@elyracode/ai 0.9.37 → 0.9.39

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,
@@ -2355,13 +2392,31 @@ 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,
2399
+ },
2400
+ contextWindow: 1000000,
2401
+ maxTokens: 128000,
2402
+ },
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)",
2406
+ api: "bedrock-converse-stream",
2407
+ provider: "amazon-bedrock",
2408
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2409
+ reasoning: true,
2410
+ thinkingType: "budget",
2411
+ input: ["text", "image"],
2412
+ cost: {
2413
+ input: 3,
2414
+ output: 15,
2415
+ cacheRead: 0.3,
2416
+ cacheWrite: 3.75,
2362
2417
  },
2363
- contextWindow: 1000000,
2364
- maxTokens: 128000,
2418
+ contextWindow: 200000,
2419
+ maxTokens: 64000,
2365
2420
  },
2366
2421
  "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {
2367
2422
  id: "us.anthropic.claude-sonnet-4-5-20250929-v1:0",
@@ -2373,10 +2428,10 @@ export const MODELS = {
2373
2428
  thinkingType: "budget",
2374
2429
  input: ["text", "image"],
2375
2430
  cost: {
2376
- input: 3,
2377
- output: 15,
2378
- cacheRead: 0.3,
2379
- cacheWrite: 3.75,
2431
+ input: 3.3,
2432
+ output: 16.5,
2433
+ cacheRead: 0.33,
2434
+ cacheWrite: 4.125,
2380
2435
  },
2381
2436
  contextWindow: 200000,
2382
2437
  maxTokens: 64000,
@@ -2391,13 +2446,13 @@ export const MODELS = {
2391
2446
  thinkingType: "adaptive",
2392
2447
  input: ["text", "image"],
2393
2448
  cost: {
2394
- input: 3,
2395
- output: 15,
2396
- cacheRead: 0.3,
2397
- cacheWrite: 3.75,
2449
+ input: 3.3,
2450
+ output: 16.5,
2451
+ cacheRead: 0.33,
2452
+ cacheWrite: 4.125,
2398
2453
  },
2399
2454
  contextWindow: 1000000,
2400
- maxTokens: 64000,
2455
+ maxTokens: 128000,
2401
2456
  },
2402
2457
  "us.anthropic.claude-sonnet-5": {
2403
2458
  id: "us.anthropic.claude-sonnet-5",
@@ -2409,31 +2464,14 @@ export const MODELS = {
2409
2464
  thinkingType: "adaptive",
2410
2465
  input: ["text", "image"],
2411
2466
  cost: {
2412
- input: 2,
2413
- output: 10,
2414
- cacheRead: 0.2,
2415
- cacheWrite: 2.5,
2467
+ input: 2.2,
2468
+ output: 11,
2469
+ cacheRead: 0.22,
2470
+ cacheWrite: 2.75,
2416
2471
  },
2417
2472
  contextWindow: 1000000,
2418
2473
  maxTokens: 128000,
2419
2474
  },
2420
- "us.deepseek.r1-v1:0": {
2421
- id: "us.deepseek.r1-v1:0",
2422
- name: "DeepSeek-R1 (US)",
2423
- api: "bedrock-converse-stream",
2424
- provider: "amazon-bedrock",
2425
- baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2426
- reasoning: true,
2427
- input: ["text"],
2428
- cost: {
2429
- input: 1.35,
2430
- output: 5.4,
2431
- cacheRead: 0,
2432
- cacheWrite: 0,
2433
- },
2434
- contextWindow: 128000,
2435
- maxTokens: 32768,
2436
- },
2437
2475
  "us.meta.llama3-1-70b-instruct-v1:0": {
2438
2476
  id: "us.meta.llama3-1-70b-instruct-v1:0",
2439
2477
  name: "Llama 3.1 70B Instruct (US)",
@@ -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": {
@@ -4254,6 +4292,23 @@ export const MODELS = {
4254
4292
  contextWindow: 1050000,
4255
4293
  maxTokens: 128000,
4256
4294
  },
4295
+ "gpt-6-astra": {
4296
+ id: "gpt-6-astra",
4297
+ name: "GPT-6 Astra",
4298
+ api: "openai-responses",
4299
+ provider: "cloudflare-ai-gateway",
4300
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/openai",
4301
+ reasoning: true,
4302
+ input: ["text", "image"],
4303
+ cost: {
4304
+ input: 10,
4305
+ output: 50,
4306
+ cacheRead: 1,
4307
+ cacheWrite: 12,
4308
+ },
4309
+ contextWindow: 1050000,
4310
+ maxTokens: 128000,
4311
+ },
4257
4312
  "o3": {
4258
4313
  id: "o3",
4259
4314
  name: "o3",
@@ -4709,6 +4764,23 @@ export const MODELS = {
4709
4764
  contextWindow: 1000000,
4710
4765
  maxTokens: 384000,
4711
4766
  },
4767
+ "accounts/fireworks/models/deepseek-v4-pro": {
4768
+ id: "accounts/fireworks/models/deepseek-v4-pro",
4769
+ name: "DeepSeek V4 Pro",
4770
+ api: "anthropic-messages",
4771
+ provider: "fireworks",
4772
+ baseUrl: "https://api.fireworks.ai/inference",
4773
+ reasoning: true,
4774
+ input: ["text"],
4775
+ cost: {
4776
+ input: 1.2,
4777
+ output: 1.2,
4778
+ cacheRead: 0.6,
4779
+ cacheWrite: 0,
4780
+ },
4781
+ contextWindow: 1000000,
4782
+ maxTokens: 384000,
4783
+ },
4712
4784
  "accounts/fireworks/models/deepseek-v4-pro-0813": {
4713
4785
  id: "accounts/fireworks/models/deepseek-v4-pro-0813",
4714
4786
  name: "DeepSeek V4 Pro 0813",
@@ -4879,6 +4951,23 @@ export const MODELS = {
4879
4951
  contextWindow: 1048576,
4880
4952
  maxTokens: 131072,
4881
4953
  },
4954
+ "accounts/fireworks/models/minimax-m2p7": {
4955
+ id: "accounts/fireworks/models/minimax-m2p7",
4956
+ name: "MiniMax-M2.7",
4957
+ api: "anthropic-messages",
4958
+ provider: "fireworks",
4959
+ baseUrl: "https://api.fireworks.ai/inference",
4960
+ reasoning: true,
4961
+ input: ["text"],
4962
+ cost: {
4963
+ input: 1.2,
4964
+ output: 1.2,
4965
+ cacheRead: 0.6,
4966
+ cacheWrite: 0,
4967
+ },
4968
+ contextWindow: 196608,
4969
+ maxTokens: 131072,
4970
+ },
4882
4971
  "accounts/fireworks/models/minimax-m3": {
4883
4972
  id: "accounts/fireworks/models/minimax-m3",
4884
4973
  name: "MiniMax-M3",
@@ -4886,7 +4975,7 @@ export const MODELS = {
4886
4975
  provider: "fireworks",
4887
4976
  baseUrl: "https://api.fireworks.ai/inference",
4888
4977
  reasoning: true,
4889
- input: ["text", "image"],
4978
+ input: ["text"],
4890
4979
  cost: {
4891
4980
  input: 0.3,
4892
4981
  output: 1.2,
@@ -4896,174 +4985,310 @@ export const MODELS = {
4896
4985
  contextWindow: 512000,
4897
4986
  maxTokens: 512000,
4898
4987
  },
4899
- "accounts/fireworks/models/mistral-large-3-fp8": {
4900
- id: "accounts/fireworks/models/mistral-large-3-fp8",
4901
- name: "Mistral Large 3 675B Instruct 2512",
4988
+ "accounts/fireworks/models/muse-glimmer-30b": {
4989
+ id: "accounts/fireworks/models/muse-glimmer-30b",
4990
+ name: "Muse Glimmer 30B",
4902
4991
  api: "anthropic-messages",
4903
4992
  provider: "fireworks",
4904
4993
  baseUrl: "https://api.fireworks.ai/inference",
4905
- reasoning: false,
4994
+ reasoning: true,
4906
4995
  input: ["text", "image"],
4907
4996
  cost: {
4908
- input: 0,
4909
- output: 0,
4910
- cacheRead: 0,
4997
+ input: 0.35,
4998
+ output: 1.5,
4999
+ cacheRead: 0.04,
5000
+ cacheWrite: 0,
5001
+ },
5002
+ contextWindow: 131072,
5003
+ maxTokens: 131072,
5004
+ },
5005
+ "accounts/fireworks/models/nemotron-3-ultra-nvfp4": {
5006
+ id: "accounts/fireworks/models/nemotron-3-ultra-nvfp4",
5007
+ name: "Nemotron 3 Ultra 550B A55B",
5008
+ api: "anthropic-messages",
5009
+ provider: "fireworks",
5010
+ baseUrl: "https://api.fireworks.ai/inference",
5011
+ reasoning: true,
5012
+ input: ["text"],
5013
+ cost: {
5014
+ input: 0.6,
5015
+ output: 2.4,
5016
+ cacheRead: 0.12,
5017
+ cacheWrite: 0,
5018
+ },
5019
+ contextWindow: 262144,
5020
+ maxTokens: 128000,
5021
+ },
5022
+ "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b": {
5023
+ id: "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b",
5024
+ name: "Nemotron 3.5 Lightning 30B A3B",
5025
+ api: "anthropic-messages",
5026
+ provider: "fireworks",
5027
+ baseUrl: "https://api.fireworks.ai/inference",
5028
+ reasoning: true,
5029
+ input: ["text"],
5030
+ cost: {
5031
+ input: 0.05,
5032
+ output: 0.2,
5033
+ cacheRead: 0.01,
4911
5034
  cacheWrite: 0,
4912
5035
  },
4913
5036
  contextWindow: 262144,
4914
5037
  maxTokens: 262144,
4915
5038
  },
4916
- "accounts/fireworks/models/muse-glimmer-30b": {
4917
- id: "accounts/fireworks/models/muse-glimmer-30b",
4918
- name: "Muse Glimmer 30B",
5039
+ "accounts/fireworks/models/qwen3p7-plus": {
5040
+ id: "accounts/fireworks/models/qwen3p7-plus",
5041
+ name: "Qwen 3.7 Plus",
4919
5042
  api: "anthropic-messages",
4920
5043
  provider: "fireworks",
4921
5044
  baseUrl: "https://api.fireworks.ai/inference",
4922
5045
  reasoning: true,
