@fleetagent/pi-ai 0.2.2 → 0.2.4

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",
@@ -276,6 +276,57 @@ export const MODELS = {
276
276
  contextWindow: 1000000,
277
277
  maxTokens: 128000,
278
278
  },
279
+ "apac.amazon.nova-lite-v1:0": {
280
+ id: "apac.amazon.nova-lite-v1:0",
281
+ name: "Nova Lite (APAC)",
282
+ api: "bedrock-converse-stream",
283
+ provider: "amazon-bedrock",
284
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
285
+ reasoning: false,
286
+ input: ["text", "image"],
287
+ cost: {
288
+ input: 0.063,
289
+ output: 0.252,
290
+ cacheRead: 0.01575,
291
+ cacheWrite: 0.063,
292
+ },
293
+ contextWindow: 300000,
294
+ maxTokens: 10000,
295
+ },
296
+ "apac.amazon.nova-micro-v1:0": {
297
+ id: "apac.amazon.nova-micro-v1:0",
298
+ name: "Nova Micro (APAC)",
299
+ api: "bedrock-converse-stream",
300
+ provider: "amazon-bedrock",
301
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
302
+ reasoning: false,
303
+ input: ["text"],
304
+ cost: {
305
+ input: 0.037,
306
+ output: 0.148,
307
+ cacheRead: 0.00925,
308
+ cacheWrite: 0.037,
309
+ },
310
+ contextWindow: 128000,
311
+ maxTokens: 10000,
312
+ },
313
+ "apac.amazon.nova-pro-v1:0": {
314
+ id: "apac.amazon.nova-pro-v1:0",
315
+ name: "Nova Pro (APAC)",
316
+ api: "bedrock-converse-stream",
317
+ provider: "amazon-bedrock",
318
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
319
+ reasoning: false,
320
+ input: ["text", "image"],
321
+ cost: {
322
+ input: 0.84,
323
+ output: 3.36,
324
+ cacheRead: 0.21,
325
+ cacheWrite: 0.84,
326
+ },
327
+ contextWindow: 300000,
328
+ maxTokens: 10000,
329
+ },
279
330
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
280
331
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
281
332
  name: "Claude Haiku 4.5 (AU)",
@@ -311,6 +362,24 @@ export const MODELS = {
311
362
  contextWindow: 1000000,
312
363
  maxTokens: 128000,
313
364
  },
365
+ "au.anthropic.claude-opus-4-7": {
366
+ id: "au.anthropic.claude-opus-4-7",
367
+ name: "Claude Opus 4.7 (AU)",
368
+ api: "bedrock-converse-stream",
369
+ provider: "amazon-bedrock",
370
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
371
+ reasoning: true,
372
+ thinkingLevelMap: { "xhigh": "xhigh" },
373
+ input: ["text", "image"],
374
+ cost: {
375
+ input: 5.5,
376
+ output: 27.5,
377
+ cacheRead: 0.55,
378
+ cacheWrite: 6.875,
379
+ },
380
+ contextWindow: 1000000,
381
+ maxTokens: 128000,
382
+ },
314
383
  "au.anthropic.claude-opus-4-8": {
315
384
  id: "au.anthropic.claude-opus-4-8",
316
385
  name: "Claude Opus 4.8 (AU)",
@@ -396,22 +465,22 @@ export const MODELS = {
396
465
  contextWindow: 1000000,
397
466
  maxTokens: 128000,
398
467
  },
399
- "deepseek.r1-v1:0": {
400
- id: "deepseek.r1-v1:0",
401
- name: "DeepSeek-R1",
468
+ "ca.amazon.nova-lite-v1:0": {
469
+ id: "ca.amazon.nova-lite-v1:0",
470
+ name: "Nova Lite (CA)",
402
471
  api: "bedrock-converse-stream",
403
472
  provider: "amazon-bedrock",
404
473
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
405
- reasoning: true,
406
- input: ["text"],
474
+ reasoning: false,
475
+ input: ["text", "image"],
407
476
  cost: {
408
- input: 1.35,
409
- output: 5.4,
410
- cacheRead: 0,
411
- cacheWrite: 0,
477
+ input: 0.064,
478
+ output: 0.256,
479
+ cacheRead: 0.016,
480
+ cacheWrite: 0.064,
412
481
  },
413
- contextWindow: 128000,
414
- maxTokens: 32768,
482
+ contextWindow: 300000,
483
+ maxTokens: 10000,
415
484
  },
416
485
  "deepseek.v3-v1:0": {
417
486
  id: "deepseek.v3-v1:0",
@@ -432,7 +501,7 @@ export const MODELS = {
432
501
  },
433
502
  "deepseek.v3.2": {
434
503
  id: "deepseek.v3.2",
435
- name: "DeepSeek-V3.2",
504
+ name: "DeepSeek V3.2",
436
505
  api: "bedrock-converse-stream",
437
506
  provider: "amazon-bedrock",
438
507
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -447,6 +516,74 @@ export const MODELS = {
447
516
  contextWindow: 163840,
448
517
  maxTokens: 81920,
449
518
  },
519
+ "eu.amazon.nova-2-lite-v1:0": {
520
+ id: "eu.amazon.nova-2-lite-v1:0",
521
+ name: "Nova 2 Lite (EU)",
522
+ api: "bedrock-converse-stream",
523
+ provider: "amazon-bedrock",
524
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
525
+ reasoning: true,
526
+ input: ["text", "image"],
527
+ cost: {
528
+ input: 0.374,
529
+ output: 3.157,
530
+ cacheRead: 0.0935,
531
+ cacheWrite: 0.374,
532
+ },
533
+ contextWindow: 1000000,
534
+ maxTokens: 65535,
535
+ },
536
+ "eu.amazon.nova-lite-v1:0": {
537
+ id: "eu.amazon.nova-lite-v1:0",
538
+ name: "Nova Lite (EU)",
539
+ api: "bedrock-converse-stream",
540
+ provider: "amazon-bedrock",
541
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
542
+ reasoning: false,
543
+ input: ["text", "image"],
544
+ cost: {
545
+ input: 0.069,
546
+ output: 0.276,
547
+ cacheRead: 0.01725,
548
+ cacheWrite: 0.069,
549
+ },
550
+ contextWindow: 300000,
551
+ maxTokens: 10000,
552
+ },
553
+ "eu.amazon.nova-micro-v1:0": {
554
+ id: "eu.amazon.nova-micro-v1:0",
555
+ name: "Nova Micro (EU)",
556
+ api: "bedrock-converse-stream",
557
+ provider: "amazon-bedrock",
558
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
559
+ reasoning: false,
560
+ input: ["text"],
561
+ cost: {
562
+ input: 0.04,
563
+ output: 0.16,
564
+ cacheRead: 0.01,
565
+ cacheWrite: 0.04,
566
+ },
567
+ contextWindow: 128000,
568
+ maxTokens: 10000,
569
+ },
570
+ "eu.amazon.nova-pro-v1:0": {
571
+ id: "eu.amazon.nova-pro-v1:0",
572
+ name: "Nova Pro (EU)",
573
+ api: "bedrock-converse-stream",
574
+ provider: "amazon-bedrock",
575
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
576
+ reasoning: false,
577
+ input: ["text", "image"],
578
+ cost: {
579
+ input: 0.92,
580
+ output: 3.68,
581
+ cacheRead: 0.23,
582
+ cacheWrite: 0.92,
583
+ },
584
+ contextWindow: 300000,
585
+ maxTokens: 10000,
586
+ },
450
587
  "eu.anthropic.claude-fable-5": {
451
588
  id: "eu.anthropic.claude-fable-5",
452
589
  name: "Claude Fable 5 (EU)",
@@ -619,6 +756,40 @@ export const MODELS = {
619
756
  contextWindow: 1000000,
620
757
  maxTokens: 128000,
621
758
  },
759
+ "eu.mistral.pixtral-large-2502-v1:0": {
760
+ id: "eu.mistral.pixtral-large-2502-v1:0",
761
+ name: "Pixtral Large (25.02) (EU)",
762
+ api: "bedrock-converse-stream",
763
+ provider: "amazon-bedrock",
764
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
765
+ reasoning: false,
766
+ input: ["text", "image"],
767
+ cost: {
768
+ input: 2,
769
+ output: 6,
770
+ cacheRead: 0,
771
+ cacheWrite: 0,
772
+ },
773
+ contextWindow: 128000,
774
+ maxTokens: 8192,
775
+ },
776
+ "global.amazon.nova-2-lite-v1:0": {
777
+ id: "global.amazon.nova-2-lite-v1:0",
778
+ name: "Nova 2 Lite (Global)",
779
+ api: "bedrock-converse-stream",
780
+ provider: "amazon-bedrock",
781
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
782
+ reasoning: true,
783
+ input: ["text", "image"],
784
+ cost: {
785
+ input: 0.3,
786
+ output: 2.5,
787
+ cacheRead: 0.075,
788
+ cacheWrite: 0.3,
789
+ },
790
+ contextWindow: 1000000,
791
+ maxTokens: 65535,
792
+ },
622
793
  "global.anthropic.claude-fable-5": {
623
794
  id: "global.anthropic.claude-fable-5",
624
795
  name: "Claude Fable 5 (Global)",
@@ -862,6 +1033,24 @@ export const MODELS = {
862
1033
  contextWindow: 1050000,
863
1034
  maxTokens: 128000,
864
1035
  },
1036
+ "global.openai.gpt-6-astra": {
1037
+ id: "global.openai.gpt-6-astra",
1038
+ name: "GPT-6 Astra (Global)",
1039
+ api: "bedrock-converse-stream",
1040
+ provider: "amazon-bedrock",
1041
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1042
+ reasoning: true,
1043
+ thinkingLevelMap: { "xhigh": "xhigh" },
1044
+ input: ["text", "image"],
1045
+ cost: {
1046
+ input: 10,
1047
+ output: 50,
1048
+ cacheRead: 1,
1049
+ cacheWrite: 12.5,
1050
+ },
1051
+ contextWindow: 1050000,
1052
+ maxTokens: 128000,
1053
+ },
865
1054
  "global.xai.grok-4.6": {
866
1055
  id: "global.xai.grok-4.6",
867
1056
  name: "Grok 4.6 (Global)",
@@ -879,30 +1068,47 @@ export const MODELS = {
879
1068
  contextWindow: 500000,
880
1069
  maxTokens: 500000,
881
1070
  },
882
- "google.gemma-3-27b-it": {
883
- id: "google.gemma-3-27b-it",
884
- name: "Google Gemma 3 27B Instruct",
1071
+ "google.gemma-4-26b-a4b": {
1072
+ id: "google.gemma-4-26b-a4b",
1073
+ name: "Gemma 4 26B A4B IT",
885
1074
  api: "bedrock-converse-stream",
886
1075
  provider: "amazon-bedrock",
887
1076
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
888
- reasoning: false,
1077
+ reasoning: true,
889
1078
  input: ["text", "image"],
890
1079
  cost: {
891
- input: 0.12,
892
- output: 0.2,
1080
+ input: 0.13,
1081
+ output: 0.4,
893
1082
  cacheRead: 0,
894
1083
  cacheWrite: 0,
895
1084
  },
896
- contextWindow: 202752,
897
- maxTokens: 8192,
1085
+ contextWindow: 262144,
1086
+ maxTokens: 32768,
898
1087
  },
899
- "google.gemma-3-4b-it": {
900
- id: "google.gemma-3-4b-it",
901
- name: "Gemma 3 4B IT",
1088
+ "google.gemma-4-31b": {
1089
+ id: "google.gemma-4-31b",
1090
+ name: "Gemma 4 31B IT",
902
1091
  api: "bedrock-converse-stream",
903
1092
  provider: "amazon-bedrock",
904
1093
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
905
- reasoning: false,
1094
+ reasoning: true,
1095
+ input: ["text", "image"],
1096
+ cost: {
1097
+ input: 0.14,
1098
+ output: 0.4,
1099
+ cacheRead: 0,
1100
+ cacheWrite: 0,
1101
+ },
1102
+ contextWindow: 262144,
1103
+ maxTokens: 32768,
1104
+ },
1105
+ "google.gemma-4-e2b": {
1106
+ id: "google.gemma-4-e2b",
1107
+ name: "Gemma 4 E2B IT",
1108
+ api: "bedrock-converse-stream",
1109
+ provider: "amazon-bedrock",
1110
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1111
+ reasoning: true,
906
1112
  input: ["text", "image"],
907
1113
  cost: {
908
1114
  input: 0.04,
@@ -910,8 +1116,61 @@ export const MODELS = {
910
1116
  cacheRead: 0,
911
1117
  cacheWrite: 0,
912
1118
  },
913
- contextWindow: 128000,
914
- maxTokens: 4096,
1119
+ contextWindow: 131072,
1120
+ maxTokens: 8192,
1121
+ },
1122
+ "in.openai.gpt-5.6-luna": {
1123
+ id: "in.openai.gpt-5.6-luna",
1124
+ name: "GPT-5.6 Luna (India)",
1125
