@elyracode/ai 0.9.37 → 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.
@@ -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",
@@ -342,6 +342,24 @@ export const MODELS = {
342
342
  contextWindow: 300000,
343
343
  maxTokens: 10000,
344
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
+ },
345
363
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
346
364
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
347
365
  name: "Claude Haiku 4.5 (AU)",
@@ -352,10 +370,10 @@ export const MODELS = {
352
370
  thinkingType: "budget",
353
371
  input: ["text", "image"],
354
372
  cost: {
355
- input: 1,
356
- output: 5,
357
- cacheRead: 0.1,
358
- cacheWrite: 1.25,
373
+ input: 1.1,
374
+ output: 5.5,
375
+ cacheRead: 0.11,
376
+ cacheWrite: 1.375,
359
377
  },
360
378
  contextWindow: 200000,
361
379
  maxTokens: 64000,
@@ -371,8 +389,8 @@ export const MODELS = {
371
389
  thinkingLevelMap: { "xhigh": "max" },
372
390
  input: ["text", "image"],
373
391
  cost: {
374
- input: 16.5,
375
- output: 82.5,
392
+ input: 5.5,
393
+ output: 27.5,
376
394
  cacheRead: 0.5,
377
395
  cacheWrite: 6.25,
378
396
  },
@@ -409,10 +427,10 @@ export const MODELS = {
409
427
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
410
428
  input: ["text", "image"],
411
429
  cost: {
412
- input: 5,
413
- output: 25,
414
- cacheRead: 0.5,
415
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
416
434
  },
417
435
  contextWindow: 1000000,
418
436
  maxTokens: 128000,
@@ -427,10 +445,10 @@ export const MODELS = {
427
445
  thinkingType: "adaptive",
428
446
  input: ["text", "image"],
429
447
  cost: {
430
- input: 5,
431
- output: 25,
432
- cacheRead: 0.5,
433
- cacheWrite: 6.25,
448
+ input: 5.5,
449
+ output: 27.5,
450
+ cacheRead: 0.55,
451
+ cacheWrite: 6.875,
434
452
  },
435
453
  contextWindow: 1000000,
436
454
  maxTokens: 128000,
@@ -445,10 +463,10 @@ export const MODELS = {
445
463
  thinkingType: "budget",
446
464
  input: ["text", "image"],
447
465
  cost: {
448
- input: 3,
449
- output: 15,
450
- cacheRead: 0.3,
451
- cacheWrite: 3.75,
466
+ input: 3.3,
467
+ output: 16.5,
468
+ cacheRead: 0.33,
469
+ cacheWrite: 4.125,
452
470
  },
453
471
  contextWindow: 200000,
454
472
  maxTokens: 64000,
@@ -481,10 +499,10 @@ export const MODELS = {
481
499
  thinkingType: "adaptive",
482
500
  input: ["text", "image"],
483
501
  cost: {
484
- input: 2,
485
- output: 10,
486
- cacheRead: 0.2,
487
- cacheWrite: 2.5,
502
+ input: 2.2,
503
+ output: 11,
504
+ cacheRead: 0.22,
505
+ cacheWrite: 2.75,
488
506
  },
489
507
  contextWindow: 1000000,
490
508
  maxTokens: 128000,
@@ -506,23 +524,6 @@ export const MODELS = {
506
524
  contextWindow: 300000,
507
525
  maxTokens: 10000,
508
526
  },
509
- "deepseek.r1-v1:0": {
510
- id: "deepseek.r1-v1:0",
511
- name: "DeepSeek-R1",
512
- api: "bedrock-converse-stream",
513
- provider: "amazon-bedrock",
514
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
515
- reasoning: true,
516
- input: ["text"],
517
- cost: {
518
- input: 1.35,
519
- output: 5.4,
520
- cacheRead: 0,
521
- cacheWrite: 0,
522
- },
523
- contextWindow: 128000,
524
- maxTokens: 32768,
525
- },
526
527
  "deepseek.v3-v1:0": {
527
528
  id: "deepseek.v3-v1:0",
528
529
  name: "DeepSeek-V3.1",
@@ -542,7 +543,7 @@ export const MODELS = {
542
543
  },
543
544
  "deepseek.v3.2": {
544
545
  id: "deepseek.v3.2",
545
- name: "DeepSeek-V3.2",
546
+ name: "DeepSeek V3.2",
546
547
  api: "bedrock-converse-stream",
547
548
  provider: "amazon-bedrock",
548
549
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -755,6 +756,24 @@ export const MODELS = {
755
756
  contextWindow: 1000000,
756
757
  maxTokens: 128000,
757
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
+ },
758
777
  "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {
759
778
  id: "eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
760
779
  name: "Claude Sonnet 4.5 (EU)",
@@ -789,7 +808,7 @@ export const MODELS = {
789
808
  cacheWrite: 4.125,
790
809
  },
791
810
  contextWindow: 1000000,
792
- maxTokens: 64000,
811
+ maxTokens: 128000,
793
812
  },
794
813
  "eu.anthropic.claude-sonnet-5": {
795
814
  id: "eu.anthropic.claude-sonnet-5",
@@ -992,6 +1011,24 @@ export const MODELS = {
992
1011
  contextWindow: 1000000,
993
1012
  maxTokens: 128000,
994
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
+ },
995
1032
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {
996
1033
  id: "global.anthropic.claude-sonnet-4-5-20250929-v1:0",
997
1034
  name: "Claude Sonnet 4.5 (Global)",
@@ -1026,7 +1063,7 @@ export const MODELS = {
1026
1063
  cacheWrite: 3.75,
1027
1064
  },
1028
1065
  contextWindow: 1000000,
1029
- maxTokens: 64000,
1066
+ maxTokens: 128000,
1030
1067
  },
1031
1068
  "global.anthropic.claude-sonnet-5": {
1032
1069
  id: "global.anthropic.claude-sonnet-5",
@@ -1243,10 +1280,10 @@ export const MODELS = {
1243
1280
  thinkingType: "budget",
1244
1281
  input: ["text", "image"],
1245
1282
  cost: {
1246
- input: 1,
1247
- output: 5,
1248
- cacheRead: 0.1,
1249
- cacheWrite: 1.25,
1283
+ input: 1.1,
1284
+ output: 5.5,
1285
+ cacheRead: 0.11,
1286
+ cacheWrite: 1.375,
1250
1287
  },
1251
1288
  contextWindow: 200000,
1252
1289
  maxTokens: 64000,
@@ -1262,10 +1299,10 @@ export const MODELS = {
1262
1299
  thinkingLevelMap: { "xhigh": "xhigh" },
1263
1300
  input: ["text", "image"],
1264
1301
  cost: {
1265
- input: 5,
1266
- output: 25,
1267
- cacheRead: 0.5,
1268
- cacheWrite: 6.25,
1302
+ input: 5.5,
1303
+ output: 27.5,
1304
+ cacheRead: 0.55,
1305
+ cacheWrite: 6.875,
1269
1306
  },
1270
1307
  contextWindow: 1000000,
1271
1308
  maxTokens: 128000,
@@ -1281,10 +1318,10 @@ export const MODELS = {
1281
1318
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
1282
1319
  input: ["text", "image"],
1283
1320
  cost: {
1284
- input: 5,
1285
- output: 25,
1286
- cacheRead: 0.5,
1287
- cacheWrite: 6.25,
1321
+ input: 5.5,
1322
+ output: 27.5,
1323
+ cacheRead: 0.55,
1324
+ cacheWrite: 6.875,
1288
1325
  },
1289
1326
  contextWindow: 1000000,
1290
1327
  maxTokens: 128000,
@@ -1299,10 +1336,10 @@ export const MODELS = {
1299
1336
  thinkingType: "adaptive",
1300
1337
  input: ["text", "image"],
1301
1338
  cost: {
1302
- input: 5,
1303
- output: 25,
1304
- cacheRead: 0.5,
1305
- cacheWrite: 6.25,
1339
+ input: 5.5,
1340
+ output: 27.5,
1341
+ cacheRead: 0.55,
1342
+ cacheWrite: 6.875,
1306
1343
  },
1307
1344
  contextWindow: 1000000,
1308
1345
  maxTokens: 128000,
@@ -1317,10 +1354,10 @@ export const MODELS = {
1317
1354
  thinkingType: "budget",
1318
1355
  input: ["text", "image"],
1319
1356
  cost: {
1320
- input: 3,
1321
- output: 15,
1322
- cacheRead: 0.3,
1323
- cacheWrite: 3.75,
1357
+ input: 3.3,
1358
+ output: 16.5,
1359
+ cacheRead: 0.33,
1360
+ cacheWrite: 4.125,
1324
1361
  },
1325
1362
  contextWindow: 200000,
1326
1363
  maxTokens: 64000,
@@ -1335,13 +1372,13 @@ export const MODELS = {
1335
1372
  thinkingType: "adaptive",
1336
1373
  input: ["text", "image"],
1337
1374
  cost: {
1338
- input: 3,
1339
- output: 15,
1340
- cacheRead: 0.3,
1341
- cacheWrite: 3.75,
1375
+ input: 3.3,
1376
+ output: 16.5,
1377
+ cacheRead: 0.33,
1378
+ cacheWrite: 4.125,
1342
1379
  },
1343
1380
  contextWindow: 1000000,
1344
- maxTokens: 64000,
1381
+ maxTokens: 128000,
1345
1382
  },
1346
1383
  "jp.anthropic.claude-sonnet-5": {
1347
1384
  id: "jp.anthropic.claude-sonnet-5",
@@ -1353,10 +1390,10 @@ export const MODELS = {
1353
1390
  thinkingType: "adaptive",
1354
1391
