@elyracode/ai 0.9.5 → 0.9.7

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.
@@ -8,7 +8,7 @@ export const MODELS = {
8
8
  api: "bedrock-converse-stream",
9
9
  provider: "amazon-bedrock",
10
10
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
11
- reasoning: false,
11
+ reasoning: true,
12
12
  input: ["text", "image"],
13
13
  cost: {
14
14
  input: 0.33,
@@ -77,6 +77,7 @@ export const MODELS = {
77
77
  provider: "amazon-bedrock",
78
78
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
79
79
  reasoning: true,
80
+ thinkingType: "budget",
80
81
  input: ["text", "image"],
81
82
  cost: {
82
83
  input: 1,
@@ -94,6 +95,7 @@ export const MODELS = {
94
95
  provider: "amazon-bedrock",
95
96
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
96
97
  reasoning: true,
98
+ thinkingType: "budget",
97
99
  input: ["text", "image"],
98
100
  cost: {
99
101
  input: 15,
@@ -111,6 +113,7 @@ export const MODELS = {
111
113
  provider: "amazon-bedrock",
112
114
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
113
115
  reasoning: true,
116
+ thinkingType: "budget",
114
117
  input: ["text", "image"],
115
118
  cost: {
116
119
  input: 5,
@@ -128,6 +131,7 @@ export const MODELS = {
128
131
  provider: "amazon-bedrock",
129
132
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
130
133
  reasoning: true,
134
+ thinkingType: "adaptive",
131
135
  thinkingLevelMap: { "xhigh": "max" },
132
136
  input: ["text", "image"],
133
137
  cost: {
@@ -146,6 +150,7 @@ export const MODELS = {
146
150
  provider: "amazon-bedrock",
147
151
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
148
152
  reasoning: true,
153
+ thinkingType: "adaptive",
149
154
  thinkingLevelMap: { "xhigh": "xhigh" },
150
155
  input: ["text", "image"],
151
156
  cost: {
@@ -164,6 +169,7 @@ export const MODELS = {
164
169
  provider: "amazon-bedrock",
165
170
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
166
171
  reasoning: true,
172
+ thinkingType: "adaptive",
167
173
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
168
174
  input: ["text", "image"],
169
175
  cost: {
@@ -182,6 +188,7 @@ export const MODELS = {
182
188
  provider: "amazon-bedrock",
183
189
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
184
190
  reasoning: true,
191
+ thinkingType: "budget",
185
192
  input: ["text", "image"],
186
193
  cost: {
187
194
  input: 3,
@@ -199,6 +206,7 @@ export const MODELS = {
199
206
  provider: "amazon-bedrock",
200
207
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
201
208
  reasoning: true,
209
+ thinkingType: "adaptive",
202
210
  input: ["text", "image"],
203
211
  cost: {
204
212
  input: 3,
@@ -216,6 +224,7 @@ export const MODELS = {
216
224
  provider: "amazon-bedrock",
217
225
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
218
226
  reasoning: true,
227
+ thinkingType: "budget",
219
228
  input: ["text", "image"],
220
229
  cost: {
221
230
  input: 1,
@@ -233,6 +242,7 @@ export const MODELS = {
233
242
  provider: "amazon-bedrock",
234
243
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
235
244
  reasoning: true,
245
+ thinkingType: "adaptive",
236
246
  thinkingLevelMap: { "xhigh": "max" },
237
247
  input: ["text", "image"],
238
248
  cost: {
@@ -251,6 +261,7 @@ export const MODELS = {
251
261
  provider: "amazon-bedrock",
252
262
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
253
263
  reasoning: true,
264
+ thinkingType: "adaptive",
254
265
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
255
266
  input: ["text", "image"],
256
267
  cost: {
@@ -269,6 +280,7 @@ export const MODELS = {
269
280
  provider: "amazon-bedrock",
270
281
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
271
282
  reasoning: true,
283
+ thinkingType: "budget",
272
284
  input: ["text", "image"],
273
285
  cost: {
274
286
  input: 3,
@@ -286,6 +298,7 @@ export const MODELS = {
286
298
  provider: "amazon-bedrock",
287
299
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
288
300
  reasoning: true,
301
+ thinkingType: "adaptive",
289
302
  input: ["text", "image"],
290
303
  cost: {
291
304
  input: 3.3,
