
+ cacheWrite: 0,
19178
+ },
19179
+ contextWindow: 1000000,
19180
+ maxTokens: 384000,
19181
+ },
18190
19182
  "google/gemini-2.5-flash": {
18191
19183
  id: "google/gemini-2.5-flash",
18192
19184
  name: "Gemini 2.5 Flash",
@@ -18425,6 +19417,23 @@ export const MODELS = {
18425
19417
  contextWindow: 128000,
18426
19418
  maxTokens: 128000,
18427
19419
  },
19420
+ "inception/mercury-2.5": {
19421
+ id: "inception/mercury-2.5",
19422
+ name: "Mercury 2.5",
19423
+ api: "anthropic-messages",
19424
+ provider: "vercel-ai-gateway",
19425
+ baseUrl: "https://ai-gateway.vercel.sh",
19426
+ reasoning: true,
19427
+ input: ["text"],
19428
+ cost: {
19429
+ input: 0.04,
19430
+ output: 0.15,
19431
+ cacheRead: 0.004,
19432
+ cacheWrite: 0,
19433
+ },
19434
+ contextWindow: 260000,
19435
+ maxTokens: 65536,
19436
+ },
18428
19437
  "inception/mercury-coder-small": {
18429
19438
  id: "inception/mercury-coder-small",
18430
19439
  name: "Mercury Coder Small Beta",
@@ -18451,9 +19460,9 @@ export const MODELS = {
18451
19460
  reasoning: true,
18452
19461
  input: ["text"],
18453
19462
  cost: {
18454
- input: 0.06,
18455
- output: 0.18,
18456
- cacheRead: 0.012,
19463
+ input: 0.020999999999999998,
19464
+ output: 0.063,
19465
+ cacheRead: 0.004200000000000001,
18457
19466
  cacheWrite: 0,
18458
19467
  },
18459
19468
  contextWindow: 256000,
@@ -18527,6 +19536,40 @@ export const MODELS = {
18527
19536
  contextWindow: 256000,
18528
19537
  maxTokens: 32000,
18529
19538
  },
19539
+ "inclusionai/ling-3.0-flash-vl": {
19540
+ id: "inclusionai/ling-3.0-flash-vl",
19541
+ name: "Ling 3.0 Flash VL",
19542
+ api: "anthropic-messages",
19543
+ provider: "vercel-ai-gateway",
19544
+ baseUrl: "https://ai-gateway.vercel.sh",
19545
+ reasoning: true,
19546
+ input: ["text", "image"],
19547
+ cost: {
19548
+ input: 0,
19549
+ output: 0,
19550
+ cacheRead: 0,
19551
+ cacheWrite: 0,
19552
+ },
19553
+ contextWindow: 256000,
19554
+ maxTokens: 32000,
19555
+ },
19556
+ "inclusionai/ling-3.0-flash-vl-free": {
19557
+ id: "inclusionai/ling-3.0-flash-vl-free",
19558
+ name: "Ling 3.0 Flash VL (Free)",
19559
+ api: "anthropic-messages",
19560
+ provider: "vercel-ai-gateway",
19561
+ baseUrl: "https://ai-gateway.vercel.sh",
19562
+ reasoning: true,
19563
+ input: ["text", "image"],
19564
+ cost: {
19565
+ input: 0,
19566
+ output: 0,
19567
+ cacheRead: 0,
19568
+ cacheWrite: 0,
19569
+ },
19570
+ contextWindow: 256000,
19571
+ maxTokens: 32000,
19572
+ },
18530
19573
  "interfaze/interfaze-beta": {
18531
19574
  id: "interfaze/interfaze-beta",
18532
19575
  name: "Interfaze Beta",
@@ -18901,23 +19944,6 @@ export const MODELS = {
18901
19944
  contextWindow: 204800,
18902
19945
  maxTokens: 131000,
18903
19946
  },
18904
- "minimax/minimax-m2.7-free": {
18905
- id: "minimax/minimax-m2.7-free",
18906
- name: "MiniMax M2.7 (Free)",
18907
- api: "anthropic-messages",
18908
- provider: "vercel-ai-gateway",
18909
- baseUrl: "https://ai-gateway.vercel.sh",
18910
- reasoning: true,
18911
- input: ["text"],
18912
- cost: {
18913
- input: 0,
18914
- output: 0,
18915
- cacheRead: 0,
18916
- cacheWrite: 0,
18917
- },
18918
- contextWindow: 196608,
18919
- maxTokens: 196608,
18920
- },
18921
19947
  "minimax/minimax-m2.7-highspeed": {
18922
19948
  id: "minimax/minimax-m2.7-highspeed",