  input: ["text", "image"],
1355
1392
  cost: {
1356
- input: 2,
1357
- output: 10,
1358
- cacheRead: 0.2,
1359
- cacheWrite: 2.5,
1393
+ input: 2.2,
1394
+ output: 11,
1395
+ cacheRead: 0.22,
1396
+ cacheWrite: 2.75,
1360
1397
  },
1361
1398
  contextWindow: 1000000,
1362
1399
  maxTokens: 128000,
@@ -1427,7 +1464,7 @@ export const MODELS = {
1427
1464
  cacheWrite: 0,
1428
1465
  },
1429
1466
  contextWindow: 1000000,
1430
- maxTokens: 16384,
1467
+ maxTokens: 8192,
1431
1468
  },
1432
1469
  "meta.llama4-scout-17b-instruct-v1:0": {
1433
1470
  id: "meta.llama4-scout-17b-instruct-v1:0",
@@ -1443,12 +1480,12 @@ export const MODELS = {
1443
1480
  cacheRead: 0,
1444
1481
  cacheWrite: 0,
1445
1482
  },
1446
- contextWindow: 3500000,
1447
- maxTokens: 16384,
1483
+ contextWindow: 10000000,
1484
+ maxTokens: 8192,
1448
1485
  },
1449
1486
  "minimax.minimax-m2": {
1450
1487
  id: "minimax.minimax-m2",
1451
- name: "MiniMax M2",
1488
+ name: "MiniMax-M2",
1452
1489
  api: "bedrock-converse-stream",
1453
1490
  provider: "amazon-bedrock",
1454
1491
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1465,7 +1502,7 @@ export const MODELS = {
1465
1502
  },
1466
1503
  "minimax.minimax-m2.1": {
1467
1504
  id: "minimax.minimax-m2.1",
1468
- name: "MiniMax M2.1",
1505
+ name: "MiniMax-M2.1",
1469
1506
  api: "bedrock-converse-stream",
1470
1507
  provider: "amazon-bedrock",
1471
1508
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1482,7 +1519,7 @@ export const MODELS = {
1482
1519
  },
1483
1520
  "minimax.minimax-m2.5": {
1484
1521
  id: "minimax.minimax-m2.5",
1485
- name: "MiniMax M2.5",
1522
+ name: "MiniMax-M2.5",
1486
1523
  api: "bedrock-converse-stream",
1487
1524
  provider: "amazon-bedrock",
1488
1525
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1538,7 +1575,7 @@ export const MODELS = {
1538
1575
  provider: "amazon-bedrock",
1539
1576
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1540
1577
  reasoning: false,
1541
- input: ["text"],
1578
+ input: ["text", "image"],
1542
1579
  cost: {
1543
1580
  input: 0.2,
1544
1581
  output: 0.2,
@@ -1572,7 +1609,7 @@ export const MODELS = {
1572
1609
  provider: "amazon-bedrock",
1573
1610
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1574
1611
  reasoning: false,
1575
- input: ["text"],
1612
+ input: ["text", "image"],
1576
1613
  cost: {
1577
1614
  input: 0.15,
1578
1615
  output: 0.15,
@@ -1630,7 +1667,7 @@ export const MODELS = {
1630
1667
  cacheRead: 0,
1631
1668
  cacheWrite: 0,
1632
1669
  },
1633
- contextWindow: 128000,
1670
+ contextWindow: 32768,
1634
1671
  maxTokens: 4096,
1635
1672
  },
1636
1673
  "mistral.voxtral-small-24b-2507": {
@@ -1642,12 +1679,12 @@ export const MODELS = {
1642
1679
  reasoning: false,
1643
1680
  input: ["text"],
1644
1681
  cost: {
1645
- input: 0.15,
1646
- output: 0.35,
1682
+ input: 0.1,
1683
+ output: 0.3,
1647
1684
  cacheRead: 0,
1648
1685
  cacheWrite: 0,
1649
1686
  },
1650
- contextWindow: 32000,
1687
+ contextWindow: 32768,
1651
1688
  maxTokens: 8192,
1652
1689
  },
1653
1690
  "moonshot.kimi-k2-thinking": {
@@ -1682,7 +1719,7 @@ export const MODELS = {
1682
1719
  cacheWrite: 0,
1683
1720
  },
1684
1721
  contextWindow: 262143,
1685
- maxTokens: 16000,
1722
+ maxTokens: 16384,
1686
1723
  },
1687
1724
  "nvidia.nemotron-nano-12b-v2": {
1688
1725
  id: "nvidia.nemotron-nano-12b-v2",
@@ -1699,7 +1736,7 @@ export const MODELS = {
1699
1736
  cacheWrite: 0,
1700
1737
  },
1701
1738
  contextWindow: 128000,
1702
- maxTokens: 4096,
1739
+ maxTokens: 8192,
1703
1740
  },
1704
1741
  "nvidia.nemotron-nano-3-30b": {
1705
1742
  id: "nvidia.nemotron-nano-3-30b",
@@ -1715,8 +1752,8 @@ export const MODELS = {
1715
1752
  cacheRead: 0,
1716
1753
  cacheWrite: 0,
1717
1754
  },
1718
- contextWindow: 128000,
1719
- maxTokens: 4096,
1755
+ contextWindow: 262144,
1756
+ maxTokens: 8192,
1720
1757
  },
1721
1758
  "nvidia.nemotron-nano-9b-v2": {
1722
1759
  id: "nvidia.nemotron-nano-9b-v2",
@@ -1732,8 +1769,8 @@ export const MODELS = {
1732
1769
  cacheRead: 0,
1733
1770
  cacheWrite: 0,
1734
1771
  },
1735
- contextWindow: 128000,
1736
- maxTokens: 4096,
1772
+ contextWindow: 131072,
1773
+ maxTokens: 8192,
1737
1774
  },
1738
1775
  "nvidia.nemotron-super-3-120b": {
1739
1776
  id: "nvidia.nemotron-super-3-120b",
@@ -1960,7 +1997,7 @@ export const MODELS = {
1960
1997
  },
1961
1998
  "qwen.qwen3-235b-a22b-2507-v1:0": {
1962
1999
  id: "qwen.qwen3-235b-a22b-2507-v1:0",
1963
- name: "Qwen3 235B A22B 2507",
2000
+ name: "Qwen3 235B-A22B Instruct 2507",
1964
2001
  api: "bedrock-converse-stream",
1965
2002
  provider: "amazon-bedrock",
1966
2003
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1977,7 +2014,7 @@ export const MODELS = {
1977
2014
  },
1978
2015
  "qwen.qwen3-32b-v1:0": {
1979
2016
  id: "qwen.qwen3-32b-v1:0",
1980
- name: "Qwen3 32B (dense)",
2017
+ name: "Qwen3 32B",
1981
2018
  api: "bedrock-converse-stream",
1982
2019
  provider: "amazon-bedrock",
1983
2020
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -1989,12 +2026,12 @@ export const MODELS = {
1989
2026
  cacheRead: 0,
1990
2027
  cacheWrite: 0,
1991
2028
  },
1992
- contextWindow: 16384,
2029
+ contextWindow: 32768,
1993
2030
  maxTokens: 16384,
1994
2031
  },
1995
2032
  "qwen.qwen3-coder-30b-a3b-v1:0": {
1996
2033
  id: "qwen.qwen3-coder-30b-a3b-v1:0",
1997
- name: "Qwen3 Coder 30B A3B Instruct",
2034
+ name: "Qwen3-Coder 30B-A3B Instruct",
1998
2035
  api: "bedrock-converse-stream",
1999
2036
  provider: "amazon-bedrock",
2000
2037
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -2011,14 +2048,14 @@ export const MODELS = {
2011
2048
  },
2012
2049
  "qwen.qwen3-coder-480b-a35b-v1:0": {
2013
2050
  id: "qwen.qwen3-coder-480b-a35b-v1:0",
2014
- name: "Qwen3 Coder 480B A35B Instruct",
2051
+ name: "Qwen3-Coder 480B-A35B Instruct",
2015
2052
  api: "bedrock-converse-stream",
2016
2053
  provider: "amazon-bedrock",
2017
2054
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2018
2055
  reasoning: false,
2019
2056
  input: ["text"],
2020
2057
  cost: {
2021
- input: 0.22,
2058
+ input: 0.45,
2022
2059
  output: 1.8,
2023
2060
  cacheRead: 0,
2024
2061
  cacheWrite: 0,
@@ -2032,49 +2069,49 @@ export const MODELS = {
2032
2069
  api: "bedrock-converse-stream",
2033
2070
  provider: "amazon-bedrock",
2034
2071
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2035
- reasoning: true,
2072
+ reasoning: false,
2036
2073
  input: ["text"],
2037
2074
  cost: {
2038
- input: 0.22,
2039
- output: 1.8,
2075
+ input: 0.5,
2076
+ output: 1.2,
2040
2077
  cacheRead: 0,
2041
2078
  cacheWrite: 0,
2042
2079
  },
2043
- contextWindow: 131072,
2080
+ contextWindow: 262144,
2044
2081
  maxTokens: 65536,
2045
2082
  },
2046
2083
  "qwen.qwen3-next-80b-a3b": {
2047
2084
  id: "qwen.qwen3-next-80b-a3b",
2048
- name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
2085
+ name: "Qwen3-Next 80B-A3B Instruct",
2049
2086
  api: "bedrock-converse-stream",
2050
2087
  provider: "amazon-bedrock",
2051
2088
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2052
2089
  reasoning: false,
2053
2090
  input: ["text"],
2054
2091
  cost: {
2055
- input: 0.14,
2056
- output: 1.4,
2092
+ input: 0.15,
2093
+ output: 1.2,
2057
2094
  cacheRead: 0,
2058
2095
  cacheWrite: 0,
2059
2096
  },
2060
- contextWindow: 262000,
2097
+ contextWindow: 262144,
2061
2098
  maxTokens: 262000,
2062
2099
  },
2063
2100
  "qwen.qwen3-vl-235b-a22b": {
2064
2101
  id: "qwen.qwen3-vl-235b-a22b",
2065