+ api: "bedrock-converse-stream",
1126
+ provider: "amazon-bedrock",
1127
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1128
+ reasoning: true,
1129
+ thinkingLevelMap: { "xhigh": "xhigh" },
1130
+ input: ["text", "image"],
1131
+ cost: {
1132
+ input: 0.22,
1133
+ output: 1.32,
1134
+ cacheRead: 0.022,
1135
+ cacheWrite: 0.275,
1136
+ },
1137
+ contextWindow: 1050000,
1138
+ maxTokens: 128000,
1139
+ },
1140
+ "in.openai.gpt-5.6-terra": {
1141
+ id: "in.openai.gpt-5.6-terra",
1142
+ name: "GPT-5.6 Terra (India)",
1143
+ api: "bedrock-converse-stream",
1144
+ provider: "amazon-bedrock",
1145
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1146
+ reasoning: true,
1147
+ thinkingLevelMap: { "xhigh": "xhigh" },
1148
+ input: ["text", "image"],
1149
+ cost: {
1150
+ input: 2.2,
1151
+ output: 13.2,
1152
+ cacheRead: 0.22,
1153
+ cacheWrite: 2.75,
1154
+ },
1155
+ contextWindow: 1050000,
1156
+ maxTokens: 128000,
1157
+ },
1158
+ "jp.amazon.nova-2-lite-v1:0": {
1159
+ id: "jp.amazon.nova-2-lite-v1:0",
1160
+ name: "Nova 2 Lite (JP)",
1161
+ api: "bedrock-converse-stream",
1162
+ provider: "amazon-bedrock",
1163
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1164
+ reasoning: true,
1165
+ input: ["text", "image"],
1166
+ cost: {
1167
+ input: 0.396,
1168
+ output: 3.311,
1169
+ cacheRead: 0.099,
1170
+ cacheWrite: 0.396,
1171
+ },
1172
+ contextWindow: 1000000,
1173
+ maxTokens: 65535,
915
1174
  },
916
1175
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
917
1176
  id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
@@ -1099,7 +1358,7 @@ export const MODELS = {
1099
1358
  cacheWrite: 0,
1100
1359
  },
1101
1360
  contextWindow: 1000000,
1102
- maxTokens: 16384,
1361
+ maxTokens: 8192,
1103
1362
  },
1104
1363
  "meta.llama4-scout-17b-instruct-v1:0": {
1105
1364
  id: "meta.llama4-scout-17b-instruct-v1:0",
@@ -1115,12 +1374,12 @@ export const MODELS = {
1115
1374
  cacheRead: 0,
1116
1375
  cacheWrite: 0,
1117
1376
  },
1118
- contextWindow: 3500000,
1119
- maxTokens: 16384,
1377
+ contextWindow: 10000000,
1378
+ maxTokens: 8192,
1120
1379
  },
1121
1380
  "minimax.minimax-m2": {
1122
1381
  id: "minimax.minimax-m2",
1123
- name: "MiniMax M2",
1382
+ name: "MiniMax-M2",
1124
1383
  api: "bedrock-converse-stream",
1125
1384
  provider: "amazon-bedrock",
1126
1385
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1137,7 +1396,7 @@ export const MODELS = {
1137
1396
  },
1138
1397
  "minimax.minimax-m2.1": {
1139
1398
  id: "minimax.minimax-m2.1",
1140
- name: "MiniMax M2.1",
1399
+ name: "MiniMax-M2.1",
1141
1400
  api: "bedrock-converse-stream",
1142
1401
  provider: "amazon-bedrock",
1143
1402
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1154,7 +1413,7 @@ export const MODELS = {
1154
1413
  },
1155
1414
  "minimax.minimax-m2.5": {
1156
1415
  id: "minimax.minimax-m2.5",
1157
- name: "MiniMax M2.5",
1416
+ name: "MiniMax-M2.5",
1158
1417
  api: "bedrock-converse-stream",
1159
1418
  provider: "amazon-bedrock",
1160
1419
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1210,7 +1469,7 @@ export const MODELS = {
1210
1469
  provider: "amazon-bedrock",
1211
1470
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1212
1471
  reasoning: false,
1213
- input: ["text"],
1472
+ input: ["text", "image"],
1214
1473
  cost: {
1215
1474
  input: 0.2,
1216
1475
  output: 0.2,
@@ -1244,7 +1503,7 @@ export const MODELS = {
1244
1503
  provider: "amazon-bedrock",
1245
1504
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1246
1505
  reasoning: false,
1247
- input: ["text"],
1506
+ input: ["text", "image"],
1248
1507
  cost: {
1249
1508
  input: 0.15,
1250
1509
  output: 0.15,
@@ -1302,7 +1561,7 @@ export const MODELS = {
1302
1561
  cacheRead: 0,
1303
1562
  cacheWrite: 0,
1304
1563
  },
1305
- contextWindow: 128000,
1564
+ contextWindow: 32768,
1306
1565
  maxTokens: 4096,
1307
1566
  },
1308
1567
  "mistral.voxtral-small-24b-2507": {
@@ -1314,12 +1573,12 @@ export const MODELS = {
1314
1573
  reasoning: false,
1315
1574
  input: ["text"],
1316
1575
  cost: {
1317
- input: 0.15,
1318
- output: 0.35,
1576
+ input: 0.1,
1577
+ output: 0.3,
1319
1578
  cacheRead: 0,
1320
1579
  cacheWrite: 0,
1321
1580
  },
1322
- contextWindow: 32000,
1581
+ contextWindow: 32768,
1323
1582
  maxTokens: 8192,
1324
1583
  },
1325
1584
  "moonshot.kimi-k2-thinking": {
@@ -1354,7 +1613,7 @@ export const MODELS = {
1354
1613
  cacheWrite: 0,
1355
1614
  },
1356
1615
  contextWindow: 262143,
1357
- maxTokens: 16000,
1616
+ maxTokens: 16384,
1358
1617
  },
1359
1618
  "nvidia.nemotron-nano-12b-v2": {
1360
1619
  id: "nvidia.nemotron-nano-12b-v2",
@@ -1371,7 +1630,7 @@ export const MODELS = {
1371
1630
  cacheWrite: 0,
1372
1631
  },
1373
1632
  contextWindow: 128000,
1374
- maxTokens: 4096,
1633
+ maxTokens: 8192,
1375
1634
  },
1376
1635
  "nvidia.nemotron-nano-3-30b": {
1377
1636
  id: "nvidia.nemotron-nano-3-30b",
@@ -1387,9 +1646,9 @@ export const MODELS = {
1387
1646
  cacheRead: 0,
1388
1647
  cacheWrite: 0,
1389
1648
  },
1390
- contextWindow: 128000,
1391
- maxTokens: 4096,
1392
- },
1649
+ contextWindow: 262144,
1650
+ maxTokens: 8192,
1651
+ },
1393
1652
  "nvidia.nemotron-nano-9b-v2": {
1394
1653
  id: "nvidia.nemotron-nano-9b-v2",
1395
1654
  name: "NVIDIA Nemotron Nano 9B v2",
@@ -1404,8 +1663,8 @@ export const MODELS = {
1404
1663
  cacheRead: 0,
1405
1664
  cacheWrite: 0,
1406
1665
  },
1407
- contextWindow: 128000,
1408
- maxTokens: 4096,
1666
+ contextWindow: 131072,
1667
+ maxTokens: 8192,
1409
1668
  },
1410
1669
  "nvidia.nemotron-super-3-120b": {
1411
1670
  id: "nvidia.nemotron-super-3-120b",
@@ -1514,6 +1773,24 @@ export const MODELS = {
1514
1773
  contextWindow: 1050000,
1515
1774
  maxTokens: 128000,
1516
1775
  },
1776
+ "openai.gpt-6-astra": {
1777
+ id: "openai.gpt-6-astra",
1778
+ name: "GPT-6 Astra",
1779
+ api: "bedrock-converse-stream",
1780
+ provider: "amazon-bedrock",
1781
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1782
+ reasoning: true,
1783
+ thinkingLevelMap: { "xhigh": "xhigh" },
1784
+ input: ["text", "image"],
1785
+ cost: {
1786
+ input: 11,
1787
+ output: 55,
1788
+ cacheRead: 1.1,
1789
+ cacheWrite: 13.75,
1790
+ },
1791
+ contextWindow: 1050000,
1792
+ maxTokens: 128000,
1793
+ },
1517
1794
  "openai.gpt-oss-120b": {
1518
1795
  id: "openai.gpt-oss-120b",
1519
1796
  name: "gpt-oss-120b",
@@ -1588,7 +1865,7 @@ export const MODELS = {
1588
1865
  api: "bedrock-converse-stream",
1589
1866
  provider: "amazon-bedrock",
1590
1867
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1591
- reasoning: false,
1868
+ reasoning: true,
1592
1869
  input: ["text"],
1593
1870
  cost: {
1594
1871
  input: 0.15,
@@ -1605,7 +1882,7 @@ export const MODELS = {
1605
1882
  api: "bedrock-converse-stream",
1606
1883
  provider: "amazon-bedrock",
1607
1884
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1608
- reasoning: false,
1885
+ reasoning: true,
1609
1886
  input: ["text"],
1610
1887
  cost: {
1611
1888
  input: 0.07,
@@ -1618,7 +1895,7 @@ export const MODELS = {
1618
1895
  },
1619
1896
  "qwen.qwen3-235b-a22b-2507-v1:0": {
1620
1897
  id: "qwen.qwen3-235b-a22b-2507-v1:0",
1621
- name: "Qwen3 235B A22B 2507",
1898
+ name: "Qwen3 235B-A22B Instruct 2507",
1622
1899
  api: "bedrock-converse-stream",
1623
1900
  provider: "amazon-bedrock",
1624
1901
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1635,7 +1912,7 @@ export const MODELS = {
1635
1912
  },
1636
1913
  "qwen.qwen3-32b-v1:0": {
1637
1914
  id: "qwen.qwen3-32b-v1:0",
1638
- name: "Qwen3 32B (dense)",
1915
+ name: "Qwen3 32B",
1639
1916
  api: "bedrock-converse-stream",
1640
1917
  provider: "amazon-bedrock",
1641
1918
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1647,12 +1924,12 @@ export const MODELS = {
1647
1924
  cacheRead: 0,
1648
1925
  cacheWrite: 0,
1649
1926
  },
1650
- contextWindow: 16384,
1927
+ contextWindow: 32768,
1651
1928
  maxTokens: 16384,
1652
1929
  },
1653
1930
  "qwen.qwen3-coder-30b-a3b-v1:0": {
1654
1931
  id: "qwen.qwen3-coder-30b-a3b-v1:0",
1655
- name: "Qwen3 Coder 30B A3B Instruct",
1932
+ name: "Qwen3-Coder 30B-A3B Instruct",
1656
1933
  api: "bedrock-converse-stream",
1657
1934
  provider: "amazon-bedrock",
1658
1935
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1669,14 +1946,14 @@ export const MODELS = {
1669
1946
  },
1670
1947
  "qwen.qwen3-coder-480b-a35b-v1:0": {
1671
1948
  id: "qwen.qwen3-coder-480b-a35b-v1:0",
1672
- name: "Qwen3 Coder 480B A35B Instruct",
1949
+ name: "Qwen3-Coder 480B-A35B Instruct",
1673
1950
  api: "bedrock-converse-stream",
1674
1951
  provider: "amazon-bedrock",
1675
1952
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1676
1953
  reasoning: false,
1677
1954
  input: ["text"],
1678
1955
  cost: {
1679
- input: 0.22,
1956
+ input: 0.45,
1680
1957
  output: 1.8,
1681
1958
  cacheRead: 0,
1682
1959
  cacheWrite: 0,
@@ -1690,51 +1967,170 @@ export const MODELS = {
1690
1967
  api: "bedrock-converse-stream",
1691
1968
  provider: "amazon-bedrock",
1692
1969
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1693
- reasoning: true,
1970
+ reasoning: false,
1694
1971
  input: ["text"],
1695
1972
  cost: {
1696
- input: 0.22,
1697
- output: 1.8,
1973
+ input: 0.5,
1974
+ output: 1.2,
1698
1975
  cacheRead: 0,
1699
1976
  cacheWrite: 0,
1700
1977
  },
1701
- contextWindow: 131072,
1978
+ contextWindow: 262144,
1702
1979
  maxTokens: 65536,
1703
1980
  },
1704
1981
  "qwen.qwen3-next-80b-a3b": {
1705
1982
  id: "qwen.qwen3-next-80b-a3b",
1706
- name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
1983
+ name: "Qwen3-Next 80B-A3B Instruct",
1707
1984
  api: "bedrock-converse-stream",
1708
1985
  provider: "amazon-bedrock",
1709
1986
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1710
1987
  reasoning: false,
1711
1988
  input: ["text"],
1712
1989
  cost: {
1713
- input: 0.14,
1714
- output: 1.4,
1990
+ input: 0.15,
1991
+ output: 1.2,
1715
1992
  cacheRead: 0,
1716
1993
  cacheWrite: 0,
1717
1994
  },
1718
- contextWindow: 262000,
1995