18923
19949
  name: "MiniMax M2.7 High Speed",
@@ -18952,23 +19978,6 @@ export const MODELS = {
18952
19978
  contextWindow: 512000,
18953
19979
  maxTokens: 512000,
18954
19980
  },
18955
- "minimax/minimax-m3-free": {
18956
- id: "minimax/minimax-m3-free",
18957
- name: "MiniMax M3 (Free)",
18958
- api: "anthropic-messages",
18959
- provider: "vercel-ai-gateway",
18960
- baseUrl: "https://ai-gateway.vercel.sh",
18961
- reasoning: true,
18962
- input: ["text", "image"],
18963
- cost: {
18964
- input: 0,
18965
- output: 0,
18966
- cacheRead: 0,
18967
- cacheWrite: 0,
18968
- },
18969
- contextWindow: 1048576,
18970
- maxTokens: 1048576,
18971
- },
18972
19981
  "mistral/codestral": {
18973
19982
  id: "mistral/codestral",
18974
19983
  name: "Mistral Codestral",
@@ -18980,46 +19989,12 @@ export const MODELS = {
18980
19989
  cost: {
18981
19990
  input: 0.3,
18982
19991
  output: 0.8999999999999999,
18983
- cacheRead: 0,
19992
+ cacheRead: 0.03,
18984
19993
  cacheWrite: 0,
18985
19994
  },
18986
19995
  contextWindow: 128000,
18987
19996
  maxTokens: 4000,
18988
19997
  },
18989
- "mistral/devstral-2": {
18990
- id: "mistral/devstral-2",
18991
- name: "Devstral 2",
18992
- api: "anthropic-messages",
18993
- provider: "vercel-ai-gateway",
18994
- baseUrl: "https://ai-gateway.vercel.sh",
18995
- reasoning: false,
18996
- input: ["text"],
18997
- cost: {
18998
- input: 0.39999999999999997,
18999
- output: 2,
19000
- cacheRead: 0,
19001
- cacheWrite: 0,
19002
- },
19003
- contextWindow: 256000,
19004
- maxTokens: 256000,
19005
- },
19006
- "mistral/devstral-small-2": {
19007
- id: "mistral/devstral-small-2",
19008
- name: "Devstral Small 2",
19009
- api: "anthropic-messages",
19010
- provider: "vercel-ai-gateway",
19011
- baseUrl: "https://ai-gateway.vercel.sh",
19012
- reasoning: false,
19013
- input: ["text", "image"],
19014
- cost: {
19015
- input: 0.09999999999999999,
19016
- output: 0.3,
19017
- cacheRead: 0,
19018
- cacheWrite: 0,
19019
- },
19020
- contextWindow: 256000,
19021
- maxTokens: 256000,
19022
- },
19023
19998
  "mistral/ministral-14b": {
19024
19999
  id: "mistral/ministral-14b",
19025
20000
  name: "Ministral 14B",
@@ -19031,10 +20006,10 @@ export const MODELS = {
19031
20006
  cost: {
19032
20007
  input: 0.19999999999999998,
19033
20008
  output: 0.19999999999999998,
19034
- cacheRead: 0,
20009
+ cacheRead: 0.02,
19035
20010
  cacheWrite: 0,
19036
20011
  },
19037
- contextWindow: 256000,
20012
+ contextWindow: 262144,
19038
20013
  maxTokens: 256000,
19039
20014
  },
19040
20015
  "mistral/ministral-3b": {
@@ -19048,10 +20023,10 @@ export const MODELS = {
19048
20023
  cost: {
19049
20024
  input: 0.09999999999999999,
19050
20025
  output: 0.09999999999999999,
19051
- cacheRead: 0,
20026
+ cacheRead: 0.01,
19052
20027
  cacheWrite: 0,
19053
20028
  },
19054
- contextWindow: 128000,
20029
+ contextWindow: 131072,
19055
20030
  maxTokens: 4000,
19056
20031
  },
19057
20032
  "mistral/ministral-8b": {
@@ -19065,10 +20040,10 @@ export const MODELS = {
19065
20040
  cost: {
19066
20041
  input: 0.15,
19067
20042
  output: 0.15,
19068
- cacheRead: 0,
20043
+ cacheRead: 0.015,
19069
20044
  cacheWrite: 0,
19070
20045
  },
19071
- contextWindow: 128000,
20046
+ contextWindow: 262144,
19072
20047
  maxTokens: 4000,
19073