- name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
2102
+ name: "Qwen3 VL 235B A22B Instruct",
2066
2103
  api: "bedrock-converse-stream",
2067
2104
  provider: "amazon-bedrock",
2068
2105
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2069
2106
  reasoning: false,
2070
2107
  input: ["text", "image"],
2071
2108
  cost: {
2072
- input: 0.3,
2073
- output: 1.5,
2109
+ input: 0.53,
2110
+ output: 2.66,
2074
2111
  cacheRead: 0,
2075
2112
  cacheWrite: 0,
2076
2113
  },
2077
- contextWindow: 262000,
2114
+ contextWindow: 262144,
2078
2115
  maxTokens: 262000,
2079
2116
  },
2080
2117
  "us-gov.openai.gpt-oss-120b-1:0": {
@@ -2207,10 +2244,10 @@ export const MODELS = {
2207
2244
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
2208
2245
  input: ["text", "image"],
2209
2246
  cost: {
2210
- input: 10,
2211
- output: 50,
2212
- cacheRead: 1,
2213
- cacheWrite: 12.5,
2247
+ input: 11,
2248
+ output: 55,
2249
+ cacheRead: 1.1,
2250
+ cacheWrite: 13.75,
2214
2251
  },
2215
2252
  contextWindow: 1000000,
2216
2253
  maxTokens: 128000,
@@ -2244,10 +2281,10 @@ export const MODELS = {
2244
2281
  thinkingType: "budget",
2245
2282
  input: ["text", "image"],
2246
2283
  cost: {
2247
- input: 1,
2248
- output: 5,
2249
- cacheRead: 0.1,
2250
- cacheWrite: 1.25,
2284
+ input: 1.1,
2285
+ output: 5.5,
2286
+ cacheRead: 0.11,
2287
+ cacheWrite: 1.375,
2251
2288
  },
2252
2289
  contextWindow: 200000,
2253
2290
  maxTokens: 64000,
@@ -2280,10 +2317,10 @@ export const MODELS = {
2280
2317
  thinkingType: "budget",
2281
2318
  input: ["text", "image"],
2282
2319
  cost: {
2283
- input: 5,
2284
- output: 25,
2285
- cacheRead: 0.5,
2286
- cacheWrite: 6.25,
2320
+ input: 5.5,
2321
+ output: 27.5,
2322
+ cacheRead: 0.55,
2323
+ cacheWrite: 6.875,
2287
2324
  },
2288
2325
  contextWindow: 200000,
2289
2326
  maxTokens: 64000,
@@ -2299,8 +2336,8 @@ export const MODELS = {
2299
2336
  thinkingLevelMap: { "xhigh": "max" },
2300
2337
  input: ["text", "image"],
2301
2338
  cost: {
2302
- input: 5,
2303
- output: 25,
2339
+ input: 5.5,
2340
+ output: 27.5,
2304
2341
  cacheRead: 0.5,
2305
2342
  cacheWrite: 6.25,
2306
2343
  },
@@ -2318,10 +2355,10 @@ export const MODELS = {
2318
2355
  thinkingLevelMap: { "xhigh": "xhigh" },
2319
2356
  input: ["text", "image"],
2320
2357
  cost: {
2321
- input: 5,
2322
- output: 25,
2323
- cacheRead: 0.5,
2324
- cacheWrite: 6.25,
2358
+ input: 5.5,
2359
+ output: 27.5,
2360
+ cacheRead: 0.55,
2361
+ cacheWrite: 6.875,
2325
2362
  },
2326
2363
  contextWindow: 1000000,
2327
2364
  maxTokens: 128000,
@@ -2337,10 +2374,10 @@ export const MODELS = {
2337
2374
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
2338
2375
  input: ["text", "image"],
2339
2376
  cost: {
2340
- input: 5,
2341
- output: 25,
2342
- cacheRead: 0.5,
2343
- cacheWrite: 6.25,
2377
+ input: 5.5,
2378
+ output: 27.5,
2379
+ cacheRead: 0.55,
2380
+ cacheWrite: 6.875,
2344
2381
  },
2345
2382
  contextWindow: 1000000,
2346
2383
  maxTokens: 128000,
@@ -2355,17 +2392,17 @@ export const MODELS = {
2355
2392
  thinkingType: "adaptive",
2356
2393
  input: ["text", "image"],
2357
2394
  cost: {
2358
- input: 5,
2359
- output: 25,
2360
- cacheRead: 0.5,
2361
- cacheWrite: 6.25,
2395
+ input: 5.5,
2396
+ output: 27.5,
2397
+ cacheRead: 0.55,
2398
+ cacheWrite: 6.875,
2362
2399
  },
2363
2400
  contextWindow: 1000000,
2364
2401
  maxTokens: 128000,
2365
2402
  },
2366
- "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2367
- id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
2368
- 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)",
2369
2406
  api: "bedrock-converse-stream",
2370
2407
  provider: "amazon-bedrock",
2371
2408
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -2381,27 +2418,27 @@ export const MODELS = {
2381
2418
  contextWindow: 200000,
2382
2419
  maxTokens: 64000,
2383
2420
  },
2384
- "us.anthropic.claude-sonnet-4-6": {
2385
- id: "us.anthropic.claude-sonnet-4-6",
2386
- name: "Claude Sonnet 4.6 (US)",
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)",
2387
2424
  api: "bedrock-converse-stream",
2388
2425
  provider: "amazon-bedrock",
2389
2426
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2390
2427
  reasoning: true,
2391
- thinkingType: "adaptive",
2428
+ thinkingType: "budget",
2392
2429
  input: ["text", "image"],
2393
2430
  cost: {
2394
- input: 3,
2395
- output: 15,
2396
- cacheRead: 0.3,
2397
- cacheWrite: 3.75,
2431
+ input: 3.3,
2432
+ output: 16.5,
2433
+ cacheRead: 0.33,
2434
+ cacheWrite: 4.125,
2398
2435
  },
2399
- contextWindow: 1000000,
2436
+ contextWindow: 200000,
2400
2437
  maxTokens: 64000,
2401
2438
  },
2402
- "us.anthropic.claude-sonnet-5": {
2403
- id: "us.anthropic.claude-sonnet-5",
2404
- name: "Claude Sonnet 5 (US)",
2439
+ "us.anthropic.claude-sonnet-4-6": {
2440
+ id: "us.anthropic.claude-sonnet-4-6",
2441
+ name: "Claude Sonnet 4.6 (US)",
2405
2442
  api: "bedrock-converse-stream",
2406
2443
  provider: "amazon-bedrock",
2407
2444
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
@@ -2409,30 +2446,31 @@ export const MODELS = {
2409
2446
  thinkingType: "adaptive",
2410
2447
  input: ["text", "image"],
2411
2448
  cost: {
2412
- input: 2,
2413
- output: 10,
2414
- cacheRead: 0.2,
2415
- cacheWrite: 2.5,
2449
+ input: 3.3,
2450
+ output: 16.5,
2451
+ cacheRead: 0.33,
2452
+ cacheWrite: 4.125,
2416
2453
  },
2417
2454
  contextWindow: 1000000,
2418
2455
  maxTokens: 128000,
2419
2456
  },
2420
- "us.deepseek.r1-v1:0": {
2421
- id: "us.deepseek.r1-v1:0",
2422
- name: "DeepSeek-R1 (US)",
2457
+ "us.anthropic.claude-sonnet-5": {
2458
+ id: "us.anthropic.claude-sonnet-5",
2459
+ name: "Claude Sonnet 5 (US)",
2423
2460
  api: "bedrock-converse-stream",
2424
2461
  provider: "amazon-bedrock",
2425
2462
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2426
2463
  reasoning: true,
2427
- input: ["text"],
2464
+ thinkingType: "adaptive",
2465
+ input: ["text", "image"],
2428
2466
  cost: {
2429
- input: 1.35,
2430
- output: 5.4,
2431
- cacheRead: 0,
2432
- cacheWrite: 0,
2467
+ input: 2.2,
2468
+ output: 11,
2469
+ cacheRead: 0.22,
2470
+ cacheWrite: 2.75,
2433
2471
  },
2434
- contextWindow: 128000,
2435
- maxTokens: 32768,
2472
+ contextWindow: 1000000,
2473
+ maxTokens: 128000,
2436
2474
  },
2437
2475
  "us.meta.llama3-1-70b-instruct-v1:0": {
2438
2476
  id: "us.meta.llama3-1-70b-instruct-v1:0",
@@ -2500,7 +2538,7 @@ export const MODELS = {
2500
2538
  cacheWrite: 0,
2501
2539
  },
2502
2540
  contextWindow: 1000000,
2503
- maxTokens: 16384,
2541
+ maxTokens: 8192,
2504
2542
  },
2505
2543
  "us.meta.llama4-scout-17b-instruct-v1:0": {
2506
2544
  id: "us.meta.llama4-scout-17b-instruct-v1:0",
@@ -2516,8 +2554,8 @@ export const MODELS = {
2516
2554
  cacheRead: 0,
2517
2555
  cacheWrite: 0,
2518
2556
  },
2519
- contextWindow: 3500000,
2520
- maxTokens: 16384,
2557
+ contextWindow: 10000000,
2558
+ maxTokens: 8192,
2521
2559
  },
2522
2560
  "us.mistral.pixtral-large-2502-v1:0": {
2523
2561
  id: "us.mistral.pixtral-large-2502-v1:0",
@@ -2772,7 +2810,7 @@ export const MODELS = {
2772
2810
  cacheWrite: 0,
2773
2811
  },
2774
2812
  contextWindow: 202752,
2775
- maxTokens: 101376,
2813
+ maxTokens: 131072,
2776
2814
  },
2777
2815
  },
2778
2816
  "anthropic": {
@@ -4709,6 +4747,23 @@ export const MODELS = {
4709
4747
  contextWindow: 1000000,
4710
4748
  maxTokens: 384000,
4711
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
+ },
4712
4767
  "accounts/fireworks/models/deepseek-v4-pro-0813": {
4713