@@ -347,6 +360,25 @@ export const MODELS = {
347
360
  contextWindow: 163840,
348
361
  maxTokens: 81920,
349
362
  },
363
+ "eu.anthropic.claude-fable-5": {
364
+ id: "eu.anthropic.claude-fable-5",
365
+ name: "Claude Fable 5 (EU)",
366
+ api: "bedrock-converse-stream",
367
+ provider: "amazon-bedrock",
368
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
369
+ reasoning: true,
370
+ thinkingType: "adaptive",
371
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
372
+ input: ["text", "image"],
373
+ cost: {
374
+ input: 11,
375
+ output: 55,
376
+ cacheRead: 1.1,
377
+ cacheWrite: 13.75,
378
+ },
379
+ contextWindow: 1000000,
380
+ maxTokens: 128000,
381
+ },
350
382
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
351
383
  id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
352
384
  name: "Claude Haiku 4.5 (EU)",
@@ -354,6 +386,7 @@ export const MODELS = {
354
386
  provider: "amazon-bedrock",
355
387
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
356
388
  reasoning: true,
389
+ thinkingType: "budget",
357
390
  input: ["text", "image"],
358
391
  cost: {
359
392
  input: 1,
@@ -371,6 +404,7 @@ export const MODELS = {
371
404
  provider: "amazon-bedrock",
372
405
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
373
406
  reasoning: true,
407
+ thinkingType: "budget",
374
408
  input: ["text", "image"],
375
409
  cost: {
376
410
  input: 5,
@@ -388,6 +422,7 @@ export const MODELS = {
388
422
  provider: "amazon-bedrock",
389
423
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
390
424
  reasoning: true,
425
+ thinkingType: "adaptive",
391
426
  thinkingLevelMap: { "xhigh": "max" },
392
427
  input: ["text", "image"],
393
428
  cost: {
@@ -406,6 +441,7 @@ export const MODELS = {
406
441
  provider: "amazon-bedrock",
407
442
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
408
443
  reasoning: true,
444
+ thinkingType: "adaptive",
409
445
  thinkingLevelMap: { "xhigh": "xhigh" },
410
446
  input: ["text", "image"],
411
447
  cost: {
@@ -424,6 +460,7 @@ export const MODELS = {
424
460
  provider: "amazon-bedrock",
425
461
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
426
462
  reasoning: true,
463
+ thinkingType: "adaptive",
427
464
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
428
465
  input: ["text", "image"],
429
466
  cost: {
@@ -442,6 +479,7 @@ export const MODELS = {
442
479
  provider: "amazon-bedrock",
443
480
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
444
481
  reasoning: true,
482
+ thinkingType: "budget",
445
483
  input: ["text", "image"],
446
484
  cost: {
447
485
  input: 3.3,
@@ -459,6 +497,7 @@ export const MODELS = {
459
497
  provider: "amazon-bedrock",
460
498
  baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
461
499
  reasoning: true,
500
+ thinkingType: "adaptive",
462
501
  input: ["text", "image"],
463
502
  cost: {
464
503
  input: 3.3,
@@ -469,6 +508,25 @@ export const MODELS = {
469
508
  contextWindow: 1000000,
470
509
  maxTokens: 64000,
471
510
  },
511
+ "global.anthropic.claude-fable-5": {
512
+ id: "global.anthropic.claude-fable-5",
513
+ name: "Claude Fable 5 (Global)",
514
+ api: "bedrock-converse-stream",
515
+ provider: "amazon-bedrock",
516
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
517
+ reasoning: true,
518
+ thinkingType: "adaptive",
519
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
520
+ input: ["text", "image"],
521
+ cost: {
522
+ input: 10,
523
+ output: 50,
524
+ cacheRead: 1,
525
+ cacheWrite: 12.5,
526
+ },
527
+ contextWindow: 1000000,
528
+ maxTokens: 128000,
529
+ },
472
530
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
473
531
  id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
474
532
  name: "Claude Haiku 4.5 (Global)",
@@ -476,6 +534,7 @@ export const MODELS = {
476
534
  provider: "amazon-bedrock",
477
535
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
478
536
  reasoning: true,
537
+ thinkingType: "budget",
479
538
  input: ["text", "image"],
480
539
  cost: {
481
540
  input: 1,
@@ -493,6 +552,7 @@ export const MODELS = {
493
552
  provider: "amazon-bedrock",
494
553