4923
5046
  input: ["text", "image"],
4924
5047
  cost: {
4925
- input: 0.35,
4926
- output: 1.5,
4927
- cacheRead: 0.04,
5048
+ input: 0.4,
5049
+ output: 1.6,
5050
+ cacheRead: 0.08,
4928
5051
  cacheWrite: 0,
4929
5052
  },
4930
- contextWindow: 131072,
5053
+ contextWindow: 262144,
5054
+ maxTokens: 65536,
5055
+ },
5056
+ "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
5057
+ id: "accounts/fireworks/models/qwen3p8-2p4t-a95b",
5058
+ name: "Qwen3.8 2.4T A95B",
5059
+ api: "anthropic-messages",
5060
+ provider: "fireworks",
5061
+ baseUrl: "https://api.fireworks.ai/inference",
5062
+ reasoning: true,
5063
+ input: ["text"],
5064
+ cost: {
5065
+ input: 2,
5066
+ output: 6,
5067
+ cacheRead: 0.25,
5068
+ cacheWrite: 0,
5069
+ },
5070
+ contextWindow: 262144,
5071
+ maxTokens: 131072,
5072
+ },
5073
+ "accounts/fireworks/models/qwen3p8-max": {
5074
+ id: "accounts/fireworks/models/qwen3p8-max",
5075
+ name: "Qwen3.8 Max",
5076
+ api: "anthropic-messages",
5077
+ provider: "fireworks",
5078
+ baseUrl: "https://api.fireworks.ai/inference",
5079
+ reasoning: true,
5080
+ input: ["text", "image"],
5081
+ cost: {
5082
+ input: 2,
5083
+ output: 6,
5084
+ cacheRead: 0.25,
5085
+ cacheWrite: 0,
5086
+ },
5087
+ contextWindow: 262144,
5088
+ maxTokens: 131072,
5089
+ },
5090
+ "accounts/fireworks/routers/deepseek-flash-latest": {
5091
+ id: "accounts/fireworks/routers/deepseek-flash-latest",
5092
+ name: "DeepSeek Flash Latest",
5093
+ api: "anthropic-messages",
5094
+ provider: "fireworks",
5095
+ baseUrl: "https://api.fireworks.ai/inference",
5096
+ reasoning: true,
5097
+ input: ["text", "image"],
5098
+ cost: {
5099
+ input: 0.22,
5100
+ output: 0.66,
5101
+ cacheRead: 0.007,
5102
+ cacheWrite: 0,
5103
+ },
5104
+ contextWindow: 1000000,
5105
+ maxTokens: 384000,
5106
+ },
5107
+ "accounts/fireworks/routers/deepseek-pro-latest": {
5108
+ id: "accounts/fireworks/routers/deepseek-pro-latest",
5109
+ name: "DeepSeek Pro Latest",
5110
+ api: "anthropic-messages",
5111
+ provider: "fireworks",
5112
+ baseUrl: "https://api.fireworks.ai/inference",
5113
+ reasoning: true,
5114
+ input: ["text"],
5115
+ cost: {
5116
+ input: 1.32,
5117
+ output: 3.96,
5118
+ cacheRead: 0.044,
5119
+ cacheWrite: 0,
5120
+ },
5121
+ contextWindow: 1000000,
5122
+ maxTokens: 384000,
5123
+ },
5124
+ "accounts/fireworks/routers/glm-5p2-fast": {
5125
+ id: "accounts/fireworks/routers/glm-5p2-fast",
5126
+ name: "GLM 5.2 Fast",
5127
+ api: "anthropic-messages",
5128
+ provider: "fireworks",
5129
+ baseUrl: "https://api.fireworks.ai/inference",
5130
+ reasoning: true,
5131
+ input: ["text"],
5132
+ cost: {
5133
+ input: 2.1,
5134
+ output: 6.6,
5135
+ cacheRead: 0.21,
5136
+ cacheWrite: 0,
5137
+ },
5138
+ contextWindow: 1048575,
4931
5139
  maxTokens: 131072,
4932
5140
  },
4933
- "accounts/fireworks/models/nemotron-3-ultra-nvfp4": {
4934
- id: "accounts/fireworks/models/nemotron-3-ultra-nvfp4",
4935
- name: "Nemotron 3 Ultra 550B A55B",
5141
+ "accounts/fireworks/routers/glm-5p3-fast": {
5142
+ id: "accounts/fireworks/routers/glm-5p3-fast",
5143
+ name: "GLM 5.3 Fast",
5144
+ api: "anthropic-messages",
5145
+ provider: "fireworks",
5146
+ baseUrl: "https://api.fireworks.ai/inference",
5147
+ reasoning: true,
5148
+ input: ["text"],
5149
+ cost: {
5150
+ input: 2.1,
5151
+ output: 6.6,
5152
+ cacheRead: 0.39,
5153
+ cacheWrite: 0,
5154
+ },
5155
+ contextWindow: 1048572,
5156
+ maxTokens: 262144,
5157
+ },
5158
+ "accounts/fireworks/routers/glm-fast-latest": {
5159
+ id: "accounts/fireworks/routers/glm-fast-latest",
5160
+ name: "GLM 5.3 Fast (Latest)",
4936
5161
  api: "anthropic-messages",
4937
5162
  provider: "fireworks",
4938
5163
  baseUrl: "https://api.fireworks.ai/inference",
4939
5164
  reasoning: true,
4940
5165
  input: ["text"],
4941
5166
  cost: {
4942
- input: 0.6,
4943
- output: 2.4,
4944
- cacheRead: 0.119,
5167
+ input: 2.1,
5168
+ output: 6.6,
5169
+ cacheRead: 0.39,
4945
5170
  cacheWrite: 0,
4946
5171
  },
4947
- contextWindow: 262144,
4948
- maxTokens: 128000,
5172
+ contextWindow: 1048572,
5173
+ maxTokens: 262144,
4949
5174
  },
4950
- "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b": {
4951
- id: "accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b",
4952
- name: "Nemotron 3.5 Lightning 30B A3B",
5175
+ "accounts/fireworks/routers/glm-flash-latest": {
5176
+ id: "accounts/fireworks/routers/glm-flash-latest",
5177
+ name: "GLM Flash Latest (GLM 5.3 Flash)",
4953
5178
  api: "anthropic-messages",
4954
5179
  provider: "fireworks",
4955
5180
  baseUrl: "https://api.fireworks.ai/inference",
4956
5181
  reasoning: true,
4957
- input: ["text"],
5182
+ input: ["text", "image"],
4958
5183
  cost: {
4959
- input: 0.05,
4960
- output: 0.2,
4961
- cacheRead: 0.01,
5184
+ input: 0.15,
5185
+ output: 0.5,
5186
+ cacheRead: 0.03,
4962
5187
  cacheWrite: 0,
4963
5188
  },
4964
- contextWindow: 262144,
4965
- maxTokens: 262144,
5189
+ contextWindow: 1048573,
5190
+ maxTokens: 131072,
4966
5191
  },
4967
- "accounts/fireworks/models/qwen3p7-plus": {
4968
- id: "accounts/fireworks/models/qwen3p7-plus",
4969
- name: "Qwen 3.7 Plus",
5192
+ "accounts/fireworks/routers/glm-latest": {
5193
+ id: "accounts/fireworks/routers/glm-latest",
5194
+ name: "GLM Latest",
4970
5195
  api: "anthropic-messages",
4971
5196
  provider: "fireworks",
4972
5197
  baseUrl: "https://api.fireworks.ai/inference",
4973
5198
  reasoning: true,
4974
- input: ["text", "image"],
5199
+ input: ["text"],
4975
5200
  cost: {
4976
- input: 0.4,
4977
- output: 1.6,
4978
- cacheRead: 0.08,
5201
+ input: 1.4,
5202
+ output: 4.4,
5203
+ cacheRead: 0.26,
4979
5204
  cacheWrite: 0,
4980
5205
  },
4981
- contextWindow: 262144,
4982
- maxTokens: 65536,
5206
+ contextWindow: 1048573,
5207
+ maxTokens: 262144,
4983
5208
  },
4984
- "accounts/fireworks/models/qwen3p8-2p4t-a95b": {
4985
- id: "accounts/fireworks/models/qwen3p8-2p4t-a95b",
4986
- name: "Qwen3.8 2.4T A95B",
5209
+ "accounts/fireworks/routers/kimi-fast-latest": {
5210
+ id: "accounts/fireworks/routers/kimi-fast-latest",
5211
+ name: "Kimi Fast Latest",
4987
5212
  api: "anthropic-messages",
4988
5213
  provider: "fireworks",
4989
5214
  baseUrl: "https://api.fireworks.ai/inference",
4990
5215
  reasoning: true,
4991
- input: ["text"],
5216
+ input: ["text", "image"],
4992
5217
  cost: {
4993
- input: 2,
4994
- output: 6,
4995
- cacheRead: 0.25,
5218
+ input: 4.5,
5219
+ output: 22.5,
5220
+ cacheRead: 0.45,
4996
5221
  cacheWrite: 0,
4997
5222
  },
4998
- contextWindow: 262144,
5223
+ contextWindow: 1048576,
4999
5224
  maxTokens: 131072,
5000
5225
  },
5001
- "accounts/fireworks/models/qwen3p8-max": {
5002
- id: "accounts/fireworks/models/qwen3p8-max",
5003
- name: "Qwen3.8 Max",
5226
+ "accounts/fireworks/routers/kimi-k3-fast": {
5227
+ id: "accounts/fireworks/routers/kimi-k3-fast",
5228
+ name: "Kimi K3 Fast",
5004
5229
  api: "anthropic-messages",
5005
5230
  provider: "fireworks",
5006
5231
  baseUrl: "https://api.fireworks.ai/inference",
5007
5232
  reasoning: true,
5008
- input: ["text"],
5233
+ input: ["text", "image"],
5009
5234
  cost: {
5010
- input: 2,
5011
- output: 6,
5012
- cacheRead: 0.25,
5235
+ input: 4.5,
5236
+ output: 22.5,
5237
+ cacheRead: 0.45,
5013
5238
  cacheWrite: 0,
5014
5239
  },
5015
- contextWindow: 262144,
5240
+ contextWindow: 1048576,
5016
5241
  maxTokens: 131072,
5017
5242
  },
5018
- "accounts/fireworks/routers/glm-5p2-fast": {
5019
- id: "accounts/fireworks/routers/glm-5p2-fast",
5020
- name: "GLM 5.2 Fast",
5243
+ "accounts/fireworks/routers/kimi-latest": {
5244
+ id: "accounts/fireworks/routers/kimi-latest",
5245
+ name: "Kimi Latest",
5021
5246
  api: "anthropic-messages",
5022
5247
  provider: "fireworks",
5023
5248
  baseUrl: "https://api.fireworks.ai/inference",