4768
  id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4714
4769
  name: "DeepSeek V4 Pro 0813",
@@ -4879,6 +4934,23 @@ export const MODELS = {
4879
4934
  contextWindow: 1048576,
4880
4935
  maxTokens: 131072,
4881
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
+ },
4882
4954
  "accounts/fireworks/models/minimax-m3": {
4883
4955
  id: "accounts/fireworks/models/minimax-m3",
4884
4956
  name: "MiniMax-M3",
@@ -4886,7 +4958,7 @@ export const MODELS = {
4886
4958
  provider: "fireworks",
4887
4959
  baseUrl: "https://api.fireworks.ai/inference",
4888
4960
  reasoning: true,
4889
- input: ["text", "image"],
4961
+ input: ["text"],
4890
4962
  cost: {
4891
4963
  input: 0.3,
4892
4964
  output: 1.2,
@@ -4941,7 +5013,7 @@ export const MODELS = {
4941
5013
  cost: {
4942
5014
  input: 0.6,
4943
5015
  output: 2.4,
4944
- cacheRead: 0.119,
5016
+ cacheRead: 0.12,
4945
5017
  cacheWrite: 0,
4946
5018
  },
4947
5019
  contextWindow: 262144,
@@ -5005,7 +5077,7 @@ export const MODELS = {
5005
5077
  provider: "fireworks",
5006
5078
  baseUrl: "https://api.fireworks.ai/inference",
5007
5079
  reasoning: true,
5008
- input: ["text"],
5080
+ input: ["text", "image"],
5009
5081
  cost: {
5010
5082
  input: 2,
5011
5083
  output: 6,
@@ -5015,6 +5087,40 @@ export const MODELS = {
5015
5087
  contextWindow: 262144,
5016
5088
  maxTokens: 131072,
5017
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
+ },
5018
5124
  "accounts/fireworks/routers/glm-5p2-fast": {
5019
5125
  id: "accounts/fireworks/routers/glm-5p2-fast",
5020
5126
  name: "GLM 5.2 Fast",
@@ -5041,29 +5147,148 @@ export const MODELS = {
5041
5147
  reasoning: true,
5042
5148
  input: ["text"],
5043
5149
  cost: {
5044
- input: 2.1,
5045
- output: 6.6,
5046
- cacheRead: 0.39,
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,
5187
+ cacheWrite: 0,
5188
+ },
5189
+ contextWindow: 1048573,
5190
+ maxTokens: 131072,
5191
+ },
5192
+ "accounts/fireworks/routers/glm-latest": {
5193
+ id: "accounts/fireworks/routers/glm-latest",
5194
+ name: "GLM Latest",
5195
+ api: "anthropic-messages",
5196
+ provider: "fireworks",
5197
+ baseUrl: "https://api.fireworks.ai/inference",
5198
+ reasoning: true,
5199
+ input: ["text"],
5200
+ cost: {
5201
+ input: 1.4,
5202
+ output: 4.4,
5203
+ cacheRead: 0.26,
5204
+ cacheWrite: 0,
5205
+ },
5206
+ contextWindow: 1048573,
5207
+ maxTokens: 262144,
5208
+ },
5209
+ "accounts/fireworks/routers/kimi-fast-latest": {
5210
+ id: "accounts/fireworks/routers/kimi-fast-latest",
5211
+ name: "Kimi Fast Latest",
5212
+ api: "anthropic-messages",
5213
+ provider: "fireworks",
5214
+ baseUrl: "https://api.fireworks.ai/inference",
5215
+ reasoning: true,
5216
+ input: ["text", "image"],
5217
+ cost: {
5218
+ input: 4.5,
5219
+ output: 22.5,
5220
+ cacheRead: 0.45,
5221
+ cacheWrite: 0,
5222
+ },
5223
+ contextWindow: 1048576,
5224
+ maxTokens: 131072,
5225
+ },
5226
+ "accounts/fireworks/routers/kimi-k3-fast": {
5227
+ id: "accounts/fireworks/routers/kimi-k3-fast",
5228
+ name: "Kimi K3 Fast",
5229
+ api: "anthropic-messages",
5230
+ provider: "fireworks",
5231
+ baseUrl: "https://api.fireworks.ai/inference",
5232
+ reasoning: true,
5233
+ input: ["text", "image"],
5234
+ cost: {
5235
+ input: 4.5,
5236
+ output: 22.5,
5237
+ cacheRead: 0.45,
5238
+ cacheWrite: 0,
5239
+ },
5240
+ contextWindow: 1048576,
5241
+ maxTokens: 131072,
5242
+ },
5243
+ "accounts/fireworks/routers/kimi-latest": {
5244
+ id: "accounts/fireworks/routers/kimi-latest",
5245
+ name: "Kimi Latest",
5246
+ api: "anthropic-messages",
5247
+ provider: "fireworks",
5248
+ baseUrl: "https://api.fireworks.ai/inference",
5249
+ reasoning: true,
5250
+ input: ["text", "image"],
5251
+ cost: {
5252
+ input: 3,
5253
+ output: 15,
5254
+ cacheRead: 0.3,
5255
+ cacheWrite: 0,
5256
+ },
5257
+ contextWindow: 1048576,
5258
+ maxTokens: 131072,
5259
+ },
5260
+ "accounts/fireworks/routers/minimax-latest": {
5261
+ id: "accounts/fireworks/routers/minimax-latest",
5262
+ name: "MiniMax Latest",
5263
+ api: "anthropic-messages",
5264
+ provider: "fireworks",
5265
+ baseUrl: "https://api.fireworks.ai/inference",
5266
+ reasoning: true,
5267
+ input: ["text"],
5268
+ cost: {
5269
+ input: 0.3,
5270
+ output: 1.2,
5271
+ cacheRead: 0.06,
5047
5272
  cacheWrite: 0,
5048
5273
  },
5049
- contextWindow: 1048572,
5050
- maxTokens: 262144,
5274
+ contextWindow: 512000,
5275
+ maxTokens: 512000,
5051
5276
  },
5052
- "accounts/fireworks/routers/kimi-k3-fast": {
5053
- id: "accounts/fireworks/routers/kimi-k3-fast",
5054
- 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)",
5055
5280
  api: "anthropic-messages",
5056
5281
  provider: "fireworks",
5057
5282
  baseUrl: "https://api.fireworks.ai/inference",
5058
5283
  reasoning: true,
5059
5284
  input: ["text", "image"],
5060
5285
  cost: {
5061
- input: 4.5,
5062
- output: 22.5,
5063
- cacheRead: 0.45,
5286
+ input: 2,
5287
+ output: 6,
5288
+ cacheRead: 0.25,
5064
5289
  cacheWrite: 0,
5065
5290
  },
5066
- contextWindow: 1048576,
5291
+ contextWindow: 262144,
5067
5292
  maxTokens: 131072,
5068
5293
  },
5069
5294
  },
@@ -7713,7 +7938,7 @@ export const MODELS = {
7713
7938
  },
7714
7939
  "kimi-for-coding": {
7715
7940
  id: "kimi-for-coding",
7716
- name: "Kimi K2.7 Code",
7941
+ name: "kimi-for-coding",
7717
7942
  api: "anthropic-messages",
7718
7943
  provider: "kimi-coding",
7719
7944
  baseUrl: "https://api.kimi.com/coding",
@@ -7726,7 +7951,7 @@ export const MODELS = {
7726
7951
  cacheRead: 0,
7727
7952
  cacheWrite: 0,
7728
7953
  },
7729
- contextWindow: 262144,
7954
+ contextWindow: 1048576,
7730
7955
  maxTokens: 32768,
7731
7956
  },
7732
7957
  "kimi-for-coding-highspeed": {
@@ -11521,7 +11746,7 @@ export const MODELS = {
11521
11746
  },
11522
11747
  "anthropic/claude-opus-5": {
11523
11748
  id: "anthropic/claude-opus-5",
11524
- name: "Claude Opus 5",
11749
+ name: "Anthropic: Claude Opus 5",
11525
11750
  api: "openai-completions",
11526
11751
  provider: "openrouter",
11527
11752
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11538,7 +11763,7 @@ export const MODELS = {
11538
11763
  },
11539
11764
  "anthropic/claude-opus-5:batch": {
11540
11765
  id: "anthropic/claude-opus-5:batch",
11541
- name: "Claude Opus 5 (batch)",
11766
+ name: "Anthropic: Claude Opus 5 (batch)",
11542
11767
  api: "openai-completions",
11543
11768
  provider: "openrouter",
11544
11769
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12025,9 +12250,9 @@ export const MODELS = {
12025
12250
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12026
12251
  input: ["text"],
12027
12252
  cost: {
12028
- input: 0.065,
12029
- output: 0.18,
12030
- cacheRead: 0.016,
12253
+ input: 0.06,
12254
+ output: 0.12,
12255
+ cacheRead: 0.012,
12031
12256
  cacheWrite: 0,
12032
12257
  },
12033
12258
  contextWindow: 1310720,
@@ -12101,13 +12326,13 @@ export const MODELS = {
12101
12326
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12102
12327
  input: ["text"],
12103
12328
  cost: {
12104
- input: 0.9552599999999999,
12105
- output: 1.9105199999999998,
12106
- cacheRead: 0.07960500000000001,
12329
+ input: 1.5999999999999999,
12330
+ output: 3.1999999999999997,
12331
+ cacheRead: 0.135,
12107
12332
  cacheWrite: 0,
12108
12333
  },
12109
12334
  contextWindow: 1048576,
12110
- maxTokens: 384000,
12335
+ maxTokens: 393216,
12111
12336
  },
12112
12337
  "deepseek/deepseek-v4-pro-0813": {
12113
12338
  id: "deepseek/deepseek-v4-pro-0813",
@@ -12120,13 +12345,13 @@ export const MODELS = {