+ contextWindow: 262144,
1719
1996
  maxTokens: 262000,
1720
1997
  },
1721
1998
  "qwen.qwen3-vl-235b-a22b": {
1722
1999
  id: "qwen.qwen3-vl-235b-a22b",
1723
- name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
2000
+ name: "Qwen3 VL 235B A22B Instruct",
1724
2001
  api: "bedrock-converse-stream",
1725
2002
  provider: "amazon-bedrock",
1726
2003
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1727
2004
  reasoning: false,
1728
2005
  input: ["text", "image"],
1729
2006
  cost: {
1730
- input: 0.3,
1731
- output: 1.5,
2007
+ input: 0.53,
2008
+ output: 2.66,
1732
2009
  cacheRead: 0,
1733
2010
  cacheWrite: 0,
1734
2011
  },
1735
- contextWindow: 262000,
2012
+ contextWindow: 262144,
1736
2013
  maxTokens: 262000,
1737
2014
  },
2015
+ "us-gov.openai.gpt-oss-120b-1:0": {
2016
+ id: "us-gov.openai.gpt-oss-120b-1:0",
2017
+ name: "gpt-oss-120b (GovCloud)",
2018
+ api: "bedrock-converse-stream",
2019
+ provider: "amazon-bedrock",
2020
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2021
+ reasoning: true,
2022
+ input: ["text"],
2023
+ cost: {
2024
+ input: 0.18,
2025
+ output: 0.72,
2026
+ cacheRead: 0,
2027
+ cacheWrite: 0,
2028
+ },
2029
+ contextWindow: 128000,
2030
+ maxTokens: 16384,
2031
+ },
2032
+ "us-gov.openai.gpt-oss-20b-1:0": {
2033
+ id: "us-gov.openai.gpt-oss-20b-1:0",
2034
+ name: "gpt-oss-20b (GovCloud)",
2035
+ api: "bedrock-converse-stream",
2036
+ provider: "amazon-bedrock",
2037
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2038
+ reasoning: true,
2039
+ input: ["text"],
2040
+ cost: {
2041
+ input: 0.084,
2042
+ output: 0.36,
2043
+ cacheRead: 0,
2044
+ cacheWrite: 0,
2045
+ },
2046
+ contextWindow: 128000,
2047
+ maxTokens: 16384,
2048
+ },
2049
+ "us.amazon.nova-2-lite-v1:0": {
2050
+ id: "us.amazon.nova-2-lite-v1:0",
2051
+ name: "Nova 2 Lite (US)",
2052
+ api: "bedrock-converse-stream",
2053
+ provider: "amazon-bedrock",
2054
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2055
+ reasoning: true,
2056
+ input: ["text", "image"],
2057
+ cost: {
2058
+ input: 0.33,
2059
+ output: 2.75,
2060
+ cacheRead: 0.0825,
2061
+ cacheWrite: 0.33,
2062
+ },
2063
+ contextWindow: 1000000,
2064
+ maxTokens: 65535,
2065
+ },
2066
+ "us.amazon.nova-lite-v1:0": {
2067
+ id: "us.amazon.nova-lite-v1:0",
2068
+ name: "Nova Lite (US)",
2069
+ api: "bedrock-converse-stream",
2070
+ provider: "amazon-bedrock",
2071
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2072
+ reasoning: false,
2073
+ input: ["text", "image"],
2074
+ cost: {
2075
+ input: 0.06,
2076
+ output: 0.24,
2077
+ cacheRead: 0.015,
2078
+ cacheWrite: 0.06,
2079
+ },
2080
+ contextWindow: 300000,
2081
+ maxTokens: 10000,
2082
+ },
2083
+ "us.amazon.nova-micro-v1:0": {
2084
+ id: "us.amazon.nova-micro-v1:0",
2085
+ name: "Nova Micro (US)",
2086
+ api: "bedrock-converse-stream",
2087
+ provider: "amazon-bedrock",
2088
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2089
+ reasoning: false,
2090
+ input: ["text"],
2091
+ cost: {
2092
+ input: 0.035,
2093
+ output: 0.14,
2094
+ cacheRead: 0.00875,
2095
+ cacheWrite: 0.035,
2096
+ },
2097
+ contextWindow: 128000,
2098
+ maxTokens: 10000,
2099
+ },
2100
+ "us.amazon.nova-premier-v1:0": {
2101
+ id: "us.amazon.nova-premier-v1:0",
2102
+ name: "Nova Premier (US)",
2103
+ api: "bedrock-converse-stream",
2104
+ provider: "amazon-bedrock",
2105
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2106
+ reasoning: false,
2107
+ input: ["text", "image"],
2108
+ cost: {
2109
+ input: 2.5,
2110
+ output: 12.5,
2111
+ cacheRead: 0.625,
2112
+ cacheWrite: 2.5,
2113
+ },
2114
+ contextWindow: 1000000,
2115
+ maxTokens: 10000,
2116
+ },
2117
+ "us.amazon.nova-pro-v1:0": {
2118
+ id: "us.amazon.nova-pro-v1:0",
2119
+ name: "Nova Pro (US)",
2120
+ api: "bedrock-converse-stream",
2121
+ provider: "amazon-bedrock",
2122
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2123
+ reasoning: false,
2124
+ input: ["text", "image"],
2125
+ cost: {
2126
+ input: 0.8,
2127
+ output: 3.2,
2128
+ cacheRead: 0.2,
2129
+ cacheWrite: 0.8,
2130
+ },
2131
+ contextWindow: 300000,
2132
+ maxTokens: 10000,
2133
+ },
1738
2134
  "us.anthropic.claude-fable-5": {
1739
2135
  id: "us.anthropic.claude-fable-5",
1740
2136
  name: "Claude Fable 5 (US)",
@@ -1870,127 +2266,284 @@ export const MODELS = {
1870
2266
  cacheRead: 0.5,
1871
2267
  cacheWrite: 6.25,
1872
2268
  },
1873
- contextWindow: 1000000,
1874
- maxTokens: 128000,
2269
+ contextWindow: 1000000,
2270
+ maxTokens: 128000,
2271
+ },
2272
+ "us.anthropic.claude-opus-5": {
2273
+ id: "us.anthropic.claude-opus-5",
2274
+ name: "Claude Opus 5 (US)",
2275
+ api: "bedrock-converse-stream",
2276
+ provider: "amazon-bedrock",
2277
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2278
+ reasoning: true,
2279
+ input: ["text", "image"],
2280
+ cost: {
2281
+ input: 5,
2282
+ output: 25,
2283
+ cacheRead: 0.5,
2284
+ cacheWrite: 6.25,
2285
+ },
2286
+ contextWindow: 1000000,
2287
+ maxTokens: 128000,
2288
+ },
2289
+ "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2290
+ id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
2291
+ name: "Claude Sonnet 4.5 (US)",
2292
+ api: "bedrock-converse-stream",
2293
+ provider: "amazon-bedrock",
2294
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2295
+ reasoning: true,
2296
+ input: ["text", "image"],
2297
+ cost: {
2298
+ input: 3,
2299
+ output: 15,
2300
+ cacheRead: 0.3,
2301
+ cacheWrite: 3.75,
2302
+ },
2303
+ contextWindow: 200000,
2304
+ maxTokens: 64000,
2305
+ },
2306
+ "us.anthropic.claude-sonnet-4-6": {
2307
+ id: "us.anthropic.claude-sonnet-4-6",
2308
+ name: "Claude Sonnet 4.6 (US)",
2309
+ api: "bedrock-converse-stream",
2310
+ provider: "amazon-bedrock",
2311
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2312
+ reasoning: true,
2313
+ input: ["text", "image"],
2314
+ cost: {
2315
+ input: 3,
2316
+ output: 15,
2317
+ cacheRead: 0.3,
2318
+ cacheWrite: 3.75,
2319
+ },
2320
+ contextWindow: 1000000,
2321
+ maxTokens: 64000,
2322
+ },
2323
+ "us.anthropic.claude-sonnet-5": {
2324
+ id: "us.anthropic.claude-sonnet-5",
2325
+ name: "Claude Sonnet 5 (US)",
2326
+ api: "bedrock-converse-stream",
2327
+ provider: "amazon-bedrock",
2328
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2329
+ reasoning: true,
2330
+ input: ["text", "image"],
2331
+ cost: {
2332
+ input: 2,
2333
+ output: 10,
2334
+ cacheRead: 0.2,
2335
+ cacheWrite: 2.5,
2336
+ },
2337
+ contextWindow: 1000000,
2338
+ maxTokens: 128000,
2339
+ },
2340
+ "us.meta.llama3-1-70b-instruct-v1:0": {
2341
+ id: "us.meta.llama3-1-70b-instruct-v1:0",
2342
+ name: "Llama 3.1 70B Instruct (US)",
2343
+ api: "bedrock-converse-stream",
2344
+ provider: "amazon-bedrock",
2345
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2346
+ reasoning: false,
2347
+ input: ["text"],
2348
+ cost: {
2349
+ input: 0.72,
2350
+ output: 0.72,
2351
+ cacheRead: 0,
2352
+ cacheWrite: 0,
2353
+ },
2354
+ contextWindow: 128000,
2355
+ maxTokens: 4096,
2356
+ },
2357
+ "us.meta.llama3-1-8b-instruct-v1:0": {
2358
+ id: "us.meta.llama3-1-8b-instruct-v1:0",
2359
+ name: "Llama 3.1 8B Instruct (US)",
2360
+ api: "bedrock-converse-stream",
2361
+ provider: "amazon-bedrock",
2362
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2363
+ reasoning: false,
2364
+ input: ["text"],
2365
+ cost: {
2366
+ input: 0.22,
2367
+ output: 0.22,
2368
+ cacheRead: 0,
2369
+ cacheWrite: 0,
2370
+ },
2371
+ contextWindow: 128000,
2372
+ maxTokens: 4096,
2373
+ },
2374
+ "us.meta.llama3-3-70b-instruct-v1:0": {
2375
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
2376
+ name: "Llama 3.3 70B Instruct (US)",
2377
+ api: "bedrock-converse-stream",
2378
+ provider: "amazon-bedrock",
2379
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2380
+ reasoning: false,
2381
+ input: ["text"],
2382
+ cost: {
2383
+ input: 0.72,
2384
+ output: 0.72,
2385
+ cacheRead: 0,
2386
+ cacheWrite: 0,
2387
+ },
2388
+ contextWindow: 128000,
2389
+ maxTokens: 4096,
2390
+ },
2391
+ "us.meta.llama4-maverick-17b-instruct-v1:0": {
2392
+ id: "us.meta.llama4-maverick-17b-instruct-v1:0",
2393
+ name: "Llama 4 Maverick 17B Instruct (US)",
2394
+ api: "bedrock-converse-stream",
2395
+ provider: "amazon-bedrock",
2396
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2397
+ reasoning: false,
2398
+ input: ["text", "image"],
2399
+ cost: {
2400
+ input: 0.24,
2401
+ output: 0.97,
2402
+ cacheRead: 0,
2403
+ cacheWrite: 0,
2404
+ },
2405
+ contextWindow: 1000000,
2406
+ maxTokens: 8192,
2407
+ },
2408
+ "us.meta.llama4-scout-17b-instruct-v1:0": {
2409
+ id: "us.meta.llama4-scout-17b-instruct-v1:0",
2410
+ name: "Llama 4 Scout 17B Instruct (US)",
2411
+ api: "bedrock-converse-stream",
2412
+ provider: "amazon-bedrock",
2413
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2414
+ reasoning: false,
2415
+ input: ["text", "image"],
2416
+ cost: {
2417
+ input: 0.17,
2418
+ output: 0.66,
2419
+ cacheRead: 0,
2420
+ cacheWrite: 0,
2421
+ },
2422
+ contextWindow: 10000000,
2423
+ maxTokens: 8192,
1875
2424
  },
1876
- "us.anthropic.claude-opus-5": {
1877
- id: "us.anthropic.claude-opus-5",
1878
- name: "Claude Opus 5 (US)",
2425
+ "us.mistral.pixtral-large-2502-v1:0": {
2426
+ id: "us.mistral.pixtral-large-2502-v1:0",
2427
+ name: "Pixtral Large (25.02) (US)",
1879
2428
  api: "bedrock-converse-stream",
1880
2429
  provider: "amazon-bedrock",
1881
2430
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1882
- reasoning: true,
2431
+ reasoning: false,
1883
2432
  input: ["text", "image"],
1884
2433
  cost: {
1885
- input: 5,
1886
- output: 25,
1887
- cacheRead: 0.5,
1888
- cacheWrite: 6.25,
2434
+ input: 2,
2435
+ output: 6,
2436
+ cacheRead: 0,
2437
+ cacheWrite: 0,
1889
2438
  },
1890
- contextWindow: 1000000,
1891
- maxTokens: 128000,
2439
+ contextWindow: 128000,
2440
+ maxTokens: 8192,
1892
2441
  },
1893
- "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
1894
- id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
1895
- name: "Claude Sonnet 4.5 (US)",
2442
+ "us.openai.gpt-5.6-luna": {