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
495
554
  reasoning: true,
555
+ thinkingType: "budget",
496
556
  input: ["text", "image"],
497
557
  cost: {
498
558
  input: 5,
@@ -510,6 +570,7 @@ export const MODELS = {
510
570
  provider: "amazon-bedrock",
511
571
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
512
572
  reasoning: true,
573
+ thinkingType: "adaptive",
513
574
  thinkingLevelMap: { "xhigh": "max" },
514
575
  input: ["text", "image"],
515
576
  cost: {
@@ -528,6 +589,7 @@ export const MODELS = {
528
589
  provider: "amazon-bedrock",
529
590
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
530
591
  reasoning: true,
592
+ thinkingType: "adaptive",
531
593
  thinkingLevelMap: { "xhigh": "xhigh" },
532
594
  input: ["text", "image"],
533
595
  cost: {
@@ -546,6 +608,7 @@ export const MODELS = {
546
608
  provider: "amazon-bedrock",
547
609
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
548
610
  reasoning: true,
611
+ thinkingType: "adaptive",
549
612
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
550
613
  input: ["text", "image"],
551
614
  cost: {
@@ -564,6 +627,7 @@ export const MODELS = {
564
627
  provider: "amazon-bedrock",
565
628
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
566
629
  reasoning: true,
630
+ thinkingType: "budget",
567
631
  input: ["text", "image"],
568
632
  cost: {
569
633
  input: 3,
@@ -581,6 +645,7 @@ export const MODELS = {
581
645
  provider: "amazon-bedrock",
582
646
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
583
647
  reasoning: true,
648
+ thinkingType: "adaptive",
584
649
  input: ["text", "image"],
585
650
  cost: {
586
651
  input: 3,
@@ -632,6 +697,7 @@ export const MODELS = {
632
697
  provider: "amazon-bedrock",
633
698
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
634
699
  reasoning: true,
700
+ thinkingType: "adaptive",
635
701
  thinkingLevelMap: { "xhigh": "xhigh" },
636
702
  input: ["text", "image"],
637
703
  cost: {
@@ -650,6 +716,7 @@ export const MODELS = {
650
716
  provider: "amazon-bedrock",
651
717
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
652
718
  reasoning: true,
719
+ thinkingType: "adaptive",
653
720
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
654
721
  input: ["text", "image"],
655
722
  cost: {
@@ -668,6 +735,7 @@ export const MODELS = {
668
735
  provider: "amazon-bedrock",
669
736
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
670
737
  reasoning: true,
738
+ thinkingType: "budget",
671
739
  input: ["text", "image"],
672
740
  cost: {
673
741
  input: 3,
@@ -685,6 +753,7 @@ export const MODELS = {
685
753
  provider: "amazon-bedrock",
686
754
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
687
755
  reasoning: true,
756
+ thinkingType: "adaptive",
688
757
  input: ["text", "image"],
689
758
  cost: {
690
759
  input: 3,
@@ -1128,7 +1197,7 @@ export const MODELS = {
1128
1197
  api: "bedrock-converse-stream",
1129
1198
  provider: "amazon-bedrock",
1130
1199
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1131
- reasoning: false,
1200
+ reasoning: true,
1132
1201
  input: ["text"],
1133
1202
  cost: {
1134
1203
  input: 0.15,
@@ -1145,7 +1214,7 @@ export const MODELS = {
1145
1214
  api: "bedrock-converse-stream",
1146
1215
  provider: "amazon-bedrock",
1147
1216
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1148
- reasoning: false,
1217
+ reasoning: true,
1149
1218
  input: ["text"],
1150
1219
  cost: {
1151
1220
  input: 0.15,
@@ -1162,7 +1231,7 @@ export const MODELS = {
1162
1231
  api: "bedrock-converse-stream",
1163
1232
  provider: "amazon-bedrock",
1164
1233
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1165
- reasoning: false,
1234
+ reasoning: true,
1166
1235
  input: ["text"],
1167
1236
  cost: {
1168
1237
  input: 0.07,
@@ -1179,7 +1248,7 @@ export const MODELS = {
1179
1248
  api: "bedrock-converse-stream",
1180
1249
  provider: "amazon-bedrock",
1181
1250
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1182
- reasoning: false,
1251
+ reasoning: true,
1183
1252
  input: ["text"],
1184
1253
  cost: {
1185
1254
  input: 0.07,