12120
12345
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12121
12346
  input: ["text"],
12122
12347
  cost: {
12123
- input: 0.66,
12124
- output: 1.9800000000000002,
12125
- cacheRead: 0.06599999999999999,
12348
+ input: 0.9833999999999999,
12349
+ output: 2.9502,
12350
+ cacheRead: 0.03278,
12126
12351
  cacheWrite: 0,
12127
12352
  },
12128
12353
  contextWindow: 1048576,
12129
- maxTokens: 943718,
12354
+ maxTokens: 384000,
12130
12355
  },
12131
12356
  "deepseek/deepseek-v4-pro-0813:batch": {
12132
12357
  id: "deepseek/deepseek-v4-pro-0813:batch",
@@ -12158,9 +12383,9 @@ export const MODELS = {
12158
12383
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12159
12384
  input: ["text", "image"],
12160
12385
  cost: {
12161
- input: 0.15,
12162
- output: 0.6,
12163
- cacheRead: 0.003,
12386
+ input: 0.3,
12387
+ output: 1.2,
12388
+ cacheRead: 0.006,
12164
12389
  cacheWrite: 0,
12165
12390
  },
12166
12391
  contextWindow: 1048576,
@@ -12285,23 +12510,6 @@ export const MODELS = {
12285
12510
  contextWindow: 1048576,
12286
12511
  maxTokens: 65536,
12287
12512
  },
12288
- "google/gemini-2.5-pro-preview-05-06": {
12289
- id: "google/gemini-2.5-pro-preview-05-06",
12290
- name: "Google: Gemini 2.5 Pro Preview 05-06",
12291
- api: "openai-completions",
12292
- provider: "openrouter",
12293
- baseUrl: "https://openrouter.ai/api/v1",
12294
- reasoning: true,
12295
- input: ["text", "image"],
12296
- cost: {
12297
- input: 1.25,
12298
- output: 10,
12299
- cacheRead: 0.125,
12300
- cacheWrite: 0.375,
12301
- },
12302
- contextWindow: 1048576,
12303
- maxTokens: 65535,
12304
- },
12305
12513
  "google/gemini-2.5-pro:batch": {
12306
12514
  id: "google/gemini-2.5-pro:batch",
12307
12515
  name: "Google: Gemini 2.5 Pro (batch)",
@@ -12685,13 +12893,13 @@ export const MODELS = {
12685
12893
  reasoning: true,
12686
12894
  input: ["text", "image"],
12687
12895
  cost: {
12688
- input: 0.041999999999999996,
12689
- output: 0.22,
12690
- cacheRead: 0,
12896
+ input: 0.09,
12897
+ output: 0.3,
12898
+ cacheRead: 0.049999999999999996,
12691
12899
  cacheWrite: 0,
12692
12900
  },
12693
12901
  contextWindow: 262144,
12694
- maxTokens: 32768,
12902
+ maxTokens: 235929,
12695
12903
  },
12696
12904
  "google/gemma-4-26b-a4b-it:free": {
12697
12905
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12727,23 +12935,6 @@ export const MODELS = {
12727
12935
  contextWindow: 262144,
12728
12936
  maxTokens: 16384,
12729
12937
  },
12730
- "google/gemma-4-31b-it:batch": {
12731
- id: "google/gemma-4-31b-it:batch",
12732
- name: "Google: Gemma 4 31B (batch)",
12733
- api: "openai-completions",
12734
- provider: "openrouter",
12735
- baseUrl: "https://openrouter.ai/api/v1",
12736
- reasoning: true,
12737
- input: ["text", "image"],
12738
- cost: {
12739
- input: 0.39,
12740
- output: 0.9700000000000001,
12741
- cacheRead: 0,
12742
- cacheWrite: 0,
12743
- },
12744
- contextWindow: 262144,
12745
- maxTokens: 235929,
12746
- },
12747
12938
  "google/gemma-4-31b-it:free": {
12748
12939
  id: "google/gemma-4-31b-it:free",
12749
12940
  name: "Google: Gemma 4 31B (free)",
@@ -12880,6 +13071,23 @@ export const MODELS = {
12880
13071
  contextWindow: 262144,
12881
13072
  maxTokens: 32768,
12882
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
+ },
12883
13091
  "inclusionai/ling-3.0-flash-vl:free": {
12884
13092
  id: "inclusionai/ling-3.0-flash-vl:free",
12885
13093
  name: "inclusionAI: Ling 3.0 Flash VL (free)",
@@ -13025,13 +13233,13 @@ export const MODELS = {
13025
13233
  reasoning: false,
13026
13234
  input: ["text", "image"],
13027
13235
  cost: {
13028
- input: 0.19999999999999998,
13029
- output: 0.696,
13236
+ input: 0.1875,
13237
+ output: 0.6525,
13030
13238
  cacheRead: 0,
13031
13239
  cacheWrite: 0,
13032
13240
  },
13033
13241
  contextWindow: 1048576,
13034
- maxTokens: 115200,
13242
+ maxTokens: 16384,
13035
13243
  },
13036
13244
  "meta-llama/llama-4-scout": {
13037
13245
  id: "meta-llama/llama-4-scout",
@@ -13059,8 +13267,8 @@ export const MODELS = {
13059
13267
  reasoning: true,
13060
13268
  input: ["text", "image"],
13061
13269
  cost: {
13062
- input: 0.3,
13063
- output: 1.1,
13270
+ input: 0.35,
13271
+ output: 1.5,
13064
13272
  cacheRead: 0.04,
13065
13273
  cacheWrite: 0,
13066
13274
  },
@@ -13178,7 +13386,7 @@ export const MODELS = {
13178
13386
  reasoning: true,
13179
13387
  input: ["text"],
13180
13388
  cost: {
13181
- input: 0.55,
13389
+ input: 0.39999999999999997,
13182
13390
  output: 2.2,
13183
13391
  cacheRead: 0,
13184
13392
  cacheWrite: 0,
@@ -13229,13 +13437,13 @@ export const MODELS = {
13229
13437
  reasoning: true,
13230
13438
  input: ["text"],
13231
13439
  cost: {
13232
- input: 0.3,
13233
- output: 1.2,
13234
- cacheRead: 0.03,
13440
+ input: 0.27,
13441
+ output: 1.08,
13442
+ cacheRead: 0.027,
13235
13443
  cacheWrite: 0,
13236
13444
  },
13237
13445
  contextWindow: 204800,
13238
- maxTokens: 131072,
13446
+ maxTokens: 128000,
13239
13447
  },
13240
13448
  "minimax/minimax-m2.7": {
13241
13449
  id: "minimax/minimax-m2.7",
@@ -13637,12 +13845,12 @@ export const MODELS = {
13637
13845
  reasoning: false,
13638
13846
  input: ["text", "image"],
13639
13847
  cost: {
13640
- input: 0.075,
13641
- output: 0.19999999999999998,
13848
+ input: 0.09375,
13849
+ output: 0.25,
13642
13850
  cacheRead: 0,
13643
13851
  cacheWrite: 0,
13644
13852
  },
13645
- contextWindow: 131072,
13853
+ contextWindow: 256000,
13646
13854
  maxTokens: 16384,
13647
13855
  },
13648
13856
  "mistralai/mixtral-8x22b-instruct": {
@@ -13694,7 +13902,7 @@ export const MODELS = {
13694
13902
  cacheWrite: 0,
13695
13903
  },
13696
13904
  contextWindow: 131072,
13697
- maxTokens: 100352,
13905
+ maxTokens: 98304,
13698
13906
  },
13699
13907
  "moonshotai/kimi-k2-0905": {
13700
13908
  id: "moonshotai/kimi-k2-0905",
@@ -13711,7 +13919,7 @@ export const MODELS = {
13711
13919
  cacheWrite: 0,
13712
13920
  },
13713
13921
  contextWindow: 262144,
13714
- maxTokens: 100352,
13922
+ maxTokens: 98304,
13715
13923
  },
13716
13924
  "moonshotai/kimi-k2-thinking": {
13717
13925
  id: "moonshotai/kimi-k2-thinking",
@@ -13728,7 +13936,7 @@ export const MODELS = {
13728
13936
  cacheWrite: 0,
13729
13937
  },
13730
13938
  contextWindow: 262144,
13731
- maxTokens: 100352,
13939
+ maxTokens: 98304,
13732
13940
  },
13733
13941
  "moonshotai/kimi-k2.5": {
13734
13942
  id: "moonshotai/kimi-k2.5",
@@ -13773,9 +13981,9 @@ export const MODELS = {
13773
13981
  reasoning: true,
13774
13982
  input: ["text", "image"],
13775
13983
  cost: {
13776
- input: 0.71,
13777
- output: 3.5,
13778
- cacheRead: 0.15,
13984
+ input: 0.7062,
13985
+ output: 3.21,
13986
+ cacheRead: 0.18,
13779
13987
  cacheWrite: 0,
13780
13988
  },
13781
13989
  contextWindow: 262144,
@@ -13790,9 +13998,9 @@ export const MODELS = {
13790
13998
  reasoning: true,
13791
13999
  input: ["text", "image"],
13792
14000
  cost: {
13793
- input: 2.34,
13794
- output: 11.7,
13795
- cacheRead: 0.261,
14001
+ input: 2.6481380629999998,
14002
+ output: 13.282724250000001,
14003
+ cacheRead: 0.30264435,
13796
14004
  cacheWrite: 0,
13797
14005
  },
13798
14006
  contextWindow: 1048576,
@@ -13892,13 +14100,13 @@ export const MODELS = {
13892
14100
  reasoning: true,
13893
14101
  input: ["text"],
13894
14102
  cost: {
13895
- input: 0.08499999999999999,
13896
- output: 0.39999999999999997,
14103
+ input: 0.08,
14104
+ output: 0.44999999999999996,
13897
14105
  cacheRead: 0,
13898