5024
5249
  reasoning: true,
5025
- input: ["text"],
5250
+ input: ["text", "image"],
5026
5251
  cost: {
5027
- input: 2.1,
5028
- output: 6.6,
5029
- cacheRead: 0.21,
5252
+ input: 3,
5253
+ output: 15,
5254
+ cacheRead: 0.3,
5030
5255
  cacheWrite: 0,
5031
5256
  },
5032
- contextWindow: 1048575,
5257
+ contextWindow: 1048576,
5033
5258
  maxTokens: 131072,
5034
5259
  },
5035
- "accounts/fireworks/routers/glm-5p3-fast": {
5036
- id: "accounts/fireworks/routers/glm-5p3-fast",
5037
- name: "GLM 5.3 Fast",
5260
+ "accounts/fireworks/routers/minimax-latest": {
5261
+ id: "accounts/fireworks/routers/minimax-latest",
5262
+ name: "MiniMax Latest",
5038
5263
  api: "anthropic-messages",
5039
5264
  provider: "fireworks",
5040
5265
  baseUrl: "https://api.fireworks.ai/inference",
5041
5266
  reasoning: true,
5042
5267
  input: ["text"],
5043
5268
  cost: {
5044
- input: 2.1,
5045
- output: 6.6,
5046
- cacheRead: 0.39,
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": {
@@ -8399,6 +8624,23 @@ export const MODELS = {
8399
8624
  contextWindow: 1000000,
8400
8625
  maxTokens: 131072,
8401
8626
  },
8627
+ "zai-glm-5-3": {
8628
+ id: "zai-glm-5-3",
8629
+ name: "GLM-5.3",
8630
+ api: "mistral-conversations",
8631
+ provider: "mistral",
8632
+ baseUrl: "https://api.mistral.ai",
8633
+ reasoning: true,
8634
+ input: ["text"],
8635
+ cost: {
8636
+ input: 1.4,
8637
+ output: 4.4,
8638
+ cacheRead: 0.14,
8639
+ cacheWrite: 0,
8640
+ },
8641
+ contextWindow: 1000000,
8642
+ maxTokens: 131072,
8643
+ },
8402
8644
  },
8403
8645
  "moonshotai": {
8404
8646
  "kimi-k2.6": {
@@ -10203,7 +10445,7 @@ export const MODELS = {
10203
10445
  },
10204
10446
  "gpt-5.6-sol": {
10205
10447
  id: "gpt-5.6-sol",
10206
- name: "GPT-5.6 Sol (50% Off)",
10448
+ name: "GPT-5.6 Sol",
10207
10449
  api: "openai-responses",
10208
10450
  provider: "opencode",
10209
10451
  baseUrl: "https://opencode.ai/zen/v1",
@@ -10211,10 +10453,10 @@ export const MODELS = {
10211
10453
  thinkingLevelMap: { "off": null },
10212
10454
  input: ["text", "image"],
10213
10455
  cost: {
10214
- input: 2,
10215
- output: 10,
10216
- cacheRead: 0.2,
10217
- cacheWrite: 2.5,
10456
+ input: 4,
10457
+ output: 20,
10458
+ cacheRead: 0.4,
10459
+ cacheWrite: 5,
10218
10460
  },
10219
10461
  contextWindow: 1050000,
10220
10462
  maxTokens: 128000,
@@ -11521,7 +11763,7 @@ export const MODELS = {
11521
11763
  },
11522
11764
  "anthropic/claude-opus-5": {
11523
11765
  id: "anthropic/claude-opus-5",
11524
- name: "Claude Opus 5",
11766
+ name: "Anthropic: Claude Opus 5",
11525
11767
  api: "openai-completions",
11526
11768
  provider: "openrouter",
11527
11769
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11538,7 +11780,7 @@ export const MODELS = {
11538
11780
  },
11539
11781
  "anthropic/claude-opus-5:batch": {
11540
11782
  id: "anthropic/claude-opus-5:batch",
11541
- name: "Claude Opus 5 (batch)",
11783
+ name: "Anthropic: Claude Opus 5 (batch)",
11542
11784
  api: "openai-completions",
11543
11785
  provider: "openrouter",
11544
11786
  baseUrl: "https://openrouter.ai/api/v1",
@@ -11868,13 +12110,13 @@ export const MODELS = {
11868
12110
  reasoning: false,
11869
12111
  input: ["text"],
11870
12112
  cost: {
11871
- input: 0.2574,
11872
- output: 1.0287,
12113
+ input: 0.32,
12114
+ output: 0.8899999999999999,
11873
12115
  cacheRead: 0,
11874
12116
  cacheWrite: 0,
11875
12117
  },
11876
12118
  contextWindow: 163840,
11877
- maxTokens: 16000,
12119
+ maxTokens: 16384,
11878
12120
  },
11879
12121
  "deepseek/deepseek-chat-v3-0324": {
11880
12122
  id: "deepseek/deepseek-chat-v3-0324",
@@ -12006,9 +12248,9 @@ export const MODELS = {
12006
12248
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12007
12249
  input: ["text"],
12008
12250
  cost: {
12009
- input: 0.088606,
12010
- output: 0.177212,
12011
- cacheRead: 0.017721200000000003,
12251
+ input: 0.04956,
12252
+ output: 0.09912,
12253
+ cacheRead: 0.009911999999999999,
12012
12254
  cacheWrite: 0,
12013
12255
  },
12014
12256
  contextWindow: 1048576,
@@ -12025,9 +12267,9 @@ export const MODELS = {
12025
12267
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12026
12268
  input: ["text"],
12027
12269
  cost: {
12028
- input: 0.065,
12029
- output: 0.18,
12030
- cacheRead: 0.016,
12270
+ input: 0.06,
12271
+ output: 0.12,
12272
+ cacheRead: 0.012,
12031
12273
  cacheWrite: 0,
12032
12274
  },
12033
12275
  contextWindow: 1310720,
@@ -12052,6 +12294,25 @@ export const MODELS = {
12052
12294
  contextWindow: 1048576,
12053
12295
  maxTokens: 943718,
12054
12296
  },
12297
+ "deepseek/deepseek-v4-flash-0731:free": {
12298
+ id: "deepseek/deepseek-v4-flash-0731:free",
12299
+ name: "DeepSeek: DeepSeek V4 Flash 0731 (free)",
12300
+ api: "openai-completions",
12301
+ provider: "openrouter",
12302
+ baseUrl: "https://openrouter.ai/api/v1",
12303
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
12304
+ reasoning: true,
12305
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12306
+ input: ["text"],
12307
+ cost: {
12308
+ input: 0,
12309
+ output: 0,
12310
+ cacheRead: 0,
12311
+ cacheWrite: 0,
12312
+ },
12313
+ contextWindow: 1048576,
12314
+ maxTokens: 393216,
12315
+ },
12055
12316
  "deepseek/deepseek-v4-flash-vision-exp": {
12056
12317
  id: "deepseek/deepseek-v4-flash-vision-exp",
12057
12318
  name: "DeepSeek: DeepSeek V4 Flash Vision Exp",
@@ -12063,13 +12324,13 @@ export const MODELS = {
12063
12324
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12064
12325
  input: ["text", "image"],
12065
12326
  cost: {
12066
- input: 0.22,
12067
- output: 0.66,
12068
- cacheRead: 0.007,
12327
+ input: 0.21559999999999999,
12328
+ output: 0.6468,
12329
+ cacheRead: 0.00686,
12069
12330
  cacheWrite: 0,
12070
12331
  },
12071
12332
  contextWindow: 1048576,
12072
- maxTokens: 943718,
12333
+ maxTokens: 262144,
12073
12334
  },
12074
12335
  "deepseek/deepseek-v4-flash-vision-exp:batch": {
12075
12336
  id: "deepseek/deepseek-v4-flash-vision-exp:batch",
@@ -12101,13 +12362,13 @@ export const MODELS = {
12101
12362
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12102
12363
  input: ["text"],
12103
12364
  cost: {
12104
- input: 0.9552599999999999,
12105
- output: 1.9105199999999998,
12106
- cacheRead: 0.07960500000000001,
12365
+ input: 1.5999999999999999,
12366
+ output: 3.1999999999999997,
12367
+ cacheRead: 0.135,
12107
12368
  cacheWrite: 0,
12108
12369
  },
12109
12370
  contextWindow: 1048576,
12110
- maxTokens: 384000,
12371
+ maxTokens: 393216,
12111
12372
  },
12112
12373
  "deepseek/deepseek-v4-pro-0813": {
12113
12374
  id: "deepseek/deepseek-v4-pro-0813",
@@ -12120,13 +12381,13 @@ export const MODELS = {
12120
12381
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12121
12382
  input: ["text"],
12122
12383
  cost: {
12123
- input: 0.66,
12124
- output: 1.9800000000000002,
12125
- cacheRead: 0.06599999999999999,
12384
+ input: 0.57816,
12385
+ output: 1.73448,
12386
+ cacheRead: 0.018396000000000003,
12126
12387
  cacheWrite: 0,
12127
12388
  },
12128
12389
  contextWindow: 1048576,
12129
- maxTokens: 943718,
12390
+ maxTokens: 393216,
12130
12391
  },
12131
12392
  "deepseek/deepseek-v4-pro-0813:batch": {
12132
12393
  id: "deepseek/deepseek-v4-pro-0813:batch",
@@ -12285,23 +12546,6 @@ export const MODELS = {
12285
12546
  contextWindow: 1048576,
12286
12547
  maxTokens: 65536,
12287
12548
  },
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
12549
  "google/gemini-2.5-pro:batch": {
12306
12550
  id: "google/gemini-2.5-pro:batch",
12307
12551
  name: "Google: Gemini 2.5 Pro (batch)",
@@ -12685,13 +12929,13 @@ export const MODELS = {
12685
12929
  reasoning: true,
12686
12930
  input: ["text", "image"],
12687
12931
  cost: {
12688
- input: 0.041999999999999996,
12689