2443
+ id: "us.openai.gpt-5.6-luna",
2444
+ name: "GPT-5.6 Luna (US)",
1896
2445
  api: "bedrock-converse-stream",
1897
2446
  provider: "amazon-bedrock",
1898
2447
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1899
2448
  reasoning: true,
2449
+ thinkingLevelMap: { "xhigh": "xhigh" },
1900
2450
  input: ["text", "image"],
1901
2451
  cost: {
1902
- input: 3,
1903
- output: 15,
1904
- cacheRead: 0.3,
1905
- cacheWrite: 3.75,
2452
+ input: 0.22,
2453
+ output: 1.32,
2454
+ cacheRead: 0.022,
2455
+ cacheWrite: 0.275,
1906
2456
  },
1907
- contextWindow: 200000,
1908
- maxTokens: 64000,
2457
+ contextWindow: 1050000,
2458
+ maxTokens: 128000,
1909
2459
  },
1910
- "us.anthropic.claude-sonnet-4-6": {
1911
- id: "us.anthropic.claude-sonnet-4-6",
1912
- name: "Claude Sonnet 4.6 (US)",
2460
+ "us.openai.gpt-5.6-sol": {
2461
+ id: "us.openai.gpt-5.6-sol",
2462
+ name: "GPT-5.6 Sol (US)",
1913
2463
  api: "bedrock-converse-stream",
1914
2464
  provider: "amazon-bedrock",
1915
2465
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1916
2466
  reasoning: true,
2467
+ thinkingLevelMap: { "xhigh": "xhigh" },
1917
2468
  input: ["text", "image"],
1918
2469
  cost: {
1919
- input: 3,
1920
- output: 15,
1921
- cacheRead: 0.3,
1922
- cacheWrite: 3.75,
2470
+ input: 4.4,
2471
+ output: 22,
2472
+ cacheRead: 0.44,
2473
+ cacheWrite: 5.5,
1923
2474
  },
1924
- contextWindow: 1000000,
1925
- maxTokens: 64000,
2475
+ contextWindow: 1050000,
2476
+ maxTokens: 128000,
1926
2477
  },
1927
- "us.anthropic.claude-sonnet-5": {
1928
- id: "us.anthropic.claude-sonnet-5",
1929
- name: "Claude Sonnet 5 (US)",
2478
+ "us.openai.gpt-5.6-terra": {
2479
+ id: "us.openai.gpt-5.6-terra",
2480
+ name: "GPT-5.6 Terra (US)",
1930
2481
  api: "bedrock-converse-stream",
1931
2482
  provider: "amazon-bedrock",
1932
2483
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1933
2484
  reasoning: true,
2485
+ thinkingLevelMap: { "xhigh": "xhigh" },
1934
2486
  input: ["text", "image"],
1935
2487
  cost: {
1936
- input: 2,
1937
- output: 10,
1938
- cacheRead: 0.2,
1939
- cacheWrite: 2.5,
2488
+ input: 2.2,
2489
+ output: 13.2,
2490
+ cacheRead: 0.22,
2491
+ cacheWrite: 2.75,
1940
2492
  },
1941
- contextWindow: 1000000,
2493
+ contextWindow: 1050000,
1942
2494
  maxTokens: 128000,
1943
2495
  },
1944
- "us.deepseek.r1-v1:0": {
1945
- id: "us.deepseek.r1-v1:0",
1946
- name: "DeepSeek-R1 (US)",
2496
+ "us.openai.gpt-6-astra": {
2497
+ id: "us.openai.gpt-6-astra",
2498
+ name: "GPT-6 Astra (US)",
1947
2499
  api: "bedrock-converse-stream",
1948
2500
  provider: "amazon-bedrock",
1949
2501
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1950
2502
  reasoning: true,
1951
- input: ["text"],
2503
+ thinkingLevelMap: { "xhigh": "xhigh" },
2504
+ input: ["text", "image"],
1952
2505
  cost: {
1953
- input: 1.35,
1954
- output: 5.4,
1955
- cacheRead: 0,
1956
- cacheWrite: 0,
2506
+ input: 11,
2507
+ output: 55,
2508
+ cacheRead: 1.1,
2509
+ cacheWrite: 13.75,
1957
2510
  },
1958
- contextWindow: 128000,
1959
- maxTokens: 32768,
2511
+ contextWindow: 1050000,
2512
+ maxTokens: 128000,
1960
2513
  },
1961
- "us.meta.llama4-maverick-17b-instruct-v1:0": {
1962
- id: "us.meta.llama4-maverick-17b-instruct-v1:0",
1963
- name: "Llama 4 Maverick 17B Instruct (US)",
2514
+ "us.writer.palmyra-x4-v1:0": {
2515
+ id: "us.writer.palmyra-x4-v1:0",
2516
+ name: "Palmyra X4 (US)",
1964
2517
  api: "bedrock-converse-stream",
1965
2518
  provider: "amazon-bedrock",
1966
2519
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1967
- reasoning: false,
1968
- input: ["text", "image"],
2520
+ reasoning: true,
2521
+ input: ["text"],
1969
2522
  cost: {
1970
- input: 0.24,
1971
- output: 0.97,
2523
+ input: 2.5,
2524
+ output: 10,
1972
2525
  cacheRead: 0,
1973
2526
  cacheWrite: 0,
1974
2527
  },
1975
- contextWindow: 1000000,
1976
- maxTokens: 16384,
2528
+ contextWindow: 122880,
2529
+ maxTokens: 8192,
1977
2530
  },
1978
- "us.meta.llama4-scout-17b-instruct-v1:0": {
1979
- id: "us.meta.llama4-scout-17b-instruct-v1:0",
1980
- name: "Llama 4 Scout 17B Instruct (US)",
2531
+ "us.writer.palmyra-x5-v1:0": {
2532
+ id: "us.writer.palmyra-x5-v1:0",
2533
+ name: "Palmyra X5 (US)",
1981
2534
  api: "bedrock-converse-stream",
1982
2535
  provider: "amazon-bedrock",
1983
2536
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1984
- reasoning: false,
1985
- input: ["text", "image"],
2537
+ reasoning: true,
2538
+ input: ["text"],
1986
2539
  cost: {
1987
- input: 0.17,
1988
- output: 0.66,
2540
+ input: 0.6,
2541
+ output: 6,
1989
2542
  cacheRead: 0,
1990
2543
  cacheWrite: 0,
1991
2544
  },
1992
- contextWindow: 3500000,
1993
- maxTokens: 16384,
2545
+ contextWindow: 1040000,
2546
+ maxTokens: 8192,
1994
2547
  },
1995
2548
  "us.xai.grok-4.6": {
1996
2549
  id: "us.xai.grok-4.6",
@@ -2126,7 +2679,7 @@ export const MODELS = {
2126
2679
  cacheWrite: 0,
2127
2680
  },
2128
2681
  contextWindow: 202752,
2129
- maxTokens: 101376,
2682
+ maxTokens: 131072,
2130
2683
  },
2131
2684
  },
2132
2685
  "anthropic": {
@@ -2153,7 +2706,9 @@ export const MODELS = {
2153
2706
  api: "anthropic-messages",
2154
2707
  provider: "anthropic",
2155
2708
  baseUrl: "https://api.anthropic.com",
2709
+ compat: { "supportsTemperature": false },
2156
2710
  reasoning: true,
2711
+ thinkingLevelMap: { "off": "low", "minimal": "low", "xhigh": "xhigh" },
2157
2712
  input: ["text", "image"],
2158
2713
  cost: {
2159
2714
  input: 10,
@@ -4158,6 +4713,24 @@ export const MODELS = {
4158
4713
  contextWindow: 1000000,
4159
4714
  maxTokens: 384000,
4160
4715
  },
4716
+ "accounts/fireworks/models/deepseek-v4p1-flash": {
4717
+ id: "accounts/fireworks/models/deepseek-v4p1-flash",
4718
+ name: "DeepSeek V4.1 Flash",
4719
+ api: "anthropic-messages",
4720
+ provider: "fireworks",
4721
+ baseUrl: "https://api.fireworks.ai/inference",
4722
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4723
+ reasoning: true,
4724
+ input: ["text", "image"],
4725
+ cost: {
4726
+ input: 0.22,
4727
+ output: 0.66,
4728
+ cacheRead: 0.007,
4729
+ cacheWrite: 0,
4730
+ },
4731
+ contextWindow: 1000000,
4732
+ maxTokens: 384000,
4733
+ },
4161
4734
  "accounts/fireworks/models/glm-5p2": {
4162
4735
  id: "accounts/fireworks/models/glm-5p2",
4163
4736
  name: "GLM 5.2",
@@ -4320,6 +4893,24 @@ export const MODELS = {
4320
4893
  contextWindow: 512000,
4321
4894
  maxTokens: 512000,
4322
4895
  },
4896
+ "accounts/fireworks/models/mistral-large-3-fp8": {
4897
+ id: "accounts/fireworks/models/mistral-large-3-fp8",
4898
+ name: "Mistral Large 3 675B Instruct 2512",
4899
+ api: "anthropic-messages",
4900
+ provider: "fireworks",
4901
+ baseUrl: "https://api.fireworks.ai/inference",
4902
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
4903
+ reasoning: false,
4904
+ input: ["text", "image"],
4905
+ cost: {
4906
+ input: 0,
4907
+ output: 0,
4908
+ cacheRead: 0,
4909
+ cacheWrite: 0,
4910
+ },
4911
+ contextWindow: 262144,
4912
+ maxTokens: 262144,
4913
+ },
4323
4914
  "accounts/fireworks/models/muse-glimmer-30b": {
4324
4915
  id: "accounts/fireworks/models/muse-glimmer-30b",
4325
4916
  name: "Muse Glimmer 30B",
@@ -6514,6 +7105,78 @@ export const MODELS = {
6514
7105
  contextWindow: 1000000,
6515
7106
  maxTokens: 384000,
6516
7107
  },
7108
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
7109
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
7110
+ name: "DeepSeek V4.1 Flash",
7111
+ api: "openai-completions",
7112
+ provider: "huggingface",
7113
+ baseUrl: "https://router.huggingface.co/v1",
7114
+ compat: { "supportsDeveloperRole": false },
7115
+ reasoning: true,
7116
+ input: ["text", "image"],
7117
+ cost: {
7118
+ input: 0.3,
7119
+ output: 1.2,
7120
+ cacheRead: 0,
7121
+ cacheWrite: 0,
7122
+ },
7123
+ contextWindow: 1048576,
7124
+ maxTokens: 384000,
7125
+ },
7126
+ "google/gemma-3-12b-it": {
7127
+ id: "google/gemma-3-12b-it",
7128
+ name: "Gemma 3 12B IT",
7129
+ api: "openai-completions",
7130
+ provider: "huggingface",
7131
+ baseUrl: "https://router.huggingface.co/v1",
7132
+ compat: { "supportsDeveloperRole": false },
7133
+ reasoning: false,
7134
+ input: ["text", "image"],
7135
+ cost: {
7136
+ input: 0.05,
7137
+ output: 0.15,
7138
+ cacheRead: 0,
7139
+ cacheWrite: 0,
7140
+ },
7141
+ contextWindow: 131072,
7142
+ maxTokens: 131072,
7143
+ },
7144
+ "google/gemma-3-27b-it": {
7145
+ id: "google/gemma-3-27b-it",
7146
+ name: "Gemma 3 27B IT",
7147
+ api: "openai-completions",
7148
+ provider: "huggingface",
7149
+ baseUrl: "https://router.huggingface.co/v1",
7150
+ compat: { "supportsDeveloperRole": false },
7151
+ reasoning: false,
7152
+ input: ["text", "image"],
7153
+ cost: {
7154
+ input: 0.08,
7155
+ output: 0.16,
7156
+ cacheRead: 0,
7157
+ cacheWrite: 0,
7158
+ },
7159
+ contextWindow: 131072,
7160
+ maxTokens: 131072,
7161
+ },
7162
+ "google/gemma-3-4b-it": {
7163
+ id: "google/gemma-3-4b-it",
7164
+ name: "Gemma 3 4B IT",
7165
+ api: "openai-completions",
7166
+ provider: "huggingface",
7167
+ baseUrl: "https://router.huggingface.co/v1",
7168
+ compat: { "supportsDeveloperRole": false },
7169
+ reasoning: false,
7170
+ input: ["text", "image"],
7171
+ cost: {
7172
+ input: 0.05,
7173
+ output: 0.1,
7174
+ cacheRead: 0,
7175
+ cacheWrite: 0,
7176
+ },
7177
+ contextWindow: 131072,
7178
+ maxTokens: 131072,
7179
+ },
6517
7180
  "google/gemma-4-26B-A4B-it": {
6518
7181
  id: "google/gemma-4-26B-A4B-it",
6519
7182
  name: "Gemma 4 26B A4B IT",
@@ -10048,9 +10711,9 @@ export const MODELS = {
10048
10711
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10049
10712
  input: ["text"],
10050
10713
  cost: {
10051
- input: 0.22,
10052
- output: 0.66,
10053
- cacheRead: 0.007,
10714
+ input: 0.15,
10715
+ output: 0.6,
10716
+ cacheRead: 0.003,
10054
10717
  cacheWrite: 0,
10055
10718
  },
10056
10719
  contextWindow: 1000000,
@@ -10067,9 +10730,9 @@ export const MODELS = {
10067
10730
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10068
10731
  input: ["text", "image"],
10069
10732
  cost: {