20048
  },
19074
20049
  "mistral/mistral-large-3": {
@@ -19082,29 +20057,12 @@ export const MODELS = {
19082
20057
  cost: {
19083
20058
  input: 0.5,
19084
20059
  output: 1.5,
19085
- cacheRead: 0,
20060
+ cacheRead: 0.049999999999999996,
19086
20061
  cacheWrite: 0,
19087
20062
  },
19088
- contextWindow: 256000,
20063
+ contextWindow: 262144,
19089
20064
  maxTokens: 256000,
19090
20065
  },
19091
- "mistral/mistral-medium": {
19092
- id: "mistral/mistral-medium",
19093
- name: "Mistral Medium 3.1",
19094
- api: "anthropic-messages",
19095
- provider: "vercel-ai-gateway",
19096
- baseUrl: "https://ai-gateway.vercel.sh",
19097
- reasoning: false,
19098
- input: ["text", "image"],
19099
- cost: {
19100
- input: 0.39999999999999997,
19101
- output: 2,
19102
- cacheRead: 0,
19103
- cacheWrite: 0,
19104
- },
19105
- contextWindow: 128000,
19106
- maxTokens: 64000,
19107
- },
19108
20066
  "mistral/mistral-medium-3.5": {
19109
20067
  id: "mistral/mistral-medium-3.5",
19110
20068
  name: "Mistral Medium Latest",
@@ -19116,10 +20074,10 @@ export const MODELS = {
19116
20074
  cost: {
19117
20075
  input: 1.5,
19118
20076
  output: 7.5,
19119
- cacheRead: 0,
20077
+ cacheRead: 0.15,
19120
20078
  cacheWrite: 0,
19121
20079
  },
19122
- contextWindow: 256000,
20080
+ contextWindow: 262144,
19123
20081
  maxTokens: 256000,
19124
20082
  },
19125
20083
  "mistral/mistral-nemo": {
@@ -19129,15 +20087,15 @@ export const MODELS = {
19129
20087
  provider: "vercel-ai-gateway",
19130
20088
  baseUrl: "https://ai-gateway.vercel.sh",
19131
20089
  reasoning: false,
19132
- input: ["text", "image"],
20090
+ input: ["text"],
19133
20091
  cost: {
19134
- input: 0.15,
19135
- output: 0.15,
20092
+ input: 0.04,
20093
+ output: 0.16999999999999998,
19136
20094
  cacheRead: 0,
19137
20095
  cacheWrite: 0,
19138
20096
  },
19139
- contextWindow: 128000,
19140
- maxTokens: 128000,
20097
+ contextWindow: 60288,
20098
+ maxTokens: 16000,
19141
20099
  },
19142
20100
  "mistral/mistral-small": {
19143
20101
  id: "mistral/mistral-small",
@@ -19147,30 +20105,13 @@ export const MODELS = {
19147
20105
  baseUrl: "https://ai-gateway.vercel.sh",
19148
20106
  reasoning: false,
19149
20107
  input: ["text", "image"],
19150
- cost: {
19151
- input: 0.09999999999999999,
19152
- output: 0.3,
19153
- cacheRead: 0,
19154
- cacheWrite: 0,
19155
- },
19156
- contextWindow: 32000,
19157
- maxTokens: 4000,
19158
- },
19159
- "mistral/pixtral-12b": {
19160
- id: "mistral/pixtral-12b",
19161
- name: "Pixtral 12B 2409",
19162
- api: "anthropic-messages",
19163
- provider: "vercel-ai-gateway",
19164
- baseUrl: "https://ai-gateway.vercel.sh",
19165
- reasoning: false,
19166
- input: ["text", "image"],
19167
20108
  cost: {
19168
20109
  input: 0.15,
19169
- output: 0.15,
19170
- cacheRead: 0,
20110
+ output: 0.6,
20111
+ cacheRead: 0.015,
19171
20112
  cacheWrite: 0,
19172
20113
  },
19173
- contextWindow: 128000,
20114
+ contextWindow: 262144,
19174
20115
  maxTokens: 4000,
19175
20116
  },
19176
20117
  "moonshotai/kimi-k2": {
@@ -19319,8 +20260,8 @@ export const MODELS = {
19319
20260
  input: ["text"],
19320
20261
  cost: {
19321
20262
  input: 0.049999999999999996,
19322
- output: 0.24,
19323
- cacheRead: 0,
20263
+ output: 0.19999999999999998,
20264
+ cacheRead: 0.024999999999999998,