- output: 0.22,
12690
- cacheRead: 0,
12932
+ input: 0.09,
12933
+ output: 0.3,
12934
+ cacheRead: 0.049999999999999996,
12691
12935
  cacheWrite: 0,
12692
12936
  },
12693
12937
  contextWindow: 262144,
12694
- maxTokens: 32768,
12938
+ maxTokens: 235929,
12695
12939
  },
12696
12940
  "google/gemma-4-26b-a4b-it:free": {
12697
12941
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12727,23 +12971,6 @@ export const MODELS = {
12727
12971
  contextWindow: 262144,
12728
12972
  maxTokens: 16384,
12729
12973
  },
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
12974
  "google/gemma-4-31b-it:free": {
12748
12975
  id: "google/gemma-4-31b-it:free",
12749
12976
  name: "Google: Gemma 4 31B (free)",
@@ -12880,6 +13107,23 @@ export const MODELS = {
12880
13107
  contextWindow: 262144,
12881
13108
  maxTokens: 32768,
12882
13109
  },
13110
+ "inclusionai/ling-3.0-flash-vl": {
13111
+ id: "inclusionai/ling-3.0-flash-vl",
13112
+ name: "inclusionAI: Ling 3.0 Flash VL",
13113
+ api: "openai-completions",
13114
+ provider: "openrouter",
13115
+ baseUrl: "https://openrouter.ai/api/v1",
13116
+ reasoning: true,
13117
+ input: ["text", "image"],
13118
+ cost: {
13119
+ input: 0.06,
13120
+ output: 0.18,
13121
+ cacheRead: 0.012,
13122
+ cacheWrite: 0,
13123
+ },
13124
+ contextWindow: 131072,
13125
+ maxTokens: 32768,
13126
+ },
12883
13127
  "inclusionai/ling-3.0-flash-vl:free": {
12884
13128
  id: "inclusionai/ling-3.0-flash-vl:free",
12885
13129
  name: "inclusionAI: Ling 3.0 Flash VL (free)",
@@ -13025,13 +13269,13 @@ export const MODELS = {
13025
13269
  reasoning: false,
13026
13270
  input: ["text", "image"],
13027
13271
  cost: {
13028
- input: 0.19999999999999998,
13029
- output: 0.696,
13272
+ input: 0.1875,
13273
+ output: 0.6525,
13030
13274
  cacheRead: 0,
13031
13275
  cacheWrite: 0,
13032
13276
  },
13033
13277
  contextWindow: 1048576,
13034
- maxTokens: 115200,
13278
+ maxTokens: 16384,
13035
13279
  },
13036
13280
  "meta-llama/llama-4-scout": {
13037
13281
  id: "meta-llama/llama-4-scout",
@@ -13059,8 +13303,8 @@ export const MODELS = {
13059
13303
  reasoning: true,
13060
13304
  input: ["text", "image"],
13061
13305
  cost: {
13062
- input: 0.3,
13063
- output: 1.1,
13306
+ input: 0.35,
13307
+ output: 1.5,
13064
13308
  cacheRead: 0.04,
13065
13309
  cacheWrite: 0,
13066
13310
  },
@@ -13178,7 +13422,7 @@ export const MODELS = {
13178
13422
  reasoning: true,
13179
13423
  input: ["text"],
13180
13424
  cost: {
13181
- input: 0.55,
13425
+ input: 0.39999999999999997,
13182
13426
  output: 2.2,
13183
13427
  cacheRead: 0,
13184
13428
  cacheWrite: 0,
@@ -13229,13 +13473,13 @@ export const MODELS = {
13229
13473
  reasoning: true,
13230
13474
  input: ["text"],
13231
13475
  cost: {
13232
- input: 0.3,
13233
- output: 1.2,
13234
- cacheRead: 0.03,
13476
+ input: 0.27,
13477
+ output: 1.08,
13478
+ cacheRead: 0.027,
13235
13479
  cacheWrite: 0,
13236
13480
  },
13237
13481
  contextWindow: 204800,
13238
- maxTokens: 131072,
13482
+ maxTokens: 128000,
13239
13483
  },
13240
13484
  "minimax/minimax-m2.7": {
13241
13485
  id: "minimax/minimax-m2.7",
@@ -13433,30 +13677,13 @@ export const MODELS = {
13433
13677
  reasoning: false,
13434
13678
  input: ["text"],
13435
13679
  cost: {
13436
- input: 2,
13437
- output: 6,
13438
- cacheRead: 0.19999999999999998,
13439
- cacheWrite: 0,
13440
- },
13441
- contextWindow: 131072,
13442
- maxTokens: 104857,
13443
- },
13444
- "mistralai/mistral-large-2512": {
13445
- id: "mistralai/mistral-large-2512",
13446
- name: "Mistral: Mistral Large 3 2512",
13447
- api: "openai-completions",
13448
- provider: "openrouter",
13449
- baseUrl: "https://openrouter.ai/api/v1",
13450
- reasoning: false,
13451
- input: ["text", "image"],
13452
- cost: {
13453
- input: 0.5,
13454
- output: 1.5,
13455
- cacheRead: 0.049999999999999996,
13680
+ input: 2,
13681
+ output: 6,
13682
+ cacheRead: 0.19999999999999998,
13456
13683
  cacheWrite: 0,
13457
13684
  },
13458
- contextWindow: 262144,
13459
- maxTokens: 209715,
13685
+ contextWindow: 131072,
13686
+ maxTokens: 104857,
13460
13687
  },
13461
13688
  "mistralai/mistral-large-2512:batch": {
13462
13689
  id: "mistralai/mistral-large-2512:batch",
@@ -13637,12 +13864,12 @@ export const MODELS = {
13637
13864
  reasoning: false,
13638
13865
  input: ["text", "image"],
13639
13866
  cost: {
13640
- input: 0.075,
13641
- output: 0.19999999999999998,
13867
+ input: 0.09375,
13868
+ output: 0.25,
13642
13869
  cacheRead: 0,
13643
13870
  cacheWrite: 0,
13644
13871
  },
13645
- contextWindow: 131072,
13872
+ contextWindow: 256000,
13646
13873
  maxTokens: 16384,
13647
13874
  },
13648
13875
  "mistralai/mixtral-8x22b-instruct": {
@@ -13694,7 +13921,7 @@ export const MODELS = {
13694
13921
  cacheWrite: 0,
13695
13922
  },
13696
13923
  contextWindow: 131072,
13697
- maxTokens: 100352,
13924
+ maxTokens: 98304,
13698
13925
  },
13699
13926
  "moonshotai/kimi-k2-0905": {
13700
13927
  id: "moonshotai/kimi-k2-0905",
@@ -13711,7 +13938,7 @@ export const MODELS = {
13711
13938
  cacheWrite: 0,
13712
13939
  },
13713
13940
  contextWindow: 262144,
13714
- maxTokens: 100352,
13941
+ maxTokens: 98304,
13715
13942
  },
13716
13943
  "moonshotai/kimi-k2-thinking": {
13717
13944
  id: "moonshotai/kimi-k2-thinking",
@@ -13728,7 +13955,7 @@ export const MODELS = {
13728
13955
  cacheWrite: 0,
13729
13956
  },
13730
13957
  contextWindow: 262144,
13731
- maxTokens: 100352,
13958
+ maxTokens: 98304,
13732
13959
  },
13733
13960
  "moonshotai/kimi-k2.5": {
13734
13961
  id: "moonshotai/kimi-k2.5",
@@ -13773,9 +14000,9 @@ export const MODELS = {
13773
14000
  reasoning: true,
13774
14001
  input: ["text", "image"],
13775
14002
  cost: {
13776
- input: 0.71,
13777
- output: 3.5,
13778
- cacheRead: 0.15,
14003
+ input: 0.7062,
14004
+ output: 3.21,
14005
+ cacheRead: 0.18,
13779
14006
  cacheWrite: 0,
13780
14007
  },
13781
14008
  contextWindow: 262144,
@@ -13790,9 +14017,9 @@ export const MODELS = {
13790
14017
  reasoning: true,
13791
14018
  input: ["text", "image"],
13792
14019
  cost: {
13793
- input: 2.34,
13794
- output: 11.7,
13795
- cacheRead: 0.261,
14020
+ input: 2.0999999999999996,
14021
+ output: 10.950000000000001,
14022
+ cacheRead: 0.22999999999999998,
13796
14023
  cacheWrite: 0,
13797
14024
  },
13798
14025
  contextWindow: 1048576,
@@ -13858,9 +14085,9 @@ export const MODELS = {
13858
14085
  reasoning: true,
13859
14086
  input: ["text"],
13860
14087
  cost: {
13861
- input: 0.049999999999999996,
13862
- output: 0.19999999999999998,
13863
- cacheRead: 0.03,
14088
+ input: 0.06,
14089
+ output: 0.24,
14090
+ cacheRead: 0,
13864
14091
  cacheWrite: 0,
13865
14092
  },
13866
14093
  contextWindow: 262144,
@@ -13892,13 +14119,13 @@ export const MODELS = {
13892
14119
  reasoning: true,
13893
14120
  input: ["text"],
13894
14121
  cost: {
13895
- input: 0.08499999999999999,
13896
- output: 0.39999999999999997,
14122
+ input: 0.08,
14123
+ output: 0.44999999999999996,
13897
14124
  cacheRead: 0,
13898
14125
  cacheWrite: 0,
13899
14126
  },
13900
14127
  contextWindow: 262144,
13901
- maxTokens: 16384,
14128
+ maxTokens: 235929,
13902
14129
  },
13903
14130
  "nvidia/nemotron-3-super-120b-a12b:free": {
13904
14131
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -15152,13 +15379,13 @@ export const MODELS = {
15152
15379
  reasoning: true,
15153
15380
  input: ["text"],
15154
15381
  cost: {
15155
- input: 0.037,
15156
- output: 0.16999999999999998,
15157
- cacheRead: 0,
15382
+ input: 0.15,
15383
+ output: 0.6,
15384
+ cacheRead: 0.075,
15158
15385
  cacheWrite: 0,
15159
15386
  },
15160
15387
  contextWindow: 131072,
15161
- maxTokens: 117964,
15388
+ maxTokens: 65536,
15162
15389
  },
15163
15390
  "openai/gpt-oss-120b:batch": {
15164
15391