10070
- input: 0.22,
10071
- output: 0.66,
10072
- cacheRead: 0.007,
10733
+ input: 0.15,
10734
+ output: 0.6,
10735
+ cacheRead: 0.003,
10073
10736
  cacheWrite: 0,
10074
10737
  },
10075
10738
  contextWindow: 1000000,
@@ -10094,6 +10757,25 @@ export const MODELS = {
10094
10757
  contextWindow: 1000000,
10095
10758
  maxTokens: 384000,
10096
10759
  },
10760
+ "deepseek-v4.1-flash": {
10761
+ id: "deepseek-v4.1-flash",
10762
+ name: "DeepSeek V4.1 Flash",
10763
+ api: "openai-completions",
10764
+ provider: "opencode-go",
10765
+ baseUrl: "https://opencode.ai/zen/go/v1",
10766
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
10767
+ reasoning: true,
10768
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10769
+ input: ["text", "image"],
10770
+ cost: {
10771
+ input: 0.15,
10772
+ output: 0.6,
10773
+ cacheRead: 0.003,
10774
+ cacheWrite: 0,
10775
+ },
10776
+ contextWindow: 1000000,
10777
+ maxTokens: 384000,
10778
+ },
10097
10779
  "glm-5.1": {
10098
10780
  id: "glm-5.1",
10099
10781
  name: "GLM-5.1",
@@ -10147,16 +10829,16 @@ export const MODELS = {
10147
10829
  },
10148
10830
  "glm-5.3-flash": {
10149
10831
  id: "glm-5.3-flash",
10150
- name: "GLM-5.3-Flash (2x usage)",
10832
+ name: "GLM-5.3-Flash",
10151
10833
  api: "openai-completions",
10152
10834
  provider: "opencode-go",
10153
10835
  baseUrl: "https://opencode.ai/zen/go/v1",
10154
10836
  reasoning: true,
10155
10837
  input: ["text", "image"],
10156
10838
  cost: {
10157
- input: 0.075,
10158
- output: 0.25,
10159
- cacheRead: 0.015,
10839
+ input: 0.15,
10840
+ output: 0.5,
10841
+ cacheRead: 0.03,
10160
10842
  cacheWrite: 0,
10161
10843
  },
10162
10844
  contextWindow: 1000000,
@@ -10403,23 +11085,6 @@ export const MODELS = {
10403
11085
  contextWindow: 1048576,
10404
11086
  maxTokens: 131072,
10405
11087
  },
10406
- "omen-alpha": {
10407
- id: "omen-alpha",
10408
- name: "Omen Alpha",
10409
- api: "openai-completions",
10410
- provider: "opencode-go",
10411
- baseUrl: "https://opencode.ai/zen/go/v1",
10412
- reasoning: true,
10413
- input: ["text", "image"],
10414
- cost: {
10415
- input: 0.2,
10416
- output: 0.66,
10417
- cacheRead: 0.04,
10418
- cacheWrite: 0,
10419
- },
10420
- contextWindow: 500000,
10421
- maxTokens: 128000,
10422
- },
10423
11088
  "qwen3.6-plus": {
10424
11089
  id: "qwen3.6-plus",
10425
11090
  name: "Qwen3.6 Plus",
@@ -11303,13 +11968,13 @@ export const MODELS = {
11303
11968
  reasoning: false,
11304
11969
  input: ["text"],
11305
11970
  cost: {
11306
- input: 0.32,
11307
- output: 0.8899999999999999,
11971
+ input: 0.2574,
11972
+ output: 1.0287,
11308
11973
  cacheRead: 0,
11309
11974
  cacheWrite: 0,
11310
11975
  },
11311
11976
  contextWindow: 163840,
11312
- maxTokens: 16384,
11977
+ maxTokens: 16000,
11313
11978
  },
11314
11979
  "deepseek/deepseek-chat-v3-0324": {
11315
11980
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11320,9 +11985,9 @@ export const MODELS = {
11320
11985
  reasoning: false,
11321
11986
  input: ["text"],
11322
11987
  cost: {
11323
- input: 0.29,
11324
- output: 1.1400000000000001,
11325
- cacheRead: 0.11,
11988
+ input: 0.25,
11989
+ output: 1,
11990
+ cacheRead: 0,
11326
11991
  cacheWrite: 0,
11327
11992
  },
11328
11993
  contextWindow: 163840,
@@ -11441,9 +12106,9 @@ export const MODELS = {
11441
12106
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11442
12107
  input: ["text"],
11443
12108
  cost: {
11444
- input: 0.088606,
11445
- output: 0.177212,
11446
- cacheRead: 0.017721200000000003,
12109
+ input: 0.06678,
12110
+ output: 0.13356,
12111
+ cacheRead: 0.013356,
11447
12112
  cacheWrite: 0,
11448
12113
  },
11449
12114
  contextWindow: 1048576,
@@ -11460,9 +12125,9 @@ export const MODELS = {
11460
12125
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11461
12126
  input: ["text"],
11462
12127
  cost: {
11463
- input: 0.065,
11464
- output: 0.18,
11465
- cacheRead: 0.016,
12128
+ input: 0.04,
12129
+ output: 0.08,
12130
+ cacheRead: 0.008,
11466
12131
  cacheWrite: 0,
11467
12132
  },
11468
12133
  contextWindow: 1310720,
@@ -11479,9 +12144,9 @@ export const MODELS = {
11479
12144
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11480
12145
  input: ["text"],
11481
12146
  cost: {
11482
- input: 0.14,
11483
- output: 0.28,
11484
- cacheRead: 0.03,
12147
+ input: 0.11,
12148
+ output: 0.33,
12149
+ cacheRead: 0.0035,
11485
12150
  cacheWrite: 0,
11486
12151
  },
11487
12152
  contextWindow: 1048576,
@@ -11498,13 +12163,32 @@ export const MODELS = {
11498
12163
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11499
12164
  input: ["text", "image"],
11500
12165
  cost: {
11501
- input: 0.44,
11502
- output: 1.32,
11503
- cacheRead: 0.014,
12166
+ input: 0.22,
12167
+ output: 0.66,
12168
+ cacheRead: 0.007,
11504
12169
  cacheWrite: 0,
11505
12170
  },
11506
12171
  contextWindow: 1048576,
11507
- maxTokens: 384000,
12172
+ maxTokens: 943718,
12173
+ },
12174
+ "deepseek/deepseek-v4-flash-vision-exp:batch": {
12175
+ id: "deepseek/deepseek-v4-flash-vision-exp:batch",
12176
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp (batch)",
12177
+ api: "openai-completions",
12178
+ provider: "openrouter",
12179
+ baseUrl: "https://openrouter.ai/api/v1",
12180
+ compat: { "requiresReasoningContentOnAssistantMessages": true },
12181
+ reasoning: true,
12182
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12183
+ input: ["text", "image"],
12184
+ cost: {
12185
+ input: 0.11,
12186
+ output: 0.33,
12187
+ cacheRead: 0.0035,
12188
+ cacheWrite: 0,
12189
+ },
12190
+ contextWindow: 1048576,
12191
+ maxTokens: 943718,
11508
12192
  },
11509
12193
  "deepseek/deepseek-v4-pro": {
11510
12194
  id: "deepseek/deepseek-v4-pro",
@@ -11517,9 +12201,9 @@ export const MODELS = {
11517
12201
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11518
12202
  input: ["text"],
11519
12203
  cost: {
11520
- input: 0.9552599999999999,
11521
- output: 1.9105199999999998,
11522
- cacheRead: 0.07960500000000001,
12204
+ input: 0.788916,
12205
+ output: 1.577832,
12206
+ cacheRead: 0.065743,
11523
12207
  cacheWrite: 0,
11524
12208
  },
11525
12209
  contextWindow: 1048576,
@@ -11536,13 +12220,13 @@ export const MODELS = {
11536
12220
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11537
12221
  input: ["text"],
11538
12222
  cost: {
11539
- input: 1.0494,
11540
- output: 3.1482,
11541
- cacheRead: 0.03498,
12223
+ input: 0.57816,
12224
+ output: 1.73448,
12225
+ cacheRead: 0.018396000000000003,
11542
12226
  cacheWrite: 0,
11543
12227
  },
11544
12228
  contextWindow: 1048576,
11545
- maxTokens: 384000,
12229
+ maxTokens: 393216,
11546
12230
  },
11547
12231
  "deepseek/deepseek-v4-pro-0813:batch": {
11548
12232
  id: "deepseek/deepseek-v4-pro-0813:batch",
@@ -11555,14 +12239,33 @@ export const MODELS = {
11555
12239
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11556
12240
  input: ["text"],
11557
12241
  cost: {
11558
- input: 1.32,
11559
- output: 3.9600000000000004,
11560
- cacheRead: 0.13,
12242
+ input: 0.66,
12243
+ output: 1.9800000000000002,
12244
+ cacheRead: 0.022,
11561
12245
  cacheWrite: 0,
11562
12246
  },
11563
12247
  contextWindow: 1048576,
11564
12248
  maxTokens: 943718,
11565
12249
  },
12250
+ "deepseek/deepseek-v4.1-flash": {
12251
+ id: "deepseek/deepseek-v4.1-flash",
12252
+ name: "DeepSeek: DeepSeek V4.1 Flash",
12253
+ api: "openai-completions",
12254
+ provider: "openrouter",
12255
+ baseUrl: "https://openrouter.ai/api/v1",
12256
+ compat: { "requiresReasoningContentOnAssistantMessages": true },
12257
+ reasoning: true,
12258
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12259
+ input: ["text", "image"],
12260
+ cost: {
12261
+ input: 0.15,
12262
+ output: 0.6,
12263
+ cacheRead: 0.003,
12264
+ cacheWrite: 0,
12265
+ },
12266
+ contextWindow: 1048576,
12267
+ maxTokens: 384000,
12268
+ },
11566
12269
  "dots-studio/dots-3-note-preview:free": {
11567
12270
  id: "dots-studio/dots-3-note-preview:free",
11568
12271
  name: "Dots Studio: Dots3-Note Preview (free)",
@@ -12082,13 +12785,13 @@ export const MODELS = {
12082
12785
  reasoning: true,
12083
12786
  input: ["text", "image"],
12084
12787
  cost: {
12085
- input: 0.07,
12086
- output: 0.33999999999999997,
12788
+ input: 0.041999999999999996,
12789
+ output: 0.22,
12087
12790
  cacheRead: 0,
12088
12791
  cacheWrite: 0,
12089
12792
  },
12090
12793
  contextWindow: 262144,
12091
- maxTokens: 16384,
12794
+ maxTokens: 32768,
12092
12795
  },
12093
12796
  "google/gemma-4-26b-a4b-it:free": {
12094
12797
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12167,9 +12870,9 @@ export const MODELS = {
12167
12870
  reasoning: true,
12168
12871
  input: ["text"],
12169
12872
  cost: {
12170
- input: 0.09999999999999999,
12171
- output: 0.15,
12172
- cacheRead: 0.049999999999999996,
12873
+ input: 0.06,
12874
+ output: 0.25,
12875
+ cacheRead: 0.015,
12173
12876
  cacheWrite: 0,
12174
12877
  },
12175
12878
  contextWindow: 131072,
@@ -12193,9 +12896,9 @@ export const MODELS = {
12193
12896
  contextWindow: 128000,
12194
12897
  maxTokens: 50000,
12195
12898
  },
12196
- "inception/mercury-2.5-preview": {
12197
- id: "inception/mercury-2.5-preview",
12198
- name: "Inception: Mercury 2.5 Preview",
12899
+ "inception/mercury-2.5": {
12900
+ id: "inception/mercury-2.5",
12901
+ name: "Inception: Mercury 2.5",
12199
12902
  api: "openai-completions",
12200
12903
  provider: "openrouter",
12201
12904
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12279,6 +12982,40 @@ export const MODELS = {
12279
12982
  contextWindow: 262144,
12280
12983
  maxTokens: 32768,
12281
12984
  },
12985
+ "inclusionai/ling-3.0-flash-vl": {
12986
+ id: "inclusionai/ling-3.0-flash-vl",
12987
+ name: "inclusionAI: Ling 3.0 Flash VL",
12988
+ api: "openai-completions",
12989
+ provider: "openrouter",
12990
+ baseUrl: "https://openrouter.ai/api/v1",
12991
+ reasoning: true,
12992
+ input: ["text", "image"],
12993
+ cost: {
12994
+ input: 0.06,
12995
+ output: 0.18,
12996
+ cacheRead: 0.012,
12997