@@ -1343,6 +1412,25 @@ export const MODELS = {
1343
1412
  contextWindow: 262000,
1344
1413
  maxTokens: 262000,
1345
1414
  },
1415
+ "us.anthropic.claude-fable-5": {
1416
+ id: "us.anthropic.claude-fable-5",
1417
+ name: "Claude Fable 5 (US)",
1418
+ api: "bedrock-converse-stream",
1419
+ provider: "amazon-bedrock",
1420
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1421
+ reasoning: true,
1422
+ thinkingType: "adaptive",
1423
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
1424
+ input: ["text", "image"],
1425
+ cost: {
1426
+ input: 10,
1427
+ output: 50,
1428
+ cacheRead: 1,
1429
+ cacheWrite: 12.5,
1430
+ },
1431
+ contextWindow: 1000000,
1432
+ maxTokens: 128000,
1433
+ },
1346
1434
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
1347
1435
  id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
1348
1436
  name: "Claude Haiku 4.5 (US)",
@@ -1350,6 +1438,7 @@ export const MODELS = {
1350
1438
  provider: "amazon-bedrock",
1351
1439
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1352
1440
  reasoning: true,
1441
+ thinkingType: "budget",
1353
1442
  input: ["text", "image"],
1354
1443
  cost: {
1355
1444
  input: 1,
@@ -1367,6 +1456,7 @@ export const MODELS = {
1367
1456
  provider: "amazon-bedrock",
1368
1457
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1369
1458
  reasoning: true,
1459
+ thinkingType: "budget",
1370
1460
  input: ["text", "image"],
1371
1461
  cost: {
1372
1462
  input: 15,
@@ -1384,6 +1474,7 @@ export const MODELS = {
1384
1474
  provider: "amazon-bedrock",
1385
1475
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1386
1476
  reasoning: true,
1477
+ thinkingType: "budget",
1387
1478
  input: ["text", "image"],
1388
1479
  cost: {
1389
1480
  input: 5,
@@ -1401,6 +1492,7 @@ export const MODELS = {
1401
1492
  provider: "amazon-bedrock",
1402
1493
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1403
1494
  reasoning: true,
1495
+ thinkingType: "adaptive",
1404
1496
  thinkingLevelMap: { "xhigh": "max" },
1405
1497
  input: ["text", "image"],
1406
1498
  cost: {
@@ -1419,6 +1511,7 @@ export const MODELS = {
1419
1511
  provider: "amazon-bedrock",
1420
1512
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1421
1513
  reasoning: true,
1514
+ thinkingType: "adaptive",
1422
1515
  thinkingLevelMap: { "xhigh": "xhigh" },
1423
1516
  input: ["text", "image"],
1424
1517
  cost: {
@@ -1437,6 +1530,7 @@ export const MODELS = {
1437
1530
  provider: "amazon-bedrock",
1438
1531
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1439
1532
  reasoning: true,
1533
+ thinkingType: "adaptive",
1440
1534
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
1441
1535
  input: ["text", "image"],
1442
1536
  cost: {
@@ -1455,6 +1549,7 @@ export const MODELS = {
1455
1549
  provider: "amazon-bedrock",
1456
1550
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1457
1551
  reasoning: true,
1552
+ thinkingType: "budget",
1458
1553
  input: ["text", "image"],
1459
1554
  cost: {
1460
1555
  input: 3,
@@ -1472,6 +1567,7 @@ export const MODELS = {
1472
1567
  provider: "amazon-bedrock",
1473
1568
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1474
1569
  reasoning: true,
1570
+ thinkingType: "adaptive",
1475
1571
  input: ["text", "image"],
1476
1572
  cost: {
1477
1573
  input: 3,
@@ -1661,6 +1757,7 @@ export const MODELS = {
1661
1757
  provider: "anthropic",
1662
1758
  baseUrl: "https://api.anthropic.com",
1663
1759
  reasoning: true,
1760
+ thinkingType: "budget",
1664
1761
  input: ["text", "image"],
1665
1762
  cost: {
1666
1763
  input: 3,
@@ -1722,6 +1819,25 @@ export const MODELS = {
1722
1819
  contextWindow: 200000,
1723
1820
  maxTokens: 4096,
1724
1821
  },
1822
+ "claude-fable-5": {
1823
+ id: "claude-fable-5",
1824
+ name: "Claude Fable 5",
1825
+ api: "anthropic-messages",
1826
+ provider: "anthropic",
1827
+ baseUrl: "https://api.anthropic.com",
1828
+ reasoning: true,
1829
+ thinkingType: "adaptive",
1830
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
1831
+ input: ["text", "image"],
1832
+ cost: {
1833
+ input: 10,
1834
+ output: 50,
1835
+ cacheRead: 1,
1836
+ cacheWrite: 12.5,