  id: "openai/gpt-oss-120b:batch",
@@ -15577,13 +15804,13 @@ export const MODELS = {
15577
15804
  reasoning: true,
15578
15805
  input: ["text"],
15579
15806
  cost: {
15580
- input: 0.22749999999999998,
15581
- output: 0.9099999999999999,
15807
+ input: 0.12,
15808
+ output: 0.24,
15582
15809
  cacheRead: 0,
15583
15810
  cacheWrite: 0,
15584
15811
  },
15585
15812
  contextWindow: 131072,
15586
- maxTokens: 8192,
15813
+ maxTokens: 16384,
15587
15814
  },
15588
15815
  "qwen/qwen3-235b-a22b": {
15589
15816
  id: "qwen/qwen3-235b-a22b",
@@ -15611,13 +15838,13 @@ export const MODELS = {
15611
15838
  reasoning: false,
15612
15839
  input: ["text"],
15613
15840
  cost: {
15614
- input: 0.22,
15615
- output: 0.88,
15616
- cacheRead: 0,
15841
+ input: 0.0875,
15842
+ output: 0.35,
15843
+ cacheRead: 0.0175,
15617
15844
  cacheWrite: 0,
15618
15845
  },
15619
15846
  contextWindow: 262144,
15620
- maxTokens: 16384,
15847
+ maxTokens: 235929,
15621
15848
  },
15622
15849
  "qwen/qwen3-235b-a22b-thinking-2507": {
15623
15850
  id: "qwen/qwen3-235b-a22b-thinking-2507",
@@ -15662,13 +15889,13 @@ export const MODELS = {
15662
15889
  reasoning: false,
15663
15890
  input: ["text"],
15664
15891
  cost: {
15665
- input: 0.09,
15666
- output: 0.3,
15892
+ input: 0.04815,
15893
+ output: 0.19305,
15667
15894
  cacheRead: 0,
15668
15895
  cacheWrite: 0,
15669
15896
  },
15670
15897
  contextWindow: 262144,
15671
- maxTokens: 235929,
15898
+ maxTokens: 32000,
15672
15899
  },
15673
15900
  "qwen/qwen3-30b-a3b-thinking-2507": {
15674
15901
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15872,7 +16099,7 @@ export const MODELS = {
15872
16099
  cacheWrite: 0,
15873
16100
  },
15874
16101
  contextWindow: 262144,
15875
- maxTokens: 235929,
16102
+ maxTokens: 32768,
15876
16103
  },
15877
16104
  "qwen/qwen3-vl-235b-a22b-instruct": {
15878
16105
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -15917,13 +16144,13 @@ export const MODELS = {
15917
16144
  reasoning: false,
15918
16145
  input: ["text", "image"],
15919
16146
  cost: {
15920
- input: 0.15,
15921
- output: 0.6,
16147
+ input: 0.13,
16148
+ output: 0.52,
15922
16149
  cacheRead: 0,
15923
16150
  cacheWrite: 0,
15924
16151
  },
15925
16152
  contextWindow: 262144,
15926
- maxTokens: 16384,
16153
+ maxTokens: 32768,
15927
16154
  },
15928
16155
  "qwen/qwen3-vl-30b-a3b-thinking": {
15929
16156
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -16036,13 +16263,13 @@ export const MODELS = {
16036
16263
  reasoning: true,
16037
16264
  input: ["text", "image"],
16038
16265
  cost: {
16039
- input: 0.3125,
16040
- output: 1.25,
16041
- cacheRead: 0.15625,
16266
+ input: 0.1625,
16267
+ output: 1.3,
16268
+ cacheRead: 0,
16042
16269
  cacheWrite: 0,
16043
16270
  },
16044
16271
  contextWindow: 262144,
16045
- maxTokens: 16384,
16272
+ maxTokens: 65536,
16046
16273
  },
16047
16274
  "qwen/qwen3.5-397b-a17b": {
16048
16275
  id: "qwen/qwen3.5-397b-a17b",
@@ -16325,14 +16552,31 @@ export const MODELS = {
16325
16552
  reasoning: true,
16326
16553
  input: ["text", "image"],
16327
16554
  cost: {
16328
- input: 0.42,
16329
- output: 3,
16330
- cacheRead: 0.08499999999999999,
16555
+ input: 0.21400000000000002,
16556
+ output: 2.5500000000000003,
16557
+ cacheRead: 0.15,
16331
16558
  cacheWrite: 0,
16332
16559
  },
16333
16560
  contextWindow: 1000000,
16334
16561
  maxTokens: 131072,
16335
16562
  },
16563
+ "qwen/qwen3.8-27b:free": {
16564
+ id: "qwen/qwen3.8-27b:free",
16565
+ name: "Qwen: Qwen3.8 27B (free)",
16566
+ api: "openai-completions",
16567
+ provider: "openrouter",
16568
+ baseUrl: "https://openrouter.ai/api/v1",
16569
+ reasoning: true,
16570
+ input: ["text", "image"],
16571
+ cost: {
16572
+ input: 0,
16573
+ output: 0,
16574
+ cacheRead: 0,
16575
+ cacheWrite: 0,
16576
+ },
16577
+ contextWindow: 262144,
16578
+ maxTokens: 235929,
16579
+ },
16336
16580
  "qwen/qwen3.8-flash": {
16337
16581
  id: "qwen/qwen3.8-flash",
16338
16582
  name: "Qwen: Qwen3.8 Flash",
@@ -16529,9 +16773,9 @@ export const MODELS = {
16529
16773
  reasoning: true,
16530
16774
  input: ["text"],
16531
16775
  cost: {
16532
- input: 0.13199999999999998,
16533
- output: 0.5279999999999999,
16534
- cacheRead: 0.032999999999999995,
16776
+ input: 0.0825,
16777
+ output: 0.33,
16778
+ cacheRead: 0.020625,
16535
16779
  cacheWrite: 0,
16536
16780
  },
16537
16781
  contextWindow: 262144,
@@ -16554,23 +16798,6 @@ export const MODELS = {
16554
16798
  contextWindow: 1048576,
16555
16799
  maxTokens: 64000,
16556
16800
  },
16557
- "thedrummer/unslopnemo-12b": {
16558
- id: "thedrummer/unslopnemo-12b",
16559
- name: "TheDrummer: UnslopNemo 12B",
16560
- api: "openai-completions",
16561
- provider: "openrouter",
16562
- baseUrl: "https://openrouter.ai/api/v1",
16563
- reasoning: false,
16564
- input: ["text"],
16565
- cost: {
16566
- input: 0.39999999999999997,
16567
- output: 0.39999999999999997,
16568
- cacheRead: 0,
16569
- cacheWrite: 0,
16570
- },
16571
- contextWindow: 1024000,
16572
- maxTokens: 819200,
16573
- },
16574
16801
  "thinkingmachines/inkling": {
16575
16802
  id: "thinkingmachines/inkling",
16576
16803
  name: "Thinking Machines: Inkling",
@@ -16586,7 +16813,7 @@ export const MODELS = {
16586
16813
  cacheWrite: 0,
16587
16814
  },
16588
16815
  contextWindow: 1048576,
16589
- maxTokens: 32768,
16816
+ maxTokens: 471859,
16590
16817
  },
16591
16818
  "thinkingmachines/inkling-small": {
16592
16819
  id: "thinkingmachines/inkling-small",
@@ -16605,23 +16832,6 @@ export const MODELS = {
16605
16832
  contextWindow: 1048576,
16606
16833
  maxTokens: 262144,
16607
16834
  },
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
16835
  "thinkingmachines/inkling-small:free": {
16626
16836
  id: "thinkingmachines/inkling-small:free",
16627
16837
  name: "Thinking Machines: Inkling Small (free)",
@@ -16673,6 +16883,23 @@ export const MODELS = {
16673
16883
  contextWindow: 1048576,
16674
16884
  maxTokens: 262144,
16675
16885
  },
16886
+ "unbiased/pareto": {
16887
+ id: "unbiased/pareto",
16888
+ name: "Pareto",
16889
+ api: "openai-completions",
16890
+ provider: "openrouter",
16891
+ baseUrl: "https://openrouter.ai/api/v1",
16892
+ reasoning: false,
16893
+ input: ["text", "image"],
16894
+ cost: {
16895
+ input: 2.5,
16896
+ output: 7.5,
16897
+ cacheRead: 0.25,
16898
+ cacheWrite: 0,
16899
+ },
16900
+ contextWindow: 262144,
16901
+ maxTokens: 131072,
16902
+ },
16676
16903
  "upstage/solar-pro-3": {
16677
16904
  id: "upstage/solar-pro-3",
16678
16905
  name: "Upstage: Solar Pro 3",
@@ -17022,9 +17249,9 @@ export const MODELS = {
17022
17249
  reasoning: true,
17023
17250
  input: ["text"],
17024
17251
  cost: {
17025
- input: 0.966,
17026
- output: 3.036,
17027
- cacheRead: 0.1932,
17252
+ input: 0.5544,
17253
+ output: 1.7424,
17254
+ cacheRead: 0.10296000000000001,
17028
17255
  cacheWrite: 0,
17029
17256
  },
17030
17257
  contextWindow: 1048576,
@@ -17056,13 +17283,13 @@ export const MODELS = {
17056
17283
  reasoning: true,
17057
17284
  input: ["text"],
17058
17285
  cost: {
17059
- input: 1.4,
17060
- output: 4.4,
17061
- cacheRead: 0.26,
17286
+ input: 0.9099999999999999,
17287
+ output: 2.8600000000000003,
17288
+ cacheRead: 0.16899999999999998,
17062
17289
  cacheWrite: 0,
17063
17290
  },
17064
17291
  contextWindow: 1310720,
17065
- maxTokens: 943718,
17292
+ maxTokens: 131072,
17066
17293
  },
17067
17294
  "z-ai/glm-5.3-flash": {
17068
17295
  id: "z-ai/glm-5.3-flash",
@@ -17073,9 +17300,9 @@ export const MODELS = {
17073
17300
  reasoning: true,
17074
17301
  input: ["text", "image"],
17075
17302
  cost: {
17076
- input: 0.15,
17077
- output: 0.5,
17078
- cacheRead: 0.03,
17303
+ input: 0.09,
17304
+ output: 0.3,
17305
+ cacheRead: 0.018,
17079
17306
  cacheWrite: 0,