+ cacheWrite: 0,
12998
+ },
12999
+ contextWindow: 131072,
13000
+ maxTokens: 32768,
13001
+ },
13002
+ "inclusionai/ling-3.0-flash-vl:free": {
13003
+ id: "inclusionai/ling-3.0-flash-vl:free",
13004
+ name: "inclusionAI: Ling 3.0 Flash VL (free)",
13005
+ api: "openai-completions",
13006
+ provider: "openrouter",
13007
+ baseUrl: "https://openrouter.ai/api/v1",
13008
+ reasoning: true,
13009
+ input: ["text", "image"],
13010
+ cost: {
13011
+ input: 0,
13012
+ output: 0,
13013
+ cacheRead: 0,
13014
+ cacheWrite: 0,
13015
+ },
13016
+ contextWindow: 262144,
13017
+ maxTokens: 32768,
13018
+ },
12282
13019
  "kwaipilot/kat-coder-pro-v2": {
12283
13020
  id: "kwaipilot/kat-coder-pro-v2",
12284
13021
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -12356,13 +13093,13 @@ export const MODELS = {
12356
13093
  reasoning: false,
12357
13094
  input: ["text"],
12358
13095
  cost: {
12359
- input: 0.39999999999999997,
12360
- output: 0.39999999999999997,
13096
+ input: 0.72,
13097
+ output: 0.72,
12361
13098
  cacheRead: 0,
12362
13099
  cacheWrite: 0,
12363
13100
  },
12364
13101
  contextWindow: 131072,
12365
- maxTokens: 16384,
13102
+ maxTokens: 8192,
12366
13103
  },
12367
13104
  "meta-llama/llama-3.1-8b-instruct": {
12368
13105
  id: "meta-llama/llama-3.1-8b-instruct",
@@ -12458,9 +13195,9 @@ export const MODELS = {
12458
13195
  reasoning: true,
12459
13196
  input: ["text", "image"],
12460
13197
  cost: {
12461
- input: 0.35,
12462
- output: 1.5,
12463
- cacheRead: 0.04,
13198
+ input: 0.175,
13199
+ output: 0.75,
13200
+ cacheRead: 0.02,
12464
13201
  cacheWrite: 0,
12465
13202
  },
12466
13203
  contextWindow: 131072,
@@ -12560,7 +13297,7 @@ export const MODELS = {
12560
13297
  reasoning: true,
12561
13298
  input: ["text"],
12562
13299
  cost: {
12563
- input: 0.39999999999999997,
13300
+ input: 0.55,
12564
13301
  output: 2.2,
12565
13302
  cacheRead: 0,
12566
13303
  cacheWrite: 0,
@@ -12687,6 +13424,23 @@ export const MODELS = {
12687
13424
  contextWindow: 256000,
12688
13425
  maxTokens: 204800,
12689
13426
  },
13427
+ "mistralai/codestral-2508:batch": {
13428
+ id: "mistralai/codestral-2508:batch",
13429
+ name: "Mistral: Codestral 2508 (batch)",
13430
+ api: "openai-completions",
13431
+ provider: "openrouter",
13432
+ baseUrl: "https://openrouter.ai/api/v1",
13433
+ reasoning: false,
13434
+ input: ["text"],
13435
+ cost: {
13436
+ input: 0.15,
13437
+ output: 0.44999999999999996,
13438
+ cacheRead: 0.015,
13439
+ cacheWrite: 0,
13440
+ },
13441
+ contextWindow: 256000,
13442
+ maxTokens: 204800,
13443
+ },
12690
13444
  "mistralai/devstral-2512": {
12691
13445
  id: "mistralai/devstral-2512",
12692
13446
  name: "Mistral: Devstral 2 2512",
@@ -12755,6 +13509,23 @@ export const MODELS = {
12755
13509
  contextWindow: 262144,
12756
13510
  maxTokens: 209715,
12757
13511
  },
13512
+ "mistralai/ministral-8b-2512:batch": {
13513
+ id: "mistralai/ministral-8b-2512:batch",
13514
+ name: "Mistral: Ministral 3 8B 2512 (batch)",
13515
+ api: "openai-completions",
13516
+ provider: "openrouter",
13517
+ baseUrl: "https://openrouter.ai/api/v1",
13518
+ reasoning: false,
13519
+ input: ["text", "image"],
13520
+ cost: {
13521
+ input: 0.075,
13522
+ output: 0.075,
13523
+ cacheRead: 0.0075,
13524
+ cacheWrite: 0,
13525
+ },
13526
+ contextWindow: 262144,
13527
+ maxTokens: 209715,
13528
+ },
12758
13529
  "mistralai/mistral-large": {
12759
13530
  id: "mistralai/mistral-large",
12760
13531
  name: "Mistral Large",
@@ -12806,6 +13577,23 @@ export const MODELS = {
12806
13577
  contextWindow: 262144,
12807
13578
  maxTokens: 209715,
12808
13579
  },
13580
+ "mistralai/mistral-large-2512:batch": {
13581
+ id: "mistralai/mistral-large-2512:batch",
13582
+ name: "Mistral: Mistral Large 3 2512 (batch)",
13583
+ api: "openai-completions",
13584
+ provider: "openrouter",
13585
+ baseUrl: "https://openrouter.ai/api/v1",
13586
+ reasoning: false,
13587
+ input: ["text", "image"],
13588
+ cost: {
13589
+ input: 0.25,
13590
+ output: 0.75,
13591
+ cacheRead: 0.024999999999999998,
13592
+ cacheWrite: 0,
13593
+ },
13594
+ contextWindow: 262144,
13595
+ maxTokens: 209715,
13596
+ },
12809
13597
  "mistralai/mistral-medium-3": {
12810
13598
  id: "mistralai/mistral-medium-3",
12811
13599
  name: "Mistral: Mistral Medium 3",
@@ -12874,6 +13662,23 @@ export const MODELS = {
12874
13662
  contextWindow: 131072,
12875
13663
  maxTokens: 104857,
12876
13664
  },
13665
+ "mistralai/mistral-medium-3.1:batch": {
13666
+ id: "mistralai/mistral-medium-3.1:batch",
13667
+ name: "Mistral: Mistral Medium 3.1 (batch)",
13668
+ api: "openai-completions",
13669
+ provider: "openrouter",
13670
+ baseUrl: "https://openrouter.ai/api/v1",
13671
+ reasoning: false,
13672
+ input: ["text", "image"],
13673
+ cost: {
13674
+ input: 0.19999999999999998,
13675
+ output: 1,
13676
+ cacheRead: 0.02,
13677
+ cacheWrite: 0,
13678
+ },
13679
+ contextWindow: 131072,
13680
+ maxTokens: 104857,
13681
+ },
12877
13682
  "mistralai/mistral-nemo": {
12878
13683
  id: "mistralai/mistral-nemo",
12879
13684
  name: "Mistral: Mistral Nemo",
@@ -12925,6 +13730,23 @@ export const MODELS = {
12925
13730
  contextWindow: 262144,
12926
13731
  maxTokens: 209715,
12927
13732
  },
13733
+ "mistralai/mistral-small-2603:batch": {
13734
+ id: "mistralai/mistral-small-2603:batch",
13735
+ name: "Mistral: Mistral Small 4 (batch)",
13736
+ api: "openai-completions",
13737
+ provider: "openrouter",
13738
+ baseUrl: "https://openrouter.ai/api/v1",
13739
+ reasoning: true,
13740
+ input: ["text", "image"],
13741
+ cost: {
13742
+ input: 0.075,
13743
+ output: 0.3,
13744
+ cacheRead: 0.0075,
13745
+ cacheWrite: 0,
13746
+ },
13747
+ contextWindow: 262144,
13748
+ maxTokens: 209715,
13749
+ },
12928
13750
  "mistralai/mistral-small-3.2-24b-instruct": {
12929
13751
  id: "mistralai/mistral-small-3.2-24b-instruct",
12930
13752
  name: "Mistral: Mistral Small 3.2 24B",
@@ -12939,7 +13761,7 @@ export const MODELS = {
12939
13761
  cacheRead: 0,
12940
13762
  cacheWrite: 0,
12941
13763
  },
12942
- contextWindow: 131072,
13764
+ contextWindow: 256000,
12943
13765
  maxTokens: 16384,
12944
13766
  },
12945
13767
  "mistralai/mixtral-8x22b-instruct": {
@@ -13088,9 +13910,9 @@ export const MODELS = {
13088
13910
  reasoning: true,
13089
13911
  input: ["text", "image"],
13090
13912
  cost: {
13091
- input: 3,
13092
- output: 15,
13093
- cacheRead: 0.3,
13913
+ input: 2.30272875,
13914
+ output: 11.550195,
13915
+ cacheRead: 0.26316900000000004,
13094
13916
  cacheWrite: 0,
13095
13917
  },
13096
13918
  contextWindow: 1048576,
@@ -13113,35 +13935,35 @@ export const MODELS = {
13113
13935
  contextWindow: 1048576,
13114
13936
  maxTokens: 943718,
13115
13937
  },
13116
- "nex-agi/nex-n2-mini": {
13117
- id: "nex-agi/nex-n2-mini",
13118
- name: "Nex AGI: Nex-N2-Mini",
13938
+ "nex-agi/nex-n2.5-mini:free": {
13939
+ id: "nex-agi/nex-n2.5-mini:free",
13940
+ name: "Nex AGI: Nex-N2.5-Mini (free)",
13119
13941
  api: "openai-completions",
13120
13942
  provider: "openrouter",
13121
13943
  baseUrl: "https://openrouter.ai/api/v1",
13122
13944
  reasoning: true,
13123
13945
  input: ["text", "image"],
13124
13946
  cost: {
13125
- input: 0.024999999999999998,
13126
- output: 0.09999999999999999,
13127
- cacheRead: 0.0025,
13947
+ input: 0,
13948
+ output: 0,
13949
+ cacheRead: 0,
13128
13950
  cacheWrite: 0,
13129
13951
  },
13130
13952
  contextWindow: 262144,
13131
13953
  maxTokens: 235929,
13132
13954
  },
13133
- "nex-agi/nex-n2-pro": {
13134
- id: "nex-agi/nex-n2-pro",
13135
- name: "Nex AGI: Nex-N2-Pro",
13955
+ "nex-agi/nex-n2.5-pro:free": {
13956
+ id: "nex-agi/nex-n2.5-pro:free",
13957
+ name: "Nex AGI: Nex-N2.5-Pro (free)",
13136
13958
  api: "openai-completions",
13137
13959
  provider: "openrouter",
13138
13960
  baseUrl: "https://openrouter.ai/api/v1",
13139
13961
  reasoning: true,
13140
13962
  input: ["text", "image"],
13141
13963
  cost: {
13142
- input: 0.25,
13143
- output: 1,
13144
- cacheRead: 0.024999999999999998,
13964
+ input: 0,
13965
+ output: 0,
13966
+ cacheRead: 0,
13145
13967
  cacheWrite: 0,
13146
13968
  },
13147
13969
  contextWindow: 262144,
@@ -13195,7 +14017,7 @@ export const MODELS = {
13195
14017
  cacheRead: 0,
13196
14018
  cacheWrite: 0,
13197
14019
  },
13198
- contextWindow: 1000000,
14020
+ contextWindow: 262144,
13199
14021
  maxTokens: 16384,
13200
14022
  },
13201
14023
  "nvidia/nemotron-3-super-120b-a12b:free": {
@@ -15027,13 +15849,13 @@ export const MODELS = {
15027
15849
  reasoning: false,
15028
15850
  input: ["text"],
15029
15851
  cost: {
15030
- input: 0.09,
15031
- output: 0.55,
15032
- cacheRead: 0,
15852
+ input: 0.0875,
15853
+ output: 0.35,
15854
+ cacheRead: 0.0175,
15033
15855
  cacheWrite: 0,
15034
15856
  },
15035
15857
  contextWindow: 262144,
15036
- maxTokens: 16384,
15858
+ maxTokens: 235929,
15037
15859
  },
15038
15860
  "qwen/qwen3-235b-a22b-thinking-2507": {
15039
15861
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -15078,13 +15900,13 @@ export const MODELS = {
15078
15900
  reasoning: false,
15079
15901
  input: ["text"],
15080
15902
  cost: {
15081
- input: 0.04815,
15082
- output: 0.19305,
15903
+ input: 0.09,
15904
+ output: 0.3,
15083
15905
  cacheRead: 0,
15084
15906
  cacheWrite: 0,
15085
15907
  },
15086
15908
  contextWindow: 262144,
15087
- maxTokens: 32000,
15909
+ maxTokens: 235929,
15088
15910
  },
15089
15911
  "qwen/qwen3-30b-a3b-thinking-2507": {
15090
15912
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15265,13 +16087,13 @@ export const MODELS = {
15265
16087
  reasoning: false,
15266
16088
  input: ["text"],
15267
16089
  cost: {
15268
- input: 0.09999999999999999,
16090
+ input: 0.09,
15269
16091
  output: 1.1,
15270
- cacheRead: 0.07,
16092
+ cacheRead: 0,
15271
16093
  cacheWrite: 0,
15272
16094
  },
15273
16095
  contextWindow: 262144,
15274
- maxTokens: 235929,
16096
+ maxTokens: 16384,
15275
16097
  },
15276
16098
  "qwen/qwen3-next-80b-a3b-thinking": {
15277
16099