14106
  cacheWrite: 0,
13899
14107
  },
13900
14108
  contextWindow: 262144,
13901
- maxTokens: 16384,
14109
+ maxTokens: 235929,
13902
14110
  },
13903
14111
  "nvidia/nemotron-3-super-120b-a12b:free": {
13904
14112
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -15577,13 +15785,13 @@ export const MODELS = {
15577
15785
  reasoning: true,
15578
15786
  input: ["text"],
15579
15787
  cost: {
15580
- input: 0.22749999999999998,
15581
- output: 0.9099999999999999,
15788
+ input: 0.12,
15789
+ output: 0.24,
15582
15790
  cacheRead: 0,
15583
15791
  cacheWrite: 0,
15584
15792
  },
15585
15793
  contextWindow: 131072,
15586
- maxTokens: 8192,
15794
+ maxTokens: 16384,
15587
15795
  },
15588
15796
  "qwen/qwen3-235b-a22b": {
15589
15797
  id: "qwen/qwen3-235b-a22b",
@@ -15611,13 +15819,13 @@ export const MODELS = {
15611
15819
  reasoning: false,
15612
15820
  input: ["text"],
15613
15821
  cost: {
15614
- input: 0.22,
15615
- output: 0.88,
15616
- cacheRead: 0,
15822
+ input: 0.0875,
15823
+ output: 0.35,
15824
+ cacheRead: 0.0175,
15617
15825
  cacheWrite: 0,
15618
15826
  },
15619
15827
  contextWindow: 262144,
15620
- maxTokens: 16384,
15828
+ maxTokens: 235929,
15621
15829
  },
15622
15830
  "qwen/qwen3-235b-a22b-thinking-2507": {
15623
15831
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -15662,13 +15870,13 @@ export const MODELS = {
15662
15870
  reasoning: false,
15663
15871
  input: ["text"],
15664
15872
  cost: {
15665
- input: 0.09,
15666
- output: 0.3,
15873
+ input: 0.04815,
15874
+ output: 0.19305,
15667
15875
  cacheRead: 0,
15668
15876
  cacheWrite: 0,
15669
15877
  },
15670
15878
  contextWindow: 262144,
15671
- maxTokens: 235929,
15879
+ maxTokens: 32000,
15672
15880
  },
15673
15881
  "qwen/qwen3-30b-a3b-thinking-2507": {
15674
15882
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15872,7 +16080,7 @@ export const MODELS = {
15872
16080
  cacheWrite: 0,
15873
16081
  },
15874
16082
  contextWindow: 262144,
15875
- maxTokens: 235929,
16083
+ maxTokens: 32768,
15876
16084
  },
15877
16085
  "qwen/qwen3-vl-235b-a22b-instruct": {
15878
16086
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -15917,13 +16125,13 @@ export const MODELS = {
15917
16125
  reasoning: false,
15918
16126
  input: ["text", "image"],
15919
16127
  cost: {
15920
- input: 0.15,
15921
- output: 0.6,
16128
+ input: 0.13,
16129
+ output: 0.52,
15922
16130
  cacheRead: 0,
15923
16131
  cacheWrite: 0,
15924
16132
  },
15925
16133
  contextWindow: 262144,
15926
- maxTokens: 16384,
16134
+ maxTokens: 32768,
15927
16135
  },
15928
16136
  "qwen/qwen3-vl-30b-a3b-thinking": {
15929
16137
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -16036,13 +16244,13 @@ export const MODELS = {
16036
16244
  reasoning: true,
16037
16245
  input: ["text", "image"],
16038
16246
  cost: {
16039
- input: 0.3125,
16040
- output: 1.25,
16041
- cacheRead: 0.15625,
16247
+ input: 0.1625,
16248
+ output: 1.3,
16249
+ cacheRead: 0,
16042
16250
  cacheWrite: 0,
16043
16251
  },
16044
16252
  contextWindow: 262144,
16045
- maxTokens: 16384,
16253
+ maxTokens: 65536,
16046
16254
  },
16047
16255
  "qwen/qwen3.5-397b-a17b": {
16048
16256
  id: "qwen/qwen3.5-397b-a17b",
@@ -16325,9 +16533,9 @@ export const MODELS = {
16325
16533
  reasoning: true,
16326
16534
  input: ["text", "image"],
16327
16535
  cost: {
16328
- input: 0.42,
16329
- output: 3,
16330
- cacheRead: 0.08499999999999999,
16536
+ input: 0.21400000000000002,
16537
+ output: 2.5500000000000003,
16538
+ cacheRead: 0.15,
16331
16539
  cacheWrite: 0,
16332
16540
  },
16333
16541
  contextWindow: 1000000,
@@ -16586,7 +16794,7 @@ export const MODELS = {
16586
16794
  cacheWrite: 0,
16587
16795
  },
16588
16796
  contextWindow: 1048576,
16589
- maxTokens: 32768,
16797
+ maxTokens: 471859,
16590
16798
  },
16591
16799
  "thinkingmachines/inkling-small": {
16592
16800
  id: "thinkingmachines/inkling-small",
@@ -16605,23 +16813,6 @@ export const MODELS = {
16605
16813
  contextWindow: 1048576,
16606
16814
  maxTokens: 262144,
16607
16815
  },
16608
- "thinkingmachines/inkling-small:batch": {
16609
- id: "thinkingmachines/inkling-small:batch",
16610
- name: "Thinking Machines: Inkling Small (batch)",
16611
- api: "openai-completions",
16612
- provider: "openrouter",
16613
- baseUrl: "https://openrouter.ai/api/v1",
16614
- reasoning: true,
16615
- input: ["text", "image"],
16616
- cost: {
16617
- input: 0.5,
16618
- output: 1.2,
16619
- cacheRead: 0.09999999999999999,
16620
- cacheWrite: 0,
16621
- },
16622
- contextWindow: 524288,
16623
- maxTokens: 471859,
16624
- },
16625
16816
  "thinkingmachines/inkling-small:free": {
16626
16817
  id: "thinkingmachines/inkling-small:free",
16627
16818
  name: "Thinking Machines: Inkling Small (free)",
@@ -17022,13 +17213,13 @@ export const MODELS = {
17022
17213
  reasoning: true,
17023
17214
  input: ["text"],
17024
17215
  cost: {
17025
- input: 0.966,
17026
- output: 3.036,
17027
- cacheRead: 0.1932,
17216
+ input: 1.4,
17217
+ output: 4.4,
17218
+ cacheRead: 0.14,
17028
17219
  cacheWrite: 0,
17029
17220
  },
17030
17221
  contextWindow: 1048576,
17031
- maxTokens: 131072,
17222
+ maxTokens: 128000,
17032
17223
  },
17033
17224
  "z-ai/glm-5.2:batch": {
17034
17225
  id: "z-ai/glm-5.2:batch",
@@ -17062,7 +17253,7 @@ export const MODELS = {
17062
17253
  cacheWrite: 0,
17063
17254
  },
17064
17255
  contextWindow: 1310720,
17065
- maxTokens: 943718,
17256
+ maxTokens: 943717,
17066
17257
  },
17067
17258
  "z-ai/glm-5.3-flash": {
17068
17259
  id: "z-ai/glm-5.3-flash",
@@ -17073,9 +17264,9 @@ export const MODELS = {
17073
17264
  reasoning: true,
17074
17265
  input: ["text", "image"],
17075
17266
  cost: {
17076
- input: 0.15,
17077
- output: 0.5,
17078
- cacheRead: 0.03,
17267
+ input: 0.09999999999999999,
17268
+ output: 0.3333,
17269
+ cacheRead: 0.02,
17079
17270
  cacheWrite: 0,
17080
17271
  },
17081
17272
  contextWindow: 1310720,
@@ -17152,7 +17343,7 @@ export const MODELS = {
17152
17343
  },
17153
17344
  "~anthropic/claude-haiku-latest": {
17154
17345
  id: "~anthropic/claude-haiku-latest",
17155
- name: "Anthropic Claude Haiku Latest",
17346
+ name: "Anthropic: Claude Haiku Latest",
17156
17347
  api: "openai-completions",
17157
17348
  provider: "openrouter",
17158
17349
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17186,7 +17377,7 @@ export const MODELS = {
17186
17377
  },
17187
17378
  "~anthropic/claude-sonnet-latest": {
17188
17379
  id: "~anthropic/claude-sonnet-latest",
17189
- name: "Anthropic Claude Sonnet Latest",
17380
+ name: "Anthropic: Claude Sonnet Latest",
17190
17381
  api: "openai-completions",
17191
17382
  provider: "openrouter",
17192
17383
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17201,9 +17392,43 @@ export const MODELS = {
17201
17392
  contextWindow: 1000000,
17202
17393
  maxTokens: 128000,
17203
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
+ },
17204
17429
  "~deepseek/deepseek-v4-flash-latest": {
17205
17430
  id: "~deepseek/deepseek-v4-flash-latest",
17206
- name: "DeepSeek V4 Flash Latest",
17431
+ name: "DeepSeek: DeepSeek V4 Flash Latest",
17207
17432
  api: "openai-completions",
17208
17433
  provider: "openrouter",
17209
17434
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17212,9 +17437,9 @@ export const MODELS = {
17212
17437
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
17213
17438
  input: ["text"],
17214
17439
  cost: {
17215
- input: 0.049999999999999996,
17216
- output: 0.16,
17217