1837
+ },
1838
+ contextWindow: 1000000,
1839
+ maxTokens: 128000,
1840
+ },
1725
1841
  "claude-haiku-4-5": {
1726
1842
  id: "claude-haiku-4-5",
1727
1843
  name: "Claude Haiku 4.5 (latest)",
@@ -1729,6 +1845,7 @@ export const MODELS = {
1729
1845
  provider: "anthropic",
1730
1846
  baseUrl: "https://api.anthropic.com",
1731
1847
  reasoning: true,
1848
+ thinkingType: "budget",
1732
1849
  input: ["text", "image"],
1733
1850
  cost: {
1734
1851
  input: 1,
@@ -1746,6 +1863,7 @@ export const MODELS = {
1746
1863
  provider: "anthropic",
1747
1864
  baseUrl: "https://api.anthropic.com",
1748
1865
  reasoning: true,
1866
+ thinkingType: "budget",
1749
1867
  input: ["text", "image"],
1750
1868
  cost: {
1751
1869
  input: 1,
@@ -1763,6 +1881,7 @@ export const MODELS = {
1763
1881
  provider: "anthropic",
1764
1882
  baseUrl: "https://api.anthropic.com",
1765
1883
  reasoning: true,
1884
+ thinkingType: "budget",
1766
1885
  input: ["text", "image"],
1767
1886
  cost: {
1768
1887
  input: 15,
@@ -1780,6 +1899,7 @@ export const MODELS = {
1780
1899
  provider: "anthropic",
1781
1900
  baseUrl: "https://api.anthropic.com",
1782
1901
  reasoning: true,
1902
+ thinkingType: "budget",
1783
1903
  input: ["text", "image"],
1784
1904
  cost: {
1785
1905
  input: 15,
@@ -1797,6 +1917,7 @@ export const MODELS = {
1797
1917
  provider: "anthropic",
1798
1918
  baseUrl: "https://api.anthropic.com",
1799
1919
  reasoning: true,
1920
+ thinkingType: "budget",
1800
1921
  input: ["text", "image"],
1801
1922
  cost: {
1802
1923
  input: 15,
@@ -1814,6 +1935,7 @@ export const MODELS = {
1814
1935
  provider: "anthropic",
1815
1936
  baseUrl: "https://api.anthropic.com",
1816
1937
  reasoning: true,
1938
+ thinkingType: "budget",
1817
1939
  input: ["text", "image"],
1818
1940
  cost: {
1819
1941
  input: 15,
@@ -1831,6 +1953,7 @@ export const MODELS = {
1831
1953
  provider: "anthropic",
1832
1954
  baseUrl: "https://api.anthropic.com",
1833
1955
  reasoning: true,
1956
+ thinkingType: "budget",
1834
1957
  input: ["text", "image"],
1835
1958
  cost: {
1836
1959
  input: 5,
@@ -1848,6 +1971,7 @@ export const MODELS = {
1848
1971
  provider: "anthropic",
1849
1972
  baseUrl: "https://api.anthropic.com",
1850
1973
  reasoning: true,
1974
+ thinkingType: "budget",
1851
1975
  input: ["text", "image"],
1852
1976
  cost: {
1853
1977
  input: 5,
@@ -1865,6 +1989,7 @@ export const MODELS = {
1865
1989
  provider: "anthropic",
1866
1990
  baseUrl: "https://api.anthropic.com",
1867
1991
  reasoning: true,
1992
+ thinkingType: "adaptive",
1868
1993
  thinkingLevelMap: { "xhigh": "max" },
1869
1994
  input: ["text", "image"],
1870
1995
  cost: {
@@ -1883,6 +2008,7 @@ export const MODELS = {
1883
2008
  provider: "anthropic",
1884
2009
  baseUrl: "https://api.anthropic.com",
1885
2010
  reasoning: true,
2011
+ thinkingType: "adaptive",
1886
2012
  thinkingLevelMap: { "xhigh": "xhigh" },
1887
2013
  input: ["text", "image"],
1888
2014
  cost: {
@@ -1901,6 +2027,7 @@ export const MODELS = {
1901
2027
  provider: "anthropic",
1902
2028
  baseUrl: "https://api.anthropic.com",
1903
2029
  reasoning: true,
2030
+ thinkingType: "adaptive",
1904
2031
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
1905
2032
  input: ["text", "image"],
1906
2033
  cost: {
@@ -1919,6 +2046,7 @@ export const MODELS = {
1919
2046
  provider: "anthropic",
1920
2047
  baseUrl: "https://api.anthropic.com",
1921
2048
  reasoning: true,
2049
+ thinkingType: "budget",
1922
2050
  input: ["text", "image"],
1923
2051
  cost: {
1924
2052
  input: 3,
@@ -1936,6 +2064,7 @@ export const MODELS = {
1936
2064
  provider: "anthropic",
1937
2065
  baseUrl: "https://api.anthropic.com",
1938
2066
  reasoning: true,
2067
+ thinkingType: "budget",
1939
2068
  input: ["text", "image"],
1940
2069
  cost: {
1941
2070
  input: 3,
@@ -1953,6 +2082,7 @@ export const MODELS = {
1953
2082
  provider: "anthropic",
1954
2083
  baseUrl: "https://api.anthropic.com",
1955
2084
  reasoning: true,
2085
+ thinkingType: "budget",
1956
2086
  input: ["text", "image"],
1957
2087
  cost: {
1958
2088
  input: 3,