17080
17307
  },
17081
17308
  contextWindow: 1310720,
@@ -17152,7 +17379,7 @@ export const MODELS = {
17152
17379
  },
17153
17380
  "~anthropic/claude-haiku-latest": {
17154
17381
  id: "~anthropic/claude-haiku-latest",
17155
- name: "Anthropic Claude Haiku Latest",
17382
+ name: "Anthropic: Claude Haiku Latest",
17156
17383
  api: "openai-completions",
17157
17384
  provider: "openrouter",
17158
17385
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17186,7 +17413,7 @@ export const MODELS = {
17186
17413
  },
17187
17414
  "~anthropic/claude-sonnet-latest": {
17188
17415
  id: "~anthropic/claude-sonnet-latest",
17189
- name: "Anthropic Claude Sonnet Latest",
17416
+ name: "Anthropic: Claude Sonnet Latest",
17190
17417
  api: "openai-completions",
17191
17418
  provider: "openrouter",
17192
17419
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17201,9 +17428,43 @@ export const MODELS = {
17201
17428
  contextWindow: 1000000,
17202
17429
  maxTokens: 128000,
17203
17430
  },
17431
+ "~deepseek/deepseek-flash-latest": {
17432
+ id: "~deepseek/deepseek-flash-latest",
17433
+ name: "DeepSeek: DeepSeek Flash Latest",
17434
+ api: "openai-completions",
17435
+ provider: "openrouter",
17436
+ baseUrl: "https://openrouter.ai/api/v1",
17437
+ reasoning: true,
17438
+ input: ["text", "image"],
17439
+ cost: {
17440
+ input: 0.15,
17441
+ output: 0.6,
17442
+ cacheRead: 0.015,
17443
+ cacheWrite: 0,
17444
+ },
17445
+ contextWindow: 1048576,
17446
+ maxTokens: 393216,
17447
+ },
17448
+ "~deepseek/deepseek-pro-latest": {
17449
+ id: "~deepseek/deepseek-pro-latest",
17450
+ name: "DeepSeek: DeepSeek Pro Latest",
17451
+ api: "openai-completions",
17452
+ provider: "openrouter",
17453
+ baseUrl: "https://openrouter.ai/api/v1",
17454
+ reasoning: true,
17455
+ input: ["text"],
17456
+ cost: {
17457
+ input: 0.57816,
17458
+ output: 1.73448,
17459
+ cacheRead: 0.018396000000000003,
17460
+ cacheWrite: 0,
17461
+ },
17462
+ contextWindow: 1048576,
17463
+ maxTokens: 393216,
17464
+ },
17204
17465
  "~deepseek/deepseek-v4-flash-latest": {
17205
17466
  id: "~deepseek/deepseek-v4-flash-latest",
17206
- name: "DeepSeek V4 Flash Latest",
17467
+ name: "DeepSeek: DeepSeek V4 Flash Latest",
17207
17468
  api: "openai-completions",
17208
17469
  provider: "openrouter",
17209
17470
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17212,17 +17473,17 @@ export const MODELS = {
17212
17473
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
17213
17474
  input: ["text"],
17214
17475
  cost: {
17215
- input: 0.049999999999999996,
17216
- output: 0.16,
17217
- cacheRead: 0.013000000000000001,
17476
+ input: 0.055,
17477
+ output: 0.165,
17478
+ cacheRead: 0.00175,
17218
17479
  cacheWrite: 0,
17219
17480
  },
17220
17481
  contextWindow: 1310720,
17221
- maxTokens: 393216,
17482
+ maxTokens: 384000,
17222
17483
  },
17223
17484
  "~google/gemini-flash-latest": {
17224
17485
  id: "~google/gemini-flash-latest",
17225
- name: "Google Gemini Flash Latest",
17486
+ name: "Google: Gemini Flash Latest",
17226
17487
  api: "openai-completions",
17227
17488
  provider: "openrouter",
17228
17489
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17239,7 +17500,7 @@ export const MODELS = {
17239
17500
  },
17240
17501
  "~google/gemini-pro-latest": {
17241
17502
  id: "~google/gemini-pro-latest",
17242
- name: "Google Gemini Pro Latest",
17503
+ name: "Google: Gemini Pro Latest",
17243
17504
  api: "openai-completions",
17244
17505
  provider: "openrouter",
17245
17506
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17256,41 +17517,58 @@ export const MODELS = {
17256
17517
  },
17257
17518
  "~moonshotai/kimi-latest": {
17258
17519
  id: "~moonshotai/kimi-latest",
17259
- name: "MoonshotAI Kimi Latest",
17520
+ name: "MoonshotAI: Kimi Latest",
17260
17521
  api: "openai-completions",
17261
17522
  provider: "openrouter",
17262
17523
  baseUrl: "https://openrouter.ai/api/v1",
17263
17524
  reasoning: true,
17264
17525
  input: ["text", "image"],
17265
17526
  cost: {
17266
- input: 2.34,
17267
- output: 11.7,
17268
- cacheRead: 0.261,
17527
+ input: 2.0999999999999996,
17528
+ output: 10.950000000000001,
17529
+ cacheRead: 0.22999999999999998,
17269
17530
  cacheWrite: 0,
17270
17531
  },
17271
17532
  contextWindow: 1048576,
17272
17533
  maxTokens: 943718,
17273
17534
  },
17274
- "~openai/gpt-latest": {
17275
- id: "~openai/gpt-latest",
17276
- name: "OpenAI GPT Latest",
17535
+ "~openai/gpt-astra-latest": {
17536
+ id: "~openai/gpt-astra-latest",
17537
+ name: "OpenAI: GPT Astra Latest",
17277
17538
  api: "openai-completions",
17278
17539
  provider: "openrouter",
17279
17540
  baseUrl: "https://openrouter.ai/api/v1",
17280
17541
  reasoning: true,
17281
17542
  input: ["text", "image"],
17282
17543
  cost: {
17283
- input: 2,
17284
- output: 10,
17285
- cacheRead: 0.19999999999999998,
17286
- cacheWrite: 2.5,
17544
+ input: 10,
17545
+ output: 50,
17546
+ cacheRead: 1,
17547
+ cacheWrite: 12.5,
17548
+ },
17549
+ contextWindow: 1050000,
17550
+ maxTokens: 128000,
17551
+ },
17552
+ "~openai/gpt-luna-latest": {
17553
+ id: "~openai/gpt-luna-latest",
17554
+ name: "OpenAI: GPT Luna Latest",
17555
+ api: "openai-completions",
17556
+ provider: "openrouter",
17557
+ baseUrl: "https://openrouter.ai/api/v1",
17558
+ reasoning: true,
17559
+ input: ["text", "image"],
17560
+ cost: {
17561
+ input: 0.19999999999999998,
17562
+ output: 1.2,
17563
+ cacheRead: 0.02,
17564
+ cacheWrite: 0.25,
17287
17565
  },
17288
17566
  contextWindow: 1050000,
17289
17567
  maxTokens: 128000,
17290
17568
  },
17291
17569
  "~openai/gpt-mini-latest": {
17292
17570
  id: "~openai/gpt-mini-latest",
17293
- name: "OpenAI GPT Mini Latest",
17571
+ name: "OpenAI: GPT Mini Latest",
17294
17572
  api: "openai-completions",
17295
17573
  provider: "openrouter",
17296
17574
  baseUrl: "https://openrouter.ai/api/v1",
@@ -17305,6 +17583,40 @@ export const MODELS = {
17305
17583
  contextWindow: 400000,
17306
17584
  maxTokens: 128000,
17307
17585
  },
17586
+ "~openai/gpt-sol-latest": {
17587
+ id: "~openai/gpt-sol-latest",
17588
+ name: "OpenAI: GPT Sol Latest",
17589
+ api: "openai-completions",
17590
+ provider: "openrouter",
17591
+ baseUrl: "https://openrouter.ai/api/v1",
17592
+ reasoning: true,
17593
+ input: ["text", "image"],
17594
+ cost: {
17595
+ input: 2,
17596
+ output: 10,
17597
+ cacheRead: 0.19999999999999998,
17598
+ cacheWrite: 2.5,
17599
+ },
17600
+ contextWindow: 1050000,
17601
+ maxTokens: 128000,
17602
+ },
17603
+ "~openai/gpt-terra-latest": {
17604
+ id: "~openai/gpt-terra-latest",
17605
+ name: "OpenAI: GPT Terra Latest",
17606
+ api: "openai-completions",
17607
+ provider: "openrouter",
17608
+ baseUrl: "https://openrouter.ai/api/v1",
17609
+ reasoning: true,
17610
+ input: ["text", "image"],
17611
+ cost: {
17612
+ input: 2,
17613
+ output: 12,
17614
+ cacheRead: 0.19999999999999998,
17615
+ cacheWrite: 2.5,
17616
+ },
17617
+ contextWindow: 1050000,
17618
+ maxTokens: 128000,
17619
+ },
17308
17620
  "~x-ai/grok-latest": {
17309
17621
  id: "~x-ai/grok-latest",
17310
17622
  name: "xAI: Grok Latest",
@@ -17348,13 +17660,13 @@ export const MODELS = {
17348
17660
  reasoning: true,
17349
17661
  input: ["text"],
17350
17662
  cost: {
17351
- input: 0.9700000000000001,
17352
- output: 3.3077,
17353
- cacheRead: 0.19399999999999998,
17663
+ input: 0.8987999999999999,
17664
+ output: 2.8248,
17665
+ cacheRead: 0.16691999999999999,
17354
17666
  cacheWrite: 0,
17355
17667
  },
17356
17668
  contextWindow: 1310720,
17357
- maxTokens: 943718,
17669
+ maxTokens: 131072,
17358
17670
  },
17359
17671
  },
17360
17672
  "together": {
@@ -17527,6 +17839,25 @@ export const MODELS = {
17527
17839
  contextWindow: 1048576,
17528
17840
  maxTokens: 384000,