- cacheRead: 0.013000000000000001,
17440
+ input: 0.04,
17441
+ output: 0.09999999999999999,
17442
+ cacheRead: 0.01,
17218
17443
  cacheWrite: 0,
17219
17444
  },
17220
17445
  contextWindow: 1310720,
@@ -17222,7 +17447,7 @@ export const MODELS = {
17222
17447
  },
17223
17448
  "~google/gemini-flash-latest": {
17224
17449
  id: "~google/gemini-flash-latest",
17225
- name: "Google Gemini Flash Latest",
17450
+ name: "Google: Gemini Flash Latest",
17226
17451
  api: "openai-completions",
17227
17452
  provider: "openrouter",
17228
17453
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17239,41 +17464,92 @@ export const MODELS = {
17239
17464
  },
17240
17465
  "~google/gemini-pro-latest": {
17241
17466
  id: "~google/gemini-pro-latest",
17242
- name: "Google Gemini Pro Latest",
17467
+ name: "Google: Gemini Pro Latest",
17468
+ api: "openai-completions",
17469
+ provider: "openrouter",
17470
+ baseUrl: "https://openrouter.ai/api/v1",
17471
+ reasoning: true,
17472
+ input: ["text", "image"],
17473
+ cost: {
17474
+ input: 2,
17475
+ output: 12,
17476
+ cacheRead: 0.19999999999999998,
17477
+ cacheWrite: 0.375,
17478
+ },
17479
+ contextWindow: 1048576,
17480
+ maxTokens: 65536,
17481
+ },
17482
+ "~moonshotai/kimi-latest": {
17483
+ id: "~moonshotai/kimi-latest",
17484
+ name: "MoonshotAI: Kimi Latest",
17485
+ api: "openai-completions",
17486
+ provider: "openrouter",
17487
+ baseUrl: "https://openrouter.ai/api/v1",
17488
+ reasoning: true,
17489
+ input: ["text", "image"],
17490
+ cost: {
17491
+ input: 2,
17492
+ output: 11.2,
17493
+ cacheRead: 0.232,
17494
+ cacheWrite: 0,
17495
+ },
17496
+ contextWindow: 1048576,
17497
+ maxTokens: 943718,
17498
+ },
17499
+ "~openai/gpt-astra-latest": {
17500
+ id: "~openai/gpt-astra-latest",
17501
+ name: "OpenAI: GPT Astra Latest",
17502
+ api: "openai-completions",
17503
+ provider: "openrouter",
17504
+ baseUrl: "https://openrouter.ai/api/v1",
17505
+ reasoning: true,
17506
+ input: ["text", "image"],
17507
+ cost: {
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",
17243
17519
  api: "openai-completions",
17244
17520
  provider: "openrouter",
17245
17521
  baseUrl: "https://openrouter.ai/api/v1",
17246
17522
  reasoning: true,
17247
17523
  input: ["text", "image"],
17248
17524
  cost: {
17249
- input: 2,
17250
- output: 12,
17251
- cacheRead: 0.19999999999999998,
17252
- cacheWrite: 0.375,
17525
+ input: 0.19999999999999998,
17526
+ output: 1.2,
17527
+ cacheRead: 0.02,
17528
+ cacheWrite: 0.25,
17253
17529
  },
17254
- contextWindow: 1048576,
17255
- maxTokens: 65536,
17530
+ contextWindow: 1050000,
17531
+ maxTokens: 128000,
17256
17532
  },
17257
- "~moonshotai/kimi-latest": {
17258
- id: "~moonshotai/kimi-latest",
17259
- name: "MoonshotAI Kimi Latest",
17533
+ "~openai/gpt-mini-latest": {
17534
+ id: "~openai/gpt-mini-latest",
17535
+ name: "OpenAI: GPT Mini Latest",
17260
17536
  api: "openai-completions",
17261
17537
  provider: "openrouter",
17262
17538
  baseUrl: "https://openrouter.ai/api/v1",
17263
17539
  reasoning: true,
17264
17540
  input: ["text", "image"],
17265
17541
  cost: {
17266
- input: 2.34,
17267
- output: 11.7,
17268
- cacheRead: 0.261,
17542
+ input: 0.75,
17543
+ output: 4.5,
17544
+ cacheRead: 0.075,
17269
17545
  cacheWrite: 0,
17270
17546
  },
17271
- contextWindow: 1048576,
17272
- maxTokens: 943718,
17547
+ contextWindow: 400000,
17548
+ maxTokens: 128000,
17273
17549
  },
17274
- "~openai/gpt-latest": {
17275
- id: "~openai/gpt-latest",
17276
- name: "OpenAI GPT Latest",
17550
+ "~openai/gpt-sol-latest": {
17551
+ id: "~openai/gpt-sol-latest",
17552
+ name: "OpenAI: GPT Sol Latest",
17277
17553
  api: "openai-completions",
17278
17554
  provider: "openrouter",
17279
17555
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17288,21 +17564,21 @@ export const MODELS = {
17288
17564
  contextWindow: 1050000,
17289
17565
  maxTokens: 128000,
17290
17566
  },
17291
- "~openai/gpt-mini-latest": {
17292
- id: "~openai/gpt-mini-latest",
17293
- name: "OpenAI GPT Mini Latest",
17567
+ "~openai/gpt-terra-latest": {
17568
+ id: "~openai/gpt-terra-latest",
17569
+ name: "OpenAI: GPT Terra Latest",
17294
17570
  api: "openai-completions",
17295
17571
  provider: "openrouter",
17296
17572
  baseUrl: "https://openrouter.ai/api/v1",
17297
17573
  reasoning: true,
17298
17574
  input: ["text", "image"],
17299
17575
  cost: {
17300
- input: 0.75,
17301
- output: 4.5,
17302
- cacheRead: 0.075,
17303
- cacheWrite: 0,
17576
+ input: 2,
17577
+ output: 12,
17578
+ cacheRead: 0.19999999999999998,
17579
+ cacheWrite: 2.5,
17304
17580
  },
17305
- contextWindow: 400000,
17581
+ contextWindow: 1050000,
17306
17582
  maxTokens: 128000,
17307
17583
  },
17308
17584
  "~x-ai/grok-latest": {
@@ -17348,13 +17624,13 @@ export const MODELS = {
17348
17624
  reasoning: true,
17349
17625
  input: ["text"],
17350
17626
  cost: {
17351
- input: 0.9700000000000001,
17352
- output: 3.3077,
17353
- cacheRead: 0.19399999999999998,
17627
+ input: 0.8775,
17628
+ output: 2.9699999999999998,
17629
+ cacheRead: 0.1755,
17354
17630
  cacheWrite: 0,
17355
17631
  },
17356
17632
  contextWindow: 1310720,
17357
- maxTokens: 943718,
17633
+ maxTokens: 235929,
17358
17634
  },
17359
17635
  },
17360
17636
  "together": {
@@ -17527,6 +17803,25 @@ export const MODELS = {
17527
17803
  contextWindow: 1048576,
17528
17804
  maxTokens: 384000,
17529
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
+ },
17530
17825
  "google/gemma-4-31B-it": {
17531
17826
  id: "google/gemma-4-31B-it",
17532
17827
  name: "Gemma 4 31B Instruct",
@@ -17835,7 +18130,7 @@ export const MODELS = {
17835
18130
  cost: {
17836
18131
  input: 1.3,
17837
18132
  output: 7.8,
17838
- cacheRead: 0.26,
18133
+ cacheRead: 0.13,
17839
18134
  cacheWrite: 1.625,
17840
18135
  },
17841
18136
  contextWindow: 240000,
@@ -18073,7 +18368,7 @@ export const MODELS = {
18073
18368
  cost: {
18074
18369
  input: 0.09999999999999999,
18075
18370
  output: 0.39999999999999997,
18076
- cacheRead: 0.001,
18371
+ cacheRead: 0.01,
18077
18372
  cacheWrite: 0.125,
18078
18373
  },
18079
18374
  contextWindow: 1000000,
@@ -18089,7 +18384,7 @@ export const MODELS = {
18089
18384
  input: ["text", "image"],
18090
18385
  cost: {
18091
18386
  input: 0.39999999999999997,
18092
- output: 2.4,
18387
+ output: 2.5,
18093
18388
  cacheRead: 0.04,
18094
18389
  cacheWrite: 0.5,
18095
18390
  },
@@ -18124,7 +18419,7 @@ export const MODELS = {
18124
18419
  cost: {
18125
18420
  input: 0.5,
18126
18421
  output: 3,
18127
- cacheRead: 0.09999999999999999,
18422
+ cacheRead: 0.049999999999999996,
18128
18423
  cacheWrite: 0.625,
18129
18424
  },
18130
18425
  contextWindow: 1000000,
@@ -18224,7 +18519,7 @@ export const MODELS = {
18224
18519
  reasoning: true,
18225
18520
  input: ["text", "image"],
18226
18521
  cost: {
18227
- input: 0.16,
18522
+ input: 0.15,
18228
18523
  output: 0.47,
18229
18524
  cacheRead: 0.016,
18230
18525
  cacheWrite: 0.19999999999999998,
@@ -18663,6 +18958,23 @@ export const MODELS = {
18663
18958
  contextWindow: 256000,
18664
18959
  maxTokens: 64000,
18665
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
+ },
18666
18978
  "cohere/command-a": {
18667
18979
  id: "cohere/command-a",
18668
18980
  name: "Command A",
@@ -18859,13 +19171,13 @@ export const MODELS = {
18859
19171
  reasoning: true,
18860
19172
  input: ["text", "image"],
18861
19173
  cost: {
18862
- input: 0.3,
18863
- output: 1.2,
18864
- cacheRead: 0.03,
19174
+ input: 0.15,
19175