@@ -1970,6 +2100,7 @@ export const MODELS = {
1970
2100
  provider: "anthropic",
1971
2101
  baseUrl: "https://api.anthropic.com",
1972
2102
  reasoning: true,
2103
+ thinkingType: "budget",
1973
2104
  input: ["text", "image"],
1974
2105
  cost: {
1975
2106
  input: 3,
@@ -1987,6 +2118,7 @@ export const MODELS = {
1987
2118
  provider: "anthropic",
1988
2119
  baseUrl: "https://api.anthropic.com",
1989
2120
  reasoning: true,
2121
+ thinkingType: "adaptive",
1990
2122
  input: ["text", "image"],
1991
2123
  cost: {
1992
2124
  input: 3,
@@ -2792,6 +2924,24 @@ export const MODELS = {
2792
2924
  contextWindow: 200000,
2793
2925
  maxTokens: 8192,
2794
2926
  },
2927
+ "claude-fable-5": {
2928
+ id: "claude-fable-5",
2929
+ name: "Claude Fable 5",
2930
+ api: "anthropic-messages",
2931
+ provider: "cloudflare-ai-gateway",
2932
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
2933
+ reasoning: true,
2934
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
2935
+ input: ["text", "image"],
2936
+ cost: {
2937
+ input: 10,
2938
+ output: 50,
2939
+ cacheRead: 1,
2940
+ cacheWrite: 12.5,
2941
+ },
2942
+ contextWindow: 1000000,
2943
+ maxTokens: 128000,
2944
+ },
2795
2945
  "claude-haiku-4-5": {
2796
2946
  id: "claude-haiku-4-5",
2797
2947
  name: "Claude Haiku 4.5 (latest)",
@@ -4692,77 +4842,9 @@ export const MODELS = {
4692
4842
  },
4693
4843
  },
4694
4844
  "groq": {
4695
- "deepseek-r1-distill-llama-70b": {
4696
- id: "deepseek-r1-distill-llama-70b",
4697
- name: "DeepSeek R1 Distill Llama 70B",
4698
- api: "openai-completions",
4699
- provider: "groq",
4700
- baseUrl: "https://api.groq.com/openai/v1",
4701
- reasoning: true,
4702
- input: ["text"],
4703
- cost: {
4704
- input: 0.75,
4705
- output: 0.99,
4706
- cacheRead: 0,
4707
- cacheWrite: 0,
4708
- },
4709
- contextWindow: 131072,
4710
- maxTokens: 8192,
4711
- },
4712
- "gemma2-9b-it": {
4713
- id: "gemma2-9b-it",
4714
- name: "Gemma 2 9B",
4715
- api: "openai-completions",
4716
- provider: "groq",
4717
- baseUrl: "https://api.groq.com/openai/v1",
4718
- reasoning: false,
4719
- input: ["text"],
4720
- cost: {
4721
- input: 0.2,
4722
- output: 0.2,
4723
- cacheRead: 0,
4724
- cacheWrite: 0,
4725
- },
4726
- contextWindow: 8192,
4727
- maxTokens: 8192,
4728
- },
4729
- "groq/compound": {
4730
- id: "groq/compound",
4731
- name: "Compound",
4732
- api: "openai-completions",
4733
- provider: "groq",
4734
- baseUrl: "https://api.groq.com/openai/v1",
4735
- reasoning: true,
4736
- input: ["text"],
4737
- cost: {
4738
- input: 0,
4739
- output: 0,
4740
- cacheRead: 0,
4741
- cacheWrite: 0,
4742
- },
4743
- contextWindow: 131072,
4744
- maxTokens: 8192,
4745
- },
4746
- "groq/compound-mini": {
4747
- id: "groq/compound-mini",
4748
- name: "Compound Mini",
4749
- api: "openai-completions",
4750
- provider: "groq",
4751
- baseUrl: "https://api.groq.com/openai/v1",
4752
- reasoning: true,
4753
- input: ["text"],
4754
- cost: {
4755
- input: 0,
4756
- output: 0,
4757
- cacheRead: 0,
4758
- cacheWrite: 0,
4759
- },
4760
- contextWindow: 131072,
4761
- maxTokens: 8192,
4762
- },
4763
4845
  "llama-3.1-8b-instant": {
4764
4846
  id: "llama-3.1-8b-instant",
4765
- name: "Llama 3.1 8B Instant",
4847
+ name: "Llama 3.1 8B",
4766
4848
  api: "openai-completions",
4767
4849
  provider: "groq",
4768
4850
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4779,7 +4861,7 @@ export const MODELS = {
4779
4861
  },
4780
4862
  "llama-3.3-70b-versatile": {
4781
4863
  id: "llama-3.3-70b-versatile",
4782
- name: "Llama 3.3 70B Versatile",
4864
+ name: "Llama 3.3 70B",
4783
4865
  api: "openai-completions",
4784
4866
  provider: "groq",
4785
4867
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4794,60 +4876,9 @@ export const MODELS = {
4794
4876
  contextWindow: 131072,
4795
4877
  maxTokens: 32768,
4796
4878
  },
4797
- "llama3-70b-8192": {
4798
- id: "llama3-70b-8192",
4799
- name: "Llama 3 70B",
4800
- api: "openai-completions",
4801
- provider: "groq",
4802
- baseUrl: "https://api.groq.com/openai/v1",
4803
- reasoning: false,
4804
- input: ["text"],
4805
- cost: {
4806
- input: 0.59,
4807
- output: 0.79,