17529
17841
  },
17842
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
17843
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
17844
+ name: "DeepSeek V4.1 Flash",
17845
+ api: "openai-completions",
17846
+ provider: "together",
17847
+ baseUrl: "https://api.together.ai/v1",
17848
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
17849
+ reasoning: true,
17850
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
17851
+ input: ["text", "image"],
17852
+ cost: {
17853
+ input: 0.3,
17854
+ output: 1.2,
17855
+ cacheRead: 0.006,
17856
+ cacheWrite: 0,
17857
+ },
17858
+ contextWindow: 1048576,
17859
+ maxTokens: 384000,
17860
+ },
17530
17861
  "google/gemma-4-31B-it": {
17531
17862
  id: "google/gemma-4-31B-it",
17532
17863
  name: "Gemma 4 31B Instruct",
@@ -17835,7 +18166,7 @@ export const MODELS = {
17835
18166
  cost: {
17836
18167
  input: 1.3,
17837
18168
  output: 7.8,
17838
- cacheRead: 0.26,
18169
+ cacheRead: 0.13,
17839
18170
  cacheWrite: 1.625,
17840
18171
  },
17841
18172
  contextWindow: 240000,
@@ -17991,8 +18322,8 @@ export const MODELS = {
17991
18322
  cacheRead: 0,
17992
18323
  cacheWrite: 0,
17993
18324
  },
17994
- contextWindow: 131072,
17995
- maxTokens: 32768,
18325
+ contextWindow: 262114,
18326
+ maxTokens: 262114,
17996
18327
  },
17997
18328
  "alibaba/qwen3-next-80b-a3b-thinking": {
17998
18329
  id: "alibaba/qwen3-next-80b-a3b-thinking",
@@ -18008,8 +18339,8 @@ export const MODELS = {
18008
18339
  cacheRead: 0,
18009
18340
  cacheWrite: 0,
18010
18341
  },
18011
- contextWindow: 131072,
18012
- maxTokens: 32768,
18342
+ contextWindow: 262144,
18343
+ maxTokens: 262144,
18013
18344
  },
18014
18345
  "alibaba/qwen3-vl-235b-a22b-instruct": {
18015
18346
  id: "alibaba/qwen3-vl-235b-a22b-instruct",
@@ -18073,7 +18404,7 @@ export const MODELS = {
18073
18404
  cost: {
18074
18405
  input: 0.09999999999999999,
18075
18406
  output: 0.39999999999999997,
18076
- cacheRead: 0.001,
18407
+ cacheRead: 0.01,
18077
18408
  cacheWrite: 0.125,
18078
18409
  },
18079
18410
  contextWindow: 1000000,
@@ -18089,7 +18420,7 @@ export const MODELS = {
18089
18420
  input: ["text", "image"],
18090
18421
  cost: {
18091
18422
  input: 0.39999999999999997,
18092
- output: 2.4,
18423
+ output: 2.5,
18093
18424
  cacheRead: 0.04,
18094
18425
  cacheWrite: 0.5,
18095
18426
  },
@@ -18124,7 +18455,7 @@ export const MODELS = {
18124
18455
  cost: {
18125
18456
  input: 0.5,
18126
18457
  output: 3,
18127
- cacheRead: 0.09999999999999999,
18458
+ cacheRead: 0.049999999999999996,
18128
18459
  cacheWrite: 0.625,
18129
18460
  },
18130
18461
  contextWindow: 1000000,
@@ -18224,7 +18555,7 @@ export const MODELS = {
18224
18555
  reasoning: true,
18225
18556
  input: ["text", "image"],
18226
18557
  cost: {
18227
- input: 0.16,
18558
+ input: 0.15,
18228
18559
  output: 0.47,
18229
18560
  cacheRead: 0.016,
18230
18561
  cacheWrite: 0.19999999999999998,
@@ -18244,9 +18575,9 @@ export const MODELS = {
18244
18575
  input: 2,
18245
18576
  output: 6,
18246
18577
  cacheRead: 0.25,
18247
- cacheWrite: 2.5,
18578
+ cacheWrite: 0,
18248
18579
  },
18249
- contextWindow: 1000000,
18580
+ contextWindow: 262144,
18250
18581
  maxTokens: 128000,
18251
18582
  },
18252
18583
  "alibaba/qwen3.8-max-0902": {
@@ -18266,6 +18597,23 @@ export const MODELS = {
18266
18597
  contextWindow: 991000,
18267
18598
  maxTokens: 128000,
18268
18599
  },
18600
+ "alibaba/qwen3.8-omni-flash": {
18601
+ id: "alibaba/qwen3.8-omni-flash",
18602
+ name: "Qwen 3.8 Omni Flash",
18603
+ api: "anthropic-messages",
18604
+ provider: "vercel-ai-gateway",
18605
+ baseUrl: "https://ai-gateway.vercel.sh",
18606
+ reasoning: true,
18607
+ input: ["text", "image"],
18608
+ cost: {
18609
+ input: 0.15,
18610
+ output: 0.47,
18611
+ cacheRead: 0.016,
18612
+ cacheWrite: 0,
18613
+ },
18614
+ contextWindow: 1000000,
18615
+ maxTokens: 131072,
18616
+ },
18269
18617
  "amazon/nova-2-lite": {
18270
18618
  id: "amazon/nova-2-lite",
18271
18619
  name: "Nova 2 Lite",
@@ -18663,6 +19011,23 @@ export const MODELS = {
18663
19011
  contextWindow: 256000,
18664
19012
  maxTokens: 64000,
18665
19013
  },
19014
+ "bytedance/seed-2.1-turbo": {
19015
+ id: "bytedance/seed-2.1-turbo",
19016
+ name: "Seed 2.1 Turbo",
19017
+ api: "anthropic-messages",
19018
+ provider: "vercel-ai-gateway",
19019
+ baseUrl: "https://ai-gateway.vercel.sh",
19020
+ reasoning: true,
19021
+ input: ["text", "image"],
19022
+ cost: {
19023
+ input: 0.5,
19024
+ output: 2.5,
19025
+ cacheRead: 0.09999999999999999,
19026
+ cacheWrite: 0,
19027
+ },
19028
+ contextWindow: 262144,
19029
+ maxTokens: 262144,
19030
+ },
18666
19031
  "cohere/command-a": {
18667
19032
  id: "cohere/command-a",
18668
19033
  name: "Command A",
@@ -19148,9 +19513,9 @@ export const MODELS = {
19148
19513
  reasoning: true,
19149
19514
  input: ["text"],
19150
19515
  cost: {
19151
- input: 0.06,
19152
- output: 0.18,
19153
- cacheRead: 0.012,
19516
+ input: 0.020999999999999998,
19517
+ output: 0.063,
19518
+ cacheRead: 0.004200000000000001,
19154
19519
  cacheWrite: 0,
19155
19520
  },
19156
19521
  contextWindow: 256000,
@@ -19224,90 +19589,56 @@ export const MODELS = {
19224
19589
  contextWindow: 256000,
19225
19590
  maxTokens: 32000,
19226
19591
  },
19227
- "interfaze/interfaze-beta": {
19228
- id: "interfaze/interfaze-beta",
19229
- name: "Interfaze Beta",
19592
+ "inclusionai/ling-3.0-flash-vl": {
19593
+ id: "inclusionai/ling-3.0-flash-vl",
19594
+ name: "Ling 3.0 Flash VL",
19230
19595
  api: "anthropic-messages",
19231
19596
  provider: "vercel-ai-gateway",
19232
19597
  baseUrl: "https://ai-gateway.vercel.sh",
19233
19598
  reasoning: true,
19234
19599
  input: ["text", "image"],
19235
19600
  cost: {
19236
- input: 1.5,
19237
- output: 3.5,
19601
+ input: 0,
19602
+ output: 0,
19238
19603
  cacheRead: 0,
19239
19604
  cacheWrite: 0,
19240
19605
  },
19241
- contextWindow: 1000000,
19606
+ contextWindow: 256000,
19242
19607
  maxTokens: 32000,
19243
19608
  },
19244
- "kwaipilot/kat-coder-air-v2.5": {
19245
- id: "kwaipilot/kat-coder-air-v2.5",
19246
- name: "Kat Coder Air V2.5",
19609
+ "inclusionai/ling-3.0-flash-vl-free": {
19610
+ id: "inclusionai/ling-3.0-flash-vl-free",
19611
+ name: "Ling 3.0 Flash VL (Free)",
19247
19612
  api: "anthropic-messages",
19248
19613
  provider: "vercel-ai-gateway",
19249
19614
  baseUrl: "https://ai-gateway.vercel.sh",
19250
19615
  reasoning: true,
19251
19616
  input: ["text", "image"],
19252
19617
  cost: {
19253
- input: 0.15,
19254
- output: 0.6,
19255
- cacheRead: 0.03,
19256
- cacheWrite: 0,
19257
- },
19258
- contextWindow: 256000,
19259
- maxTokens: 80000,
19260
- },
19261
- "kwaipilot/kat-coder-pro-v1": {
19262
- id: "kwaipilot/kat-coder-pro-v1",
19263
- name: "KAT-Coder-Pro V1",
19264
- api: "anthropic-messages",
19265
- provider: "vercel-ai-gateway",
19266
- baseUrl: "https://ai-gateway.vercel.sh",
19267
- reasoning: false,
19268
- input: ["text"],
19269
- cost: {
19270
- input: 0.3,
19271
- output: 1.2,
19272
- cacheRead: 0.06,
19618
+ input: 0,
19619
+ output: 0,
19620
+ cacheRead: 0,
19273
19621
  cacheWrite: 0,
19274
19622
  },
19275
19623
  contextWindow: 256000,
19276
19624
  maxTokens: 32000,
19277
19625
  },
19278
- "kwaipilot/kat-coder-pro-v2": {
19279
- id: "kwaipilot/kat-coder-pro-v2",
19280
- name: "Kat Coder Pro V2",
19281
- api: "anthropic-messages",
19282
- provider: "vercel-ai-gateway",
19283
- baseUrl: "https://ai-gateway.vercel.sh",
19284
- reasoning: true,
19285
- input: ["text"],
19286
- cost: {
19287
- input: 0.3,
19288
- output: 1.2,
19289
- cacheRead: 0.06,
19290
- cacheWrite: 0,
19291
- },
19292
- contextWindow: 256000,
19293
- maxTokens: 256000,
19294
- },
19295
- "kwaipilot/kat-coder-pro-v2.5": {
19296
- id: "kwaipilot/kat-coder-pro-v2.5",