+ output: 0.6,
19176
+ cacheRead: 0.015,
18865
19177
  cacheWrite: 0,
18866
19178
  },
18867
- contextWindow: 1048576,
18868
- maxTokens: 32768,
19179
+ contextWindow: 1000000,
19180
+ maxTokens: 384000,
18869
19181
  },
18870
19182
  "google/gemini-2.5-flash": {
18871
19183
  id: "google/gemini-2.5-flash",
@@ -19148,9 +19460,9 @@ export const MODELS = {
19148
19460
  reasoning: true,
19149
19461
  input: ["text"],
19150
19462
  cost: {
19151
- input: 0.06,
19152
- output: 0.18,
19153
- cacheRead: 0.012,
19463
+ input: 0.020999999999999998,
19464
+ output: 0.063,
19465
+ cacheRead: 0.004200000000000001,
19154
19466
  cacheWrite: 0,
19155
19467
  },
19156
19468
  contextWindow: 256000,
@@ -19224,6 +19536,40 @@ export const MODELS = {
19224
19536
  contextWindow: 256000,
19225
19537
  maxTokens: 32000,
19226
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
+ },
19227
19573
  "interfaze/interfaze-beta": {
19228
19574
  id: "interfaze/interfaze-beta",
19229
19575
  name: "Interfaze Beta",
@@ -19643,46 +19989,12 @@ export const MODELS = {
19643
19989
  cost: {
19644
19990
  input: 0.3,
19645
19991
  output: 0.8999999999999999,
19646
- cacheRead: 0,
19992
+ cacheRead: 0.03,
19647
19993
  cacheWrite: 0,
19648
19994
  },
19649
19995
  contextWindow: 128000,
19650
19996
  maxTokens: 4000,
19651
19997
  },
19652
- "mistral/devstral-2": {
19653
- id: "mistral/devstral-2",
19654
- name: "Devstral 2",
19655
- api: "anthropic-messages",
19656
- provider: "vercel-ai-gateway",
19657
- baseUrl: "https://ai-gateway.vercel.sh",
19658
- reasoning: false,
19659
- input: ["text"],
19660
- cost: {
19661
- input: 0.39999999999999997,
19662
- output: 2,
19663
- cacheRead: 0,
19664
- cacheWrite: 0,
19665
- },
19666
- contextWindow: 256000,
19667
- maxTokens: 256000,
19668
- },
19669
- "mistral/devstral-small-2": {
19670
- id: "mistral/devstral-small-2",
19671
- name: "Devstral Small 2",
19672
- api: "anthropic-messages",
19673
- provider: "vercel-ai-gateway",
19674
- baseUrl: "https://ai-gateway.vercel.sh",
19675
- reasoning: false,
19676
- input: ["text", "image"],
19677
- cost: {
19678
- input: 0.09999999999999999,
19679
- output: 0.3,
19680
- cacheRead: 0,
19681
- cacheWrite: 0,
19682
- },
19683
- contextWindow: 256000,
19684
- maxTokens: 256000,
19685
- },
19686
19998
  "mistral/ministral-14b": {
19687
19999
  id: "mistral/ministral-14b",
19688
20000
  name: "Ministral 14B",
@@ -19694,10 +20006,10 @@ export const MODELS = {
19694
20006
  cost: {
19695
20007
  input: 0.19999999999999998,
19696
20008
  output: 0.19999999999999998,
19697
- cacheRead: 0,
20009
+ cacheRead: 0.02,
19698
20010
  cacheWrite: 0,
19699
20011
  },
19700
- contextWindow: 256000,
20012
+ contextWindow: 262144,
19701
20013
  maxTokens: 256000,
19702
20014
  },
19703
20015
  "mistral/ministral-3b": {
@@ -19711,10 +20023,10 @@ export const MODELS = {
19711
20023
  cost: {
19712
20024
  input: 0.09999999999999999,
19713
20025
  output: 0.09999999999999999,
19714
- cacheRead: 0,
20026
+ cacheRead: 0.01,
19715
20027
  cacheWrite: 0,
19716
20028
  },
19717
- contextWindow: 128000,
20029
+ contextWindow: 131072,
19718
20030
  maxTokens: 4000,
19719
20031
  },
19720
20032
  "mistral/ministral-8b": {
@@ -19728,10 +20040,10 @@ export const MODELS = {
19728
20040
  cost: {
19729
20041
  input: 0.15,
19730
20042
  output: 0.15,
19731
- cacheRead: 0,
20043
+ cacheRead: 0.015,
19732
20044
  cacheWrite: 0,
19733
20045
  },
19734
- contextWindow: 128000,
20046
+ contextWindow: 262144,
19735
20047
  maxTokens: 4000,
19736
20048
  },
19737
20049
  "mistral/mistral-large-3": {
@@ -19745,29 +20057,12 @@ export const MODELS = {
19745
20057
  cost: {
19746
20058
  input: 0.5,
19747
20059
  output: 1.5,
19748
- cacheRead: 0,
20060
+ cacheRead: 0.049999999999999996,
19749
20061
  cacheWrite: 0,
19750
20062
  },
19751
- contextWindow: 256000,
20063
+ contextWindow: 262144,
19752
20064
  maxTokens: 256000,
19753
20065
  },
19754
- "mistral/mistral-medium": {
19755
- id: "mistral/mistral-medium",
19756
- name: "Mistral Medium 3.1",
19757
- api: "anthropic-messages",
19758
- provider: "vercel-ai-gateway",
19759
- baseUrl: "https://ai-gateway.vercel.sh",
19760
- reasoning: false,
19761
- input: ["text", "image"],
19762
- cost: {
19763
- input: 0.39999999999999997,
19764
- output: 2,
19765
- cacheRead: 0,
19766
- cacheWrite: 0,
19767
- },
19768
- contextWindow: 128000,
19769
- maxTokens: 64000,
19770
- },
19771
20066
  "mistral/mistral-medium-3.5": {
19772
20067
  id: "mistral/mistral-medium-3.5",
19773
20068
  name: "Mistral Medium Latest",
@@ -19779,10 +20074,10 @@ export const MODELS = {
19779
20074
  cost: {
19780
20075
  input: 1.5,
19781
20076
  output: 7.5,
19782
- cacheRead: 0,
20077
+ cacheRead: 0.15,
19783
20078
  cacheWrite: 0,
19784
20079
  },
19785
- contextWindow: 256000,
20080
+ contextWindow: 262144,
19786
20081
  maxTokens: 256000,
19787
20082
  },
19788
20083
  "mistral/mistral-nemo": {
@@ -19792,15 +20087,15 @@ export const MODELS = {
19792
20087
  provider: "vercel-ai-gateway",
19793
20088
  baseUrl: "https://ai-gateway.vercel.sh",
19794
20089
  reasoning: false,
19795
- input: ["text", "image"],
20090
+ input: ["text"],
19796
20091
  cost: {
19797
- input: 0.15,
19798
- output: 0.15,
20092
+ input: 0.04,
20093
+ output: 0.16999999999999998,
19799
20094
  cacheRead: 0,
19800
20095
  cacheWrite: 0,
19801
20096
  },
19802
- contextWindow: 128000,
19803
- maxTokens: 128000,
20097
+ contextWindow: 60288,
20098
+ maxTokens: 16000,
19804
20099
  },
19805
20100
  "mistral/mistral-small": {
19806
20101
  id: "mistral/mistral-small",
@@ -19810,30 +20105,13 @@ export const MODELS = {
19810
20105
  baseUrl: "https://ai-gateway.vercel.sh",
19811
20106
  reasoning: false,
19812
20107
  input: ["text", "image"],
19813
- cost: {
19814
- input: 0.09999999999999999,
19815
- output: 0.3,
19816
- cacheRead: 0,
19817
- cacheWrite: 0,
19818
- },
19819
- contextWindow: 32000,
19820
- maxTokens: 4000,
19821
- },
19822
- "mistral/pixtral-12b": {
19823
- id: "mistral/pixtral-12b",
19824
- name: "Pixtral 12B 2409",
19825
- api: "anthropic-messages",
19826
- provider: "vercel-ai-gateway",
19827
- baseUrl: "https://ai-gateway.vercel.sh",
19828
- reasoning: false,
19829
- input: ["text", "image"],
19830
20108
  cost: {
19831
20109
  input: 0.15,
19832
- output: 0.15,
19833
- cacheRead: 0,
20110
+ output: 0.6,
20111
+ cacheRead: 0.015,
19834
20112
  cacheWrite: 0,
19835
20113
  },
19836
- contextWindow: 128000,
20114
+ contextWindow: 262144,
19837
20115
  maxTokens: 4000,
19838
20116
  },
19839
20117
  "moonshotai/kimi-k2": {
@@ -19982,8 +20260,8 @@ export const MODELS = {
19982
20260
  input: ["text"],
19983
20261
  cost: {
19984
20262
  input: 0.049999999999999996,
19985
- output: 0.24,
19986
- cacheRead: 0,
20263
+ output: 0.19999999999999998,
20264
+ cacheRead: 0.024999999999999998,
19987
20265
  cacheWrite: 0,
19988
20266
  },
19989
20267
  contextWindow: 262144,
@@ -21092,6 +21370,23 @@ export const MODELS = {
21092
21370
  contextWindow: 256000,
21093
21371
  maxTokens: 32768,
21094
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
+ },
21095
21390
  "sakana/fugu-ultra": {
21096
21391
  id: "sakana/fugu-ultra",
21097
21392
  name: "Fugu Ultra",
@@ -21109,6 +21404,23 @@ export const MODELS = {
21109
21404
  contextWindow: 1000000,
21110
21405
  maxTokens: 1000000,
21111
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
+ },
21112
21424
  "sakana/namazu": {
21113
21425
  id: "sakana/namazu",
21114
21426
  name: "Sakana Namazu",