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -15418,13 +16240,13 @@ export const MODELS = {
15418
16240
  reasoning: true,
15419
16241
  input: ["text", "image"],
15420
16242
  cost: {
15421
- input: 0.29,
15422
- output: 2.4,
16243
+ input: 0.26,
16244
+ output: 2.08,
15423
16245
  cacheRead: 0,
15424
16246
  cacheWrite: 0,
15425
16247
  },
15426
16248
  contextWindow: 262144,
15427
- maxTokens: 81920,
16249
+ maxTokens: 65536,
15428
16250
  },
15429
16251
  "qwen/qwen3.5-27b": {
15430
16252
  id: "qwen/qwen3.5-27b",
@@ -15469,13 +16291,13 @@ export const MODELS = {
15469
16291
  reasoning: true,
15470
16292
  input: ["text", "image"],
15471
16293
  cost: {
15472
- input: 0.39,
15473
- output: 2.34,
15474
- cacheRead: 0,
16294
+ input: 0.55,
16295
+ output: 3.5,
16296
+ cacheRead: 0.22499999999999998,
15475
16297
  cacheWrite: 0,
15476
16298
  },
15477
16299
  contextWindow: 262144,
15478
- maxTokens: 65536,
16300
+ maxTokens: 235929,
15479
16301
  },
15480
16302
  "qwen/qwen3.5-9b": {
15481
16303
  id: "qwen/qwen3.5-9b",
@@ -15713,7 +16535,7 @@ export const MODELS = {
15713
16535
  cacheWrite: 0,
15714
16536
  },
15715
16537
  contextWindow: 1048576,
15716
- maxTokens: 262144,
16538
+ maxTokens: 131072,
15717
16539
  },
15718
16540
  "qwen/qwen3.8-2.4t-a95b:batch": {
15719
16541
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15741,9 +16563,9 @@ export const MODELS = {
15741
16563
  reasoning: true,
15742
16564
  input: ["text", "image"],
15743
16565
  cost: {
15744
- input: 0.42,
15745
- output: 3,
15746
- cacheRead: 0.08499999999999999,
16566
+ input: 0.21400000000000002,
16567
+ output: 2.5500000000000003,
16568
+ cacheRead: 0.15,
15747
16569
  cacheWrite: 0,
15748
16570
  },
15749
16571
  contextWindow: 1000000,
@@ -15814,12 +16636,46 @@ export const MODELS = {
15814
16636
  cacheRead: 0,
15815
16637
  cacheWrite: 0,
15816
16638
  },
15817
- contextWindow: 256000,
16639
+ contextWindow: 256000,
16640
+ maxTokens: 128000,
16641
+ },
16642
+ "sakana/fugu-max": {
16643
+ id: "sakana/fugu-max",
16644
+ name: "Sakana: Fugu Max",
16645
+ api: "openai-completions",
16646
+ provider: "openrouter",
16647
+ baseUrl: "https://openrouter.ai/api/v1",
16648
+ reasoning: true,
16649
+ input: ["text", "image"],
16650
+ cost: {
16651
+ input: 2,
16652
+ output: 6,
16653
+ cacheRead: 0.25,
16654
+ cacheWrite: 0,
16655
+ },
16656
+ contextWindow: 1000000,
16657
+ maxTokens: 128000,
16658
+ },
16659
+ "sakana/fugu-ultra": {
16660
+ id: "sakana/fugu-ultra",
16661
+ name: "Sakana: Fugu Ultra",
16662
+ api: "openai-completions",
16663
+ provider: "openrouter",
16664
+ baseUrl: "https://openrouter.ai/api/v1",
16665
+ reasoning: true,
16666
+ input: ["text", "image"],
16667
+ cost: {
16668
+ input: 5,
16669
+ output: 30,
16670
+ cacheRead: 0.5,
16671
+ cacheWrite: 0,
16672
+ },
16673
+ contextWindow: 1000000,
15818
16674
  maxTokens: 128000,
15819
16675
  },
15820
- "sakana/fugu-ultra": {
15821
- id: "sakana/fugu-ultra",
15822
- name: "Sakana: Fugu Ultra",
16676
+ "sakana/fugu-ultra-v2": {
16677
+ id: "sakana/fugu-ultra-v2",
16678
+ name: "Sakana: Fugu Ultra v2",
15823
16679
  api: "openai-completions",
15824
16680
  provider: "openrouter",
15825
16681
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16098,9 +16954,9 @@ export const MODELS = {
16098
16954
  reasoning: true,
16099
16955
  input: ["text"],
16100
16956
  cost: {
16101
- input: 0.03,
16102
- output: 0.12,
16103
- cacheRead: 0.006,
16957
+ input: 0.09,
16958
+ output: 0.36,
16959
+ cacheRead: 0.018,
16104
16960
  cacheWrite: 0,
16105
16961
  },
16106
16962
  contextWindow: 524288,
@@ -16302,13 +17158,13 @@ export const MODELS = {
16302
17158
  reasoning: true,
16303
17159
  input: ["text"],
16304
17160
  cost: {
16305
- input: 0.55,
16306
- output: 2.2,
16307
- cacheRead: 0.11,
17161
+ input: 0.43,
17162
+ output: 1.75,
17163
+ cacheRead: 0.08,
16308
17164
  cacheWrite: 0,
16309
17165
  },
16310
17166
  contextWindow: 204800,
16311
- maxTokens: 131072,
17167
+ maxTokens: 16384,
16312
17168
  },
16313
17169
  "z-ai/glm-4.6v": {
16314
17170
  id: "z-ai/glm-4.6v",
@@ -16358,7 +17214,7 @@ export const MODELS = {
16358
17214
  cacheRead: 0,
16359
17215
  cacheWrite: 0,
16360
17216
  },
16361
- contextWindow: 202752,
17217
+ contextWindow: 200000,
16362
17218
  maxTokens: 117964,
16363
17219
  },
16364
17220
  "z-ai/glm-5": {
@@ -16421,13 +17277,30 @@ export const MODELS = {
16421
17277
  reasoning: true,
16422
17278
  input: ["text"],
16423
17279
  cost: {
16424
- input: 0.966,
16425
- output: 3.036,
16426
- cacheRead: 0.1932,
17280
+ input: 0.6,
17281
+ output: 2,
17282
+ cacheRead: 0.15,
16427
17283
  cacheWrite: 0,
16428
17284
  },
16429
17285
  contextWindow: 1048576,
16430
- maxTokens: 131072,
17286
+ maxTokens: 182476,
17287
+ },
17288
+ "z-ai/glm-5.2:batch": {
17289
+ id: "z-ai/glm-5.2:batch",
17290
+ name: "Z.ai: GLM 5.2 (batch)",
17291
+ api: "openai-completions",
17292
+ provider: "openrouter",
17293
+ baseUrl: "https://openrouter.ai/api/v1",
17294
+ reasoning: true,
17295
+ input: ["text"],
17296
+ cost: {
17297
+ input: 0.7,
17298
+ output: 2.2,
17299
+ cacheRead: 0.07,
17300
+ cacheWrite: 0,
17301
+ },
17302
+ contextWindow: 1048576,
17303
+ maxTokens: 943718,
16431
17304
  },
16432
17305
  "z-ai/glm-5.3": {
16433
17306
  id: "z-ai/glm-5.3",
@@ -16472,13 +17345,30 @@ export const MODELS = {
16472
17345
  reasoning: true,
16473
17346
  input: ["text", "image"],
16474
17347
  cost: {
16475
- input: 0.15,
16476
- output: 0.5,
16477
- cacheRead: 0.03,
17348
+ input: 0.075,
17349
+ output: 0.25,
17350
+ cacheRead: 0.015,
16478
17351
  cacheWrite: 0,
16479
17352
  },
16480
- contextWindow: 1048575,
16481
- maxTokens: 943717,
17353
+ contextWindow: 1048576,
17354
+ maxTokens: 943718,
17355
+ },
17356
+ "z-ai/glm-5.3:batch": {
17357
+ id: "z-ai/glm-5.3:batch",
17358
+ name: "Z.ai: GLM 5.3 (batch)",
17359
+ api: "openai-completions",
17360
+ provider: "openrouter",
17361
+ baseUrl: "https://openrouter.ai/api/v1",
17362
+ reasoning: true,
17363
+ input: ["text"],
17364
+ cost: {
17365
+ input: 0.7,
17366
+ output: 2.2,
17367
+ cacheRead: 0.13,
17368
+ cacheWrite: 0,
17369
+ },
17370
+ contextWindow: 1048576,
17371
+ maxTokens: 943718,
16482
17372
  },
16483
17373
  "z-ai/glm-5v-turbo": {
16484
17374
  id: "z-ai/glm-5v-turbo",
@@ -16516,7 +17406,7 @@ export const MODELS = {
16516
17406
  },
16517
17407
  "~anthropic/claude-haiku-latest": {
16518
17408
  id: "~anthropic/claude-haiku-latest",
16519
- name: "Anthropic Claude Haiku Latest",
17409
+ name: "Anthropic: Claude Haiku Latest",
16520
17410
  api: "openai-completions",
16521
17411
  provider: "openrouter",
16522
17412
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16550,7 +17440,7 @@ export const MODELS = {
16550
17440
  },
16551
17441
  "~anthropic/claude-sonnet-latest": {
16552
17442
  id: "~anthropic/claude-sonnet-latest",
16553
- name: "Anthropic Claude Sonnet Latest",
17443
+ name: "Anthropic: Claude Sonnet Latest",
16554
17444
  api: "openai-completions",
16555
17445
  provider: "openrouter",
16556
17446
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16567,7 +17457,7 @@ export const MODELS = {
16567
17457
  },
16568
17458
  "~deepseek/deepseek-v4-flash-latest": {
16569
17459
  id: "~deepseek/deepseek-v4-flash-latest",
16570
- name: "DeepSeek V4 Flash Latest",
17460
+ name: "DeepSeek: DeepSeek V4 Flash Latest",
16571
17461
  api: "openai-completions",
16572
17462
  provider: "openrouter",
16573
17463
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16576,9 +17466,9 @@ export const MODELS = {
16576
17466
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
16577
17467
  input: ["text"],
16578
17468
  cost: {
16579
- input: 0.049999999999999996,
16580
- output: 0.16,
16581
- cacheRead: 0.013000000000000001,
17469
+ input: 0.03,
17470
+ output: 0.07,
17471
+ cacheRead: 0.003,
16582
17472
  cacheWrite: 0,
16583
17473
  },
16584
17474
  contextWindow: 1310720,
@@ -16586,7 +17476,7 @@ export const MODELS = {
16586
17476
  },
16587
17477
  "~google/gemini-flash-latest": {
16588
17478
  id: "~google/gemini-flash-latest",
16589
- name: "Google Gemini Flash Latest",
17479
+ name: "Google: Gemini Flash Latest",
16590
17480
  api: "openai-completions",
16591
17481
  provider: "openrouter",
16592
17482
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16603,7 +17493,7 @@ export const MODELS = {
16603
17493
  },
16604
17494
  "~google/gemini-pro-latest": {
16605
17495
  id: "~google/gemini-pro-latest",
16606
- name: "Google Gemini Pro Latest",
17496
+ name: "Google: Gemini Pro Latest",
16607
17497
  api: "openai-completions",
16608
17498
  provider: "openrouter",
16609
17499
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16620,41 +17510,58 @@ export const MODELS = {
16620
17510
  },
16621
17511
  "~moonshotai/kimi-latest": {
16622
17512
  id: "~moonshotai/kimi-latest",
16623
- name: "MoonshotAI Kimi Latest",
17513
+ name: "MoonshotAI: Kimi Latest",
16624
17514
  api: "openai-completions",
16625
17515
  provider: "openrouter",
16626
17516
  baseUrl: "https://openrouter.ai/api/v1",
16627
17517
  reasoning: true,
16628
17518
  input: ["text", "image"],
16629
17519
  cost: {
16630
- input: 2.5,
16631
- output: 14,
16632
- cacheRead: 0.29,
17520
+ input: 2.125,
17521
+ output: 11.9,
17522
+ cacheRead: 0.2465,
16633
17523
  cacheWrite: 0,
16634
17524
  },
16635
17525
  contextWindow: 1048576,
16636
17526
  maxTokens: 943718,
16637
17527
  },
16638
- "~openai/gpt-latest": {
16639
- id: "~openai/gpt-latest",
16640
- name: "OpenAI GPT Latest",
17528
+ "~openai/gpt-astra-latest": {
17529
+ id: "~openai/gpt-astra-latest",
17530
+ name: "OpenAI: GPT Astra Latest",
16641
17531
  api: "openai-completions",
16642
17532
  provider: "openrouter",
16643
17533
  baseUrl: "https://openrouter.ai/api/v1",
16644
17534
  reasoning: true,
16645
17535
  input: ["text", "image"],
16646
17536
  cost: {
16647
- input: 2,
16648
- output: 10,
16649
- cacheRead: 0.19999999999999998,
16650
- cacheWrite: 2.5,
17537
+ input: 10,
17538
+ output: 50,
17539