4808
- cacheRead: 0,
4809
- cacheWrite: 0,
4810
- },
4811
- contextWindow: 8192,
4812
- maxTokens: 8192,
4813
- },
4814
- "llama3-8b-8192": {
4815
- id: "llama3-8b-8192",
4816
- name: "Llama 3 8B",
4817
- api: "openai-completions",
4818
- provider: "groq",
4819
- baseUrl: "https://api.groq.com/openai/v1",
4820
- reasoning: false,
4821
- input: ["text"],
4822
- cost: {
4823
- input: 0.05,
4824
- output: 0.08,
4825
- cacheRead: 0,
4826
- cacheWrite: 0,
4827
- },
4828
- contextWindow: 8192,
4829
- maxTokens: 8192,
4830
- },
4831
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
4832
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
4833
- name: "Llama 4 Maverick 17B",
4834
- api: "openai-completions",
4835
- provider: "groq",
4836
- baseUrl: "https://api.groq.com/openai/v1",
4837
- reasoning: false,
4838
- input: ["text", "image"],
4839
- cost: {
4840
- input: 0.2,
4841
- output: 0.6,
4842
- cacheRead: 0,
4843
- cacheWrite: 0,
4844
- },
4845
- contextWindow: 131072,
4846
- maxTokens: 8192,
4847
- },
4848
4879
  "meta-llama/llama-4-scout-17b-16e-instruct": {
4849
4880
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
4850
- name: "Llama 4 Scout 17B",
4881
+ name: "Llama 4 Scout 17B 16E",
4851
4882
  api: "openai-completions",
4852
4883
  provider: "groq",
4853
4884
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4862,57 +4893,6 @@ export const MODELS = {
4862
4893
  contextWindow: 131072,
4863
4894
  maxTokens: 8192,
4864
4895
  },
4865
- "mistral-saba-24b": {
4866
- id: "mistral-saba-24b",
4867
- name: "Mistral Saba 24B",
4868
- api: "openai-completions",
4869
- provider: "groq",
4870
- baseUrl: "https://api.groq.com/openai/v1",
4871
- reasoning: false,
4872
- input: ["text"],
4873
- cost: {
4874
- input: 0.79,
4875
- output: 0.79,
4876
- cacheRead: 0,
4877
- cacheWrite: 0,
4878
- },
4879
- contextWindow: 32768,
4880
- maxTokens: 32768,
4881
- },
4882
- "moonshotai/kimi-k2-instruct": {
4883
- id: "moonshotai/kimi-k2-instruct",
4884
- name: "Kimi K2 Instruct",
4885
- api: "openai-completions",
4886
- provider: "groq",
4887
- baseUrl: "https://api.groq.com/openai/v1",
4888
- reasoning: false,
4889
- input: ["text"],
4890
- cost: {
4891
- input: 1,
4892
- output: 3,
4893
- cacheRead: 0,
4894
- cacheWrite: 0,
4895
- },
4896
- contextWindow: 131072,
4897
- maxTokens: 16384,
4898
- },
4899
- "moonshotai/kimi-k2-instruct-0905": {
4900
- id: "moonshotai/kimi-k2-instruct-0905",
4901
- name: "Kimi K2 Instruct 0905",
4902
- api: "openai-completions",
4903
- provider: "groq",
4904
- baseUrl: "https://api.groq.com/openai/v1",
4905
- reasoning: false,
4906
- input: ["text"],
4907
- cost: {
4908
- input: 1,
4909
- output: 3,
4910
- cacheRead: 0.5,
4911
- cacheWrite: 0,
4912
- },
4913
- contextWindow: 262144,
4914
- maxTokens: 16384,
4915
- },
4916
4896
  "openai/gpt-oss-120b": {
4917
4897
  id: "openai/gpt-oss-120b",
4918
4898
  name: "GPT OSS 120B",
@@ -4964,26 +4944,9 @@ export const MODELS = {
4964
4944
  contextWindow: 131072,
4965
4945
  maxTokens: 65536,
4966
4946
  },
4967
- "qwen-qwq-32b": {
4968
- id: "qwen-qwq-32b",
4969
- name: "Qwen QwQ 32B",
4970
- api: "openai-completions",
4971
- provider: "groq",
4972
- baseUrl: "https://api.groq.com/openai/v1",
4973
- reasoning: true,
4974
- input: ["text"],
4975
- cost: {
4976
- input: 0.29,
4977
- output: 0.39,
4978
- cacheRead: 0,
4979
- cacheWrite: 0,
4980
- },
4981
- contextWindow: 131072,
4982
- maxTokens: 16384,
4983
- },
4984
4947
  "qwen/qwen3-32b": {
4985
4948
  id: "qwen/qwen3-32b",
4986
- name: "Qwen3 32B",
4949
+ name: "Qwen3-32B",
4987
4950
  api: "openai-completions",
4988
4951
  provider: "groq",
4989
4952
  baseUrl: "https://api.groq.com/openai/v1",
@@ -7162,6 +7125,24 @@ export const MODELS = {
7162
7125
  contextWindow: 200000,
7163
7126
  maxTokens: 32000,
7164
7127
  },
7128
+ "claude-fable-5": {
7129
+ id: "claude-fable-5",
7130
+ name: "Claude Fable 5",
7131
+ api: "anthropic-messages",
7132
+ provider: "opencode",
7133
+ baseUrl: "https://opencode.ai/zen",
7134
+ reasoning: true,
7135
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
7136
+ input: ["text", "image"],