19297
- name: "Kat Coder Pro V2.5",
19626
+ "interfaze/interfaze-beta": {
19627
+ id: "interfaze/interfaze-beta",
19628
+ name: "Interfaze Beta",
19298
19629
  api: "anthropic-messages",
19299
19630
  provider: "vercel-ai-gateway",
19300
19631
  baseUrl: "https://ai-gateway.vercel.sh",
19301
19632
  reasoning: true,
19302
19633
  input: ["text", "image"],
19303
19634
  cost: {
19304
- input: 0.74,
19305
- output: 2.96,
19306
- cacheRead: 0.15,
19635
+ input: 1.5,
19636
+ output: 3.5,
19637
+ cacheRead: 0,
19307
19638
  cacheWrite: 0,
19308
19639
  },
19309
- contextWindow: 256000,
19310
- maxTokens: 80000,
19640
+ contextWindow: 1000000,
19641
+ maxTokens: 32000,
19311
19642
  },
19312
19643
  "meta/llama-3.1-70b": {
19313
19644
  id: "meta/llama-3.1-70b",
@@ -19643,46 +19974,12 @@ export const MODELS = {
19643
19974
  cost: {
19644
19975
  input: 0.3,
19645
19976
  output: 0.8999999999999999,
19646
- cacheRead: 0,
19977
+ cacheRead: 0.03,
19647
19978
  cacheWrite: 0,
19648
19979
  },
19649
19980
  contextWindow: 128000,
19650
19981
  maxTokens: 4000,
19651
19982
  },
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
19983
  "mistral/ministral-14b": {
19687
19984
  id: "mistral/ministral-14b",
19688
19985
  name: "Ministral 14B",
@@ -19694,10 +19991,10 @@ export const MODELS = {
19694
19991
  cost: {
19695
19992
  input: 0.19999999999999998,
19696
19993
  output: 0.19999999999999998,
19697
- cacheRead: 0,
19994
+ cacheRead: 0.02,
19698
19995
  cacheWrite: 0,
19699
19996
  },
19700
- contextWindow: 256000,
19997
+ contextWindow: 262144,
19701
19998
  maxTokens: 256000,
19702
19999
  },
19703
20000
  "mistral/ministral-3b": {
@@ -19711,10 +20008,10 @@ export const MODELS = {
19711
20008
  cost: {
19712
20009
  input: 0.09999999999999999,
19713
20010
  output: 0.09999999999999999,
19714
- cacheRead: 0,
20011
+ cacheRead: 0.01,
19715
20012
  cacheWrite: 0,
19716
20013
  },
19717
- contextWindow: 128000,
20014
+ contextWindow: 131072,
19718
20015
  maxTokens: 4000,
19719
20016
  },
19720
20017
  "mistral/ministral-8b": {
@@ -19728,10 +20025,10 @@ export const MODELS = {
19728
20025
  cost: {
19729
20026
  input: 0.15,
19730
20027
  output: 0.15,
19731
- cacheRead: 0,
20028
+ cacheRead: 0.015,
19732
20029
  cacheWrite: 0,
19733
20030
  },
19734
- contextWindow: 128000,
20031
+ contextWindow: 262144,
19735
20032
  maxTokens: 4000,
19736
20033
  },
19737
20034
  "mistral/mistral-large-3": {
@@ -19745,29 +20042,12 @@ export const MODELS = {
19745
20042
  cost: {
19746
20043
  input: 0.5,
19747
20044
  output: 1.5,
19748
- cacheRead: 0,
20045
+ cacheRead: 0.049999999999999996,
19749
20046
  cacheWrite: 0,
19750
20047
  },
19751
- contextWindow: 256000,
20048
+ contextWindow: 262144,
19752
20049
  maxTokens: 256000,
19753
20050
  },
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
20051
  "mistral/mistral-medium-3.5": {
19772
20052
  id: "mistral/mistral-medium-3.5",
19773
20053
  name: "Mistral Medium Latest",
@@ -19779,10 +20059,10 @@ export const MODELS = {
19779
20059
  cost: {
19780
20060
  input: 1.5,
19781
20061
  output: 7.5,
19782
- cacheRead: 0,
20062
+ cacheRead: 0.15,
19783
20063
  cacheWrite: 0,
19784
20064
  },
19785
- contextWindow: 256000,
20065
+ contextWindow: 262144,
19786
20066
  maxTokens: 256000,
19787
20067
  },
19788
20068
  "mistral/mistral-nemo": {
@@ -19792,15 +20072,15 @@ export const MODELS = {
19792
20072
  provider: "vercel-ai-gateway",
19793
20073
  baseUrl: "https://ai-gateway.vercel.sh",
19794
20074
  reasoning: false,
19795
- input: ["text", "image"],
20075
+ input: ["text"],
19796
20076
  cost: {
19797
- input: 0.15,
19798
- output: 0.15,
20077
+ input: 0.04,
20078
+ output: 0.16999999999999998,
19799
20079
  cacheRead: 0,
19800
20080
  cacheWrite: 0,
19801
20081
  },
19802
- contextWindow: 128000,
19803
- maxTokens: 128000,
20082
+ contextWindow: 60288,
20083
+ maxTokens: 16000,
19804
20084
  },
19805
20085
  "mistral/mistral-small": {
19806
20086
  id: "mistral/mistral-small",
@@ -19810,30 +20090,13 @@ export const MODELS = {
19810
20090
  baseUrl: "https://ai-gateway.vercel.sh",
19811
20091
  reasoning: false,
19812
20092
  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
20093
  cost: {
19831
20094
  input: 0.15,
19832
- output: 0.15,
19833
- cacheRead: 0,
20095
+ output: 0.6,
20096
+ cacheRead: 0.015,
19834
20097
  cacheWrite: 0,
19835
20098
  },
19836
- contextWindow: 128000,
20099
+ contextWindow: 262144,
19837
20100
  maxTokens: 4000,
19838
20101
  },
19839
20102
  "moonshotai/kimi-k2": {
@@ -19881,11 +20144,11 @@ export const MODELS = {
19881
20144
  cost: {
19882
20145
  input: 0.6,
19883
20146
  output: 3,
19884
- cacheRead: 0.09999999999999999,
20147
+ cacheRead: 0,
19885
20148
  cacheWrite: 0,
19886
20149
  },
19887
- contextWindow: 262114,
19888
- maxTokens: 262114,
20150
+ contextWindow: 256000,
20151
+ maxTokens: 256000,
19889
20152
  },
19890
20153
  "moonshotai/kimi-k2.6": {
19891
20154
  id: "moonshotai/kimi-k2.6",
@@ -19982,8 +20245,8 @@ export const MODELS = {
19982
20245
  input: ["text"],
19983
20246
  cost: {
19984
20247
  input: 0.049999999999999996,
19985
- output: 0.24,
19986
- cacheRead: 0,
20248
+ output: 0.19999999999999998,
20249
+ cacheRead: 0.024999999999999998,
19987
20250
  cacheWrite: 0,
19988
20251
  },
19989
20252
  contextWindow: 262144,
@@ -20897,8 +21160,8 @@ export const MODELS = {
20897
21160
  reasoning: true,
20898
21161
  input: ["text"],
20899
21162
  cost: {
20900
- input: 0.049999999999999996,
20901
- output: 0.19999999999999998,
21163
+ input: 0.03,
21164
+ output: 0.14,
20902
21165
  cacheRead: 0,
20903
21166
  cacheWrite: 0,
20904
21167
  },
@@ -21092,6 +21355,23 @@ export const MODELS = {
21092
21355
  contextWindow: 256000,
21093
21356
  maxTokens: 32768,
21094
21357
  },
21358
+ "sakana/fugu-max": {
21359
+ id: "sakana/fugu-max",
21360
+ name: "Fugu Max",
21361
+ api: "anthropic-messages",
21362
+ provider: "vercel-ai-gateway",
21363
+ baseUrl: "https://ai-gateway.vercel.sh",
21364
+ reasoning: true,
21365
+ input: ["text", "image"],
21366
+ cost: {
21367
+ input: 2,
21368
+ output: 6,
21369
+ cacheRead: 0.25,
21370
+ cacheWrite: 0,
21371
+ },
21372
+ contextWindow: 1000000,
21373
+ maxTokens: 1000000,
21374
+ },
21095
21375
  "sakana/fugu-ultra": {
21096
21376
  id: "sakana/fugu-ultra",
21097
21377
  name: "Fugu Ultra",
@@ -21109,6 +21389,23 @@ export const MODELS = {
21109
21389
  contextWindow: 1000000,
21110
21390
  maxTokens: 1000000,
21111
21391
  },
21392
+ "sakana/fugu-ultra-v2": {
21393
+ id: "sakana/fugu-ultra-v2",
21394
+ name: "Fugu Ultra v2",
21395
+ api: "anthropic-messages",
21396
+ provider: "vercel-ai-gateway",
21397
+ baseUrl: "https://ai-gateway.vercel.sh",
21398
+ reasoning: true,
21399
+ input: ["text", "image"],
21400
+ cost: {
21401
+ input: 5,
21402
+ output: 30,
21403
+ cacheRead: 0.5,
21404
+ cacheWrite: 0,
21405
+ },
21406
+ contextWindow: 1000000,
21407
+ maxTokens: 1000000,
21408
+ },
21112
21409
  "sakana/namazu": {
21113
21410
  id: "sakana/namazu",
21114
21411
  name: "Sakana Namazu",
@@ -21721,6 +22018,23 @@ export const MODELS = {
21721
22018
  contextWindow: 1000000,
21722
22019
  maxTokens: 131000,
21723
22020
  },
22021
+ "zai/glm-5.3-flashx": {
22022
+ id: "zai/glm-5.3-flashx",
22023
+ name: "GLM 5.3 FlashX",
22024
+ api: "anthropic-messages",
22025
+ provider: "vercel-ai-gateway",
22026
+ baseUrl: "https://ai-gateway.vercel.sh",
22027
+ reasoning: true,
22028
+ input: ["text", "image"],
22029
+ cost: {
22030
+ input: 0.37,
22031
+ output: 1.25,
22032
+ cacheRead: 0.075,
22033
+ cacheWrite: 0,
22034
+ },
22035
+ contextWindow: 1000000,
22036
+ maxTokens: 131072,
22037
+ },
21724
22038
  "zai/glm-5v-turbo": {
21725
22039
  id: "zai/glm-5v-turbo",
21726
22040
  name: "GLM 5V Turbo",