+ cacheRead: 1,
17540
+ cacheWrite: 12.5,
17541
+ },
17542
+ contextWindow: 1050000,
17543
+ maxTokens: 128000,
17544
+ },
17545
+ "~openai/gpt-luna-latest": {
17546
+ id: "~openai/gpt-luna-latest",
17547
+ name: "OpenAI: GPT Luna Latest",
17548
+ api: "openai-completions",
17549
+ provider: "openrouter",
17550
+ baseUrl: "https://openrouter.ai/api/v1",
17551
+ reasoning: true,
17552
+ input: ["text", "image"],
17553
+ cost: {
17554
+ input: 0.19999999999999998,
17555
+ output: 1.2,
17556
+ cacheRead: 0.02,
17557
+ cacheWrite: 0.25,
16651
17558
  },
16652
17559
  contextWindow: 1050000,
16653
17560
  maxTokens: 128000,
16654
17561
  },
16655
17562
  "~openai/gpt-mini-latest": {
16656
17563
  id: "~openai/gpt-mini-latest",
16657
- name: "OpenAI GPT Mini Latest",
17564
+ name: "OpenAI: GPT Mini Latest",
16658
17565
  api: "openai-completions",
16659
17566
  provider: "openrouter",
16660
17567
  baseUrl: "https://openrouter.ai/api/v1",
@@ -16669,6 +17576,40 @@ export const MODELS = {
16669
17576
  contextWindow: 400000,
16670
17577
  maxTokens: 128000,
16671
17578
  },
17579
+ "~openai/gpt-sol-latest": {
17580
+ id: "~openai/gpt-sol-latest",
17581
+ name: "OpenAI: GPT Sol Latest",
17582
+ api: "openai-completions",
17583
+ provider: "openrouter",
17584
+ baseUrl: "https://openrouter.ai/api/v1",
17585
+ reasoning: true,
17586
+ input: ["text", "image"],
17587
+ cost: {
17588
+ input: 2,
17589
+ output: 10,
17590
+ cacheRead: 0.19999999999999998,
17591
+ cacheWrite: 2.5,
17592
+ },
17593
+ contextWindow: 1050000,
17594
+ maxTokens: 128000,
17595
+ },
17596
+ "~openai/gpt-terra-latest": {
17597
+ id: "~openai/gpt-terra-latest",
17598
+ name: "OpenAI: GPT Terra Latest",
17599
+ api: "openai-completions",
17600
+ provider: "openrouter",
17601
+ baseUrl: "https://openrouter.ai/api/v1",
17602
+ reasoning: true,
17603
+ input: ["text", "image"],
17604
+ cost: {
17605
+ input: 2,
17606
+ output: 12,
17607
+ cacheRead: 0.19999999999999998,
17608
+ cacheWrite: 2.5,
17609
+ },
17610
+ contextWindow: 1050000,
17611
+ maxTokens: 128000,
17612
+ },
16672
17613
  "~x-ai/grok-latest": {
16673
17614
  id: "~x-ai/grok-latest",
16674
17615
  name: "xAI: Grok Latest",
@@ -16695,9 +17636,9 @@ export const MODELS = {
16695
17636
  reasoning: true,
16696
17637
  input: ["text", "image"],
16697
17638
  cost: {
16698
- input: 0.07125000000000001,
16699
- output: 0.2375,
16700
- cacheRead: 0.01425,
17639
+ input: 0.075,
17640
+ output: 0.25,
17641
+ cacheRead: 0.015,
16701
17642
  cacheWrite: 0,
16702
17643
  },
16703
17644
  contextWindow: 1310720,
@@ -16712,9 +17653,9 @@ export const MODELS = {
16712
17653
  reasoning: true,
16713
17654
  input: ["text"],
16714
17655
  cost: {
16715
- input: 1.12,
16716
- output: 3.52,
16717
- cacheRead: 0.20800000000000002,
17656
+ input: 0.8727,
17657
+ output: 3.36,
17658
+ cacheRead: 0.16390000000000002,
16718
17659
  cacheWrite: 0,
16719
17660
  },
16720
17661
  contextWindow: 1310720,
@@ -17596,23 +18537,6 @@ export const MODELS = {
17596
18537
  contextWindow: 991000,
17597
18538
  maxTokens: 128000,
17598
18539
  },
17599
- "alibaba/qwen3.8-flash-next": {
17600
- id: "alibaba/qwen3.8-flash-next",
17601
- name: "Qwen 3.8 Flash Next",
17602
- api: "anthropic-messages",
17603
- provider: "vercel-ai-gateway",
17604
- baseUrl: "https://ai-gateway.vercel.sh",
17605
- reasoning: true,
17606
- input: ["text", "image"],
17607
- cost: {
17608
- input: 0.12,
17609
- output: 0.39999999999999997,
17610
- cacheRead: 0.01,
17611
- cacheWrite: 0,
17612
- },
17613
- contextWindow: 1048576,
17614
- maxTokens: 1048576,
17615
- },
17616
18540
  "alibaba/qwen3.8-max": {
17617
18541
  id: "alibaba/qwen3.8-max",
17618
18542
  name: "Qwen 3.8 Max",
@@ -18122,9 +19046,9 @@ export const MODELS = {
18122
19046
  reasoning: false,
18123
19047
  input: ["text"],
18124
19048
  cost: {
18125
- input: 0.28,
18126
- output: 0.42,
18127
- cacheRead: 0.028,
19049
+ input: 0.62,
19050
+ output: 1.85,
19051
+ cacheRead: 0,
18128
19052
  cacheWrite: 0,
18129
19053
  },
18130
19054
  contextWindow: 128000,
@@ -18232,6 +19156,23 @@ export const MODELS = {
18232
19156
  contextWindow: 1000000,
18233
19157
  maxTokens: 384000,
18234
19158
  },
19159
+ "deepseek/deepseek-v4.1-flash": {
19160
+ id: "deepseek/deepseek-v4.1-flash",
19161
+ name: "DeepSeek V4.1 Flash",
19162
+ api: "anthropic-messages",
19163
+ provider: "vercel-ai-gateway",
19164
+ baseUrl: "https://ai-gateway.vercel.sh",
19165
+ reasoning: true,
19166
+ input: ["text", "image"],
19167
+ cost: {
19168
+ input: 0.3,
19169
+ output: 1.2,
19170
+ cacheRead: 0.03,
19171
+ cacheWrite: 0,
19172
+ },
19173
+ contextWindow: 1048576,
19174
+ maxTokens: 32768,
19175
+ },
18235
19176
  "google/gemini-2.5-flash": {
18236
19177
  id: "google/gemini-2.5-flash",
18237
19178
  name: "Gemini 2.5 Flash",
@@ -18470,6 +19411,23 @@ export const MODELS = {
18470
19411
  contextWindow: 128000,
18471
19412
  maxTokens: 128000,
18472
19413
  },
19414
+ "inception/mercury-2.5": {
19415
+ id: "inception/mercury-2.5",
19416
+ name: "Mercury 2.5",
19417
+ api: "anthropic-messages",
19418
+ provider: "vercel-ai-gateway",
19419
+ baseUrl: "https://ai-gateway.vercel.sh",
19420
+ reasoning: true,
19421
+ input: ["text"],
19422
+ cost: {
19423
+ input: 0.04,
19424
+ output: 0.15,
19425
+ cacheRead: 0.004,
19426
+ cacheWrite: 0,
19427
+ },
19428
+ contextWindow: 260000,
19429
+ maxTokens: 65536,
19430
+ },
18473
19431
  "inception/mercury-coder-small": {
18474
19432
  id: "inception/mercury-coder-small",
18475
19433
  name: "Mercury Coder Small Beta",
@@ -18572,6 +19530,40 @@ export const MODELS = {
18572
19530
  contextWindow: 256000,
18573
19531
  maxTokens: 32000,
18574
19532
  },
19533
+ "inclusionai/ling-3.0-flash-vl": {
19534
+ id: "inclusionai/ling-3.0-flash-vl",
19535
+ name: "Ling 3.0 Flash VL",
19536
+ api: "anthropic-messages",
19537
+ provider: "vercel-ai-gateway",
19538
+ baseUrl: "https://ai-gateway.vercel.sh",
19539
+ reasoning: true,
19540
+ input: ["text", "image"],
19541
+ cost: {
19542
+ input: 0,
19543
+ output: 0,
19544
+ cacheRead: 0,
19545
+ cacheWrite: 0,
19546
+ },
19547
+ contextWindow: 256000,
19548
+ maxTokens: 32000,
19549
+ },
19550
+ "inclusionai/ling-3.0-flash-vl-free": {
19551
+ id: "inclusionai/ling-3.0-flash-vl-free",
19552
+ name: "Ling 3.0 Flash VL (Free)",
19553
+ api: "anthropic-messages",
19554
+ provider: "vercel-ai-gateway",
19555
+ baseUrl: "https://ai-gateway.vercel.sh",
19556
+ reasoning: true,
19557
+ input: ["text", "image"],
19558
+ cost: {
19559
+ input: 0,
19560
+ output: 0,
19561
+ cacheRead: 0,
19562
+ cacheWrite: 0,
19563
+ },
19564
+ contextWindow: 256000,
19565
+ maxTokens: 32000,
19566
+ },
18575
19567
  "interfaze/interfaze-beta": {
18576
19568
  id: "interfaze/interfaze-beta",
18577
19569
  name: "Interfaze Beta",
@@ -20465,6 +21457,23 @@ export const MODELS = {
20465
21457
  contextWindow: 256000,
20466
21458
  maxTokens: 32768,
20467
21459
  },
21460
+ "sakana/fugu-max": {
21461
+ id: "sakana/fugu-max",
21462
+ name: "Fugu Max",
21463
+ api: "anthropic-messages",
21464
+ provider: "vercel-ai-gateway",
21465
+ baseUrl: "https://ai-gateway.vercel.sh",
21466
+ reasoning: true,
21467
+ input: ["text", "image"],
21468
+ cost: {
21469
+ input: 2,
21470
+ output: 6,
21471
+ cacheRead: 0.25,
21472
+ cacheWrite: 0,
21473
+ },
21474
+ contextWindow: 1000000,
21475
+ maxTokens: 1000000,
21476
+ },
20468
21477
  "sakana/fugu-ultra": {
20469
21478
  id: "sakana/fugu-ultra",
20470
21479
  name: "Fugu Ultra",
@@ -20482,6 +21491,23 @@ export const MODELS = {
20482
21491
  contextWindow: 1000000,
20483
21492
  maxTokens: 1000000,
20484
21493
  },
21494
+ "sakana/fugu-ultra-v2": {
21495
+ id: "sakana/fugu-ultra-v2",
21496
+ name: "Fugu Ultra v2",
21497
+ api: "anthropic-messages",
21498
+ provider: "vercel-ai-gateway",
21499
+ baseUrl: "https://ai-gateway.vercel.sh",
21500
+ reasoning: true,
21501
+ input: ["text", "image"],
21502
+ cost: {
21503
+ input: 5,
21504
+ output: 30,
21505
+ cacheRead: 0.5,
21506
+ cacheWrite: 0,
21507
+ },
21508
+ contextWindow: 1000000,
21509
+ maxTokens: 1000000,
21510
+ },
20485
21511
  "sakana/namazu": {
20486
21512
  id: "sakana/namazu",
20487
21513
  name: "Sakana Namazu",
@@ -20839,23 +21865,6 @@ export const MODELS = {
20839
21865
  contextWindow: 1050000,
20840
21866
  maxTokens: 131000,
20841
21867
  },
20842
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
20843
- id: "xiaomi/mimo-v2.5-pro-ultraspeed",
20844
- name: "MiMo V2.5 Pro UltraSpeed",
20845
- api: "anthropic-messages",
20846
- provider: "vercel-ai-gateway",
20847
- baseUrl: "https://ai-gateway.vercel.sh",
20848
- reasoning: true,
20849
- input: ["text"],
20850
- cost: {
20851
- input: 1.305,
20852
- output: 2.61,
20853
- cacheRead: 0.0108,
20854
- cacheWrite: 0,
20855
- },
20856
- contextWindow: 1048576,
20857
- maxTokens: 131072,
20858
- },
20859
21868
  "zai/glm-4.5": {
20860
21869
  id: "zai/glm-4.5",
20861
21870
  name: "GLM 4.5",
@@ -21069,9 +22078,9 @@ export const MODELS = {
21069
22078
  reasoning: true,
21070
22079
  input: ["text"],
21071
22080
  cost: {
21072
- input: 0.7,
21073
- output: 2.2,
21074
- cacheRead: 0.13,
22081
+ input: 1.4,
22082
+ output: 4.4,
22083
+ cacheRead: 0.14,
21075
22084
  cacheWrite: 0,
21076
22085
  },
21077
22086
  contextWindow: 1000000,
@@ -21111,23 +22120,6 @@ export const MODELS = {
21111
22120
  contextWindow: 1000000,
21112
22121
  maxTokens: 131000,
21113
22122
  },
21114
- "zai/glm-5.3-promo-50": {
21115
- id: "zai/glm-5.3-promo-50",
21116
- name: "GLM 5.3 (50% off)",
21117
- api: "anthropic-messages",
21118
- provider: "vercel-ai-gateway",
21119
- baseUrl: "https://ai-gateway.vercel.sh",
21120
- reasoning: true,
21121
- input: ["text"],
21122
- cost: {
21123
- input: 0.7,
21124
- output: 2.2,
21125
- cacheRead: 0.13,
21126
- cacheWrite: 0,
21127
- },
21128
- contextWindow: 1048576,
21129
- maxTokens: 1048576,
21130
- },
21131
22123
  "zai/glm-5v-turbo": {
21132
22124
  id: "zai/glm-5v-turbo",
21133
22125
  name: "GLM 5V Turbo",