19324
20265
  cacheWrite: 0,
19325
20266
  },
19326
20267
  contextWindow: 262144,
@@ -20429,6 +21370,23 @@ export const MODELS = {
20429
21370
  contextWindow: 256000,
20430
21371
  maxTokens: 32768,
20431
21372
  },
21373
+ "sakana/fugu-max": {
21374
+ id: "sakana/fugu-max",
21375
+ name: "Fugu Max",
21376
+ api: "anthropic-messages",
21377
+ provider: "vercel-ai-gateway",
21378
+ baseUrl: "https://ai-gateway.vercel.sh",
21379
+ reasoning: true,
21380
+ input: ["text", "image"],
21381
+ cost: {
21382
+ input: 2,
21383
+ output: 6,
21384
+ cacheRead: 0.25,
21385
+ cacheWrite: 0,
21386
+ },
21387
+ contextWindow: 1000000,
21388
+ maxTokens: 1000000,
21389
+ },
20432
21390
  "sakana/fugu-ultra": {
20433
21391
  id: "sakana/fugu-ultra",
20434
21392
  name: "Fugu Ultra",
@@ -20446,6 +21404,23 @@ export const MODELS = {
20446
21404
  contextWindow: 1000000,
20447
21405
  maxTokens: 1000000,
20448
21406
  },
21407
+ "sakana/fugu-ultra-v2": {
21408
+ id: "sakana/fugu-ultra-v2",
21409
+ name: "Fugu Ultra v2",
21410
+ api: "anthropic-messages",
21411
+ provider: "vercel-ai-gateway",
21412
+ baseUrl: "https://ai-gateway.vercel.sh",
21413
+ reasoning: true,
21414
+ input: ["text", "image"],
21415
+ cost: {
21416
+ input: 5,
21417
+ output: 30,
21418
+ cacheRead: 0.5,
21419
+ cacheWrite: 0,
21420
+ },
21421
+ contextWindow: 1000000,
21422
+ maxTokens: 1000000,
21423
+ },
20449
21424
  "sakana/namazu": {
20450
21425
  id: "sakana/namazu",
20451
21426
  name: "Sakana Namazu",
@@ -20803,23 +21778,6 @@ export const MODELS = {
20803
21778
  contextWindow: 1050000,
20804
21779
  maxTokens: 131000,
20805
21780
  },
20806
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
20807
- id: "xiaomi/mimo-v2.5-pro-ultraspeed",
20808
- name: "MiMo V2.5 Pro UltraSpeed",
20809
- api: "anthropic-messages",
20810
- provider: "vercel-ai-gateway",
20811
- baseUrl: "https://ai-gateway.vercel.sh",
20812
- reasoning: true,
20813
- input: ["text"],
20814
- cost: {
20815
- input: 1.305,
20816
- output: 2.61,
20817
- cacheRead: 0.0108,
20818
- cacheWrite: 0,
20819
- },
20820
- contextWindow: 1048576,
20821
- maxTokens: 131072,
20822
- },
20823
21781
  "zai/glm-4.5": {
20824
21782
  id: "zai/glm-4.5",
20825
21783
  name: "GLM 4.5",
@@ -21033,9 +21991,9 @@ export const MODELS = {
21033
21991
  reasoning: true,
21034
21992
  input: ["text"],
21035
21993
  cost: {
21036
- input: 0.7,
21037
- output: 2.2,
21038
- cacheRead: 0.13,
21994
+ input: 1.4,
21995
+ output: 4.4,
21996
+ cacheRead: 0.14,
21039
21997
  cacheWrite: 0,
21040
21998
  },
21041
21999
  contextWindow: 1000000,
@@ -21075,23 +22033,6 @@ export const MODELS = {
21075
22033
  contextWindow: 1000000,
21076
22034
  maxTokens: 131000,
21077
22035
  },
21078
- "zai/glm-5.3-promo-50": {
21079
- id: "zai/glm-5.3-promo-50",
21080
- name: "GLM 5.3 (50% off)",
21081
- api: "anthropic-messages",
21082
- provider: "vercel-ai-gateway",
21083
- baseUrl: "https://ai-gateway.vercel.sh",
21084
- reasoning: true,
21085
- input: ["text"],
21086
- cost: {
21087
- input: 0.7,
21088
- output: 2.2,
21089
- cacheRead: 0.13,
21090
- cacheWrite: 0,
21091
- },
21092
- contextWindow: 1048576,
21093
- maxTokens: 1048576,
21094
- },
21095
22036
  "zai/glm-5v-turbo": {
21096
22037
  id: "zai/glm-5v-turbo",
21097
22038
  name: "GLM 5V Turbo",