7137
+ cost: {
7138
+ input: 10,
7139
+ output: 50,
7140
+ cacheRead: 1,
7141
+ cacheWrite: 12.5,
7142
+ },
7143
+ contextWindow: 1000000,
7144
+ maxTokens: 128000,
7145
+ },
7165
7146
  "claude-haiku-4-5": {
7166
7147
  id: "claude-haiku-4-5",
7167
7148
  name: "Claude Haiku 4.5",
@@ -7331,7 +7312,7 @@ export const MODELS = {
7331
7312
  cost: {
7332
7313
  input: 0.14,
7333
7314
  output: 0.28,
7334
- cacheRead: 0.03,
7315
+ cacheRead: 0.028,
7335
7316
  cacheWrite: 0,
7336
7317
  },
7337
7318
  contextWindow: 1000000,
@@ -7356,6 +7337,25 @@ export const MODELS = {
7356
7337
  contextWindow: 200000,
7357
7338
  maxTokens: 128000,
7358
7339
  },
7340
+ "deepseek-v4-pro": {
7341
+ id: "deepseek-v4-pro",
7342
+ name: "DeepSeek V4 Pro",
7343
+ api: "openai-completions",
7344
+ provider: "opencode",
7345
+ baseUrl: "https://opencode.ai/zen/v1",
7346
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7347
+ reasoning: true,
7348
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7349
+ input: ["text"],
7350
+ cost: {
7351
+ input: 1.74,
7352
+ output: 3.84,
7353
+ cacheRead: 0.145,
7354
+ cacheWrite: 0,
7355
+ },
7356
+ contextWindow: 1000000,
7357
+ maxTokens: 384000,
7358
+ },
7359
7359
  "gemini-3-flash": {
7360
7360
  id: "gemini-3-flash",
7361
7361
  name: "Gemini 3 Flash",
@@ -7834,26 +7834,26 @@ export const MODELS = {
7834
7834
  contextWindow: 204800,
7835
7835
  maxTokens: 131072,
7836
7836
  },
7837
- "minimax-m3-free": {
7838
- id: "minimax-m3-free",
7839
- name: "MiniMax M3 Free",
7840
- api: "anthropic-messages",
7837
+ "nemotron-3-ultra-free": {
7838
+ id: "nemotron-3-ultra-free",
7839
+ name: "Nemotron 3 Ultra Free",
7840
+ api: "openai-completions",
7841
7841
  provider: "opencode",
7842
- baseUrl: "https://opencode.ai/zen",
7842
+ baseUrl: "https://opencode.ai/zen/v1",
7843
7843
  reasoning: true,
7844
- input: ["text", "image"],
7844
+ input: ["text"],
7845
7845
  cost: {
7846
7846
  input: 0,
7847
7847
  output: 0,
7848
7848
  cacheRead: 0,
7849
7849
  cacheWrite: 0,
7850
7850
  },
7851
- contextWindow: 200000,
7852
- maxTokens: 32000,
7851
+ contextWindow: 1000000,
7852
+ maxTokens: 128000,
7853
7853
  },
7854
- "nemotron-3-ultra-free": {
7855
- id: "nemotron-3-ultra-free",
7856
- name: "Nemotron 3 Ultra Free",
7854
+ "north-mini-code-free": {
7855
+ id: "north-mini-code-free",
7856
+ name: "North Mini Code Free",
7857
7857
  api: "openai-completions",
7858
7858
  provider: "opencode",
7859
7859
  baseUrl: "https://opencode.ai/zen/v1",
@@ -7865,8 +7865,8 @@ export const MODELS = {
7865
7865
  cacheRead: 0,
7866
7866
  cacheWrite: 0,
7867
7867
  },
7868
- contextWindow: 1000000,
7869
- maxTokens: 128000,
7868
+ contextWindow: 256000,
7869
+ maxTokens: 64000,
7870
7870
  },
7871
7871
  "qwen3.5-plus": {
7872
7872
  id: "qwen3.5-plus",
@@ -8087,9 +8087,9 @@ export const MODELS = {
8087
8087
  reasoning: true,
8088
8088
  input: ["text", "image"],
8089
8089
  cost: {
8090
- input: 0.6,
8091
- output: 2.4,
8092
- cacheRead: 0.12,
8090
+ input: 0.3,
8091
+ output: 1.2,
8092
+ cacheRead: 0.06,
8093
8093
  cacheWrite: 0,
8094
8094
  },
8095
8095
  contextWindow: 512000,
@@ -8268,6 +8268,24 @@ export const MODELS = {
8268
8268
  contextWindow: 200000,
8269
8269
  maxTokens: 4096,
8270
8270
  },
8271
+ "anthropic/claude-fable-5": {
8272
+ id: "anthropic/claude-fable-5",
8273
+ name: "Anthropic: Claude Fable 5",
8274
+ api: "openai-completions",
8275
+ provider: "openrouter",
8276
+ baseUrl: "https://openrouter.ai/api/v1",
8277
+ reasoning: true,
8278
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
8279
+ input: ["text", "image"],
8280
+ cost: {
8281
+ input: 10,
8282
+ output: 50,
8283
+ cacheRead: 1,
8284
+ cacheWrite: 12.5,
8285
+ },
8286
+ contextWindow: 1000000,
8287
+ maxTokens: 128000,
8288
+ },
8271
8289
  "anthropic/claude-haiku-4.5": {
8272
8290
  id: "anthropic/claude-haiku-4.5",
8273
8291
  name: "Anthropic: Claude Haiku 4.5",
@@ -9069,8 +9087,8 @@ export const MODELS = {
9069
9087
  reasoning: false,
9070
9088
  input: ["text", "image"],
9071
9089
  cost: {
9072
- input: 0.04,
9073
- output: 0.13,
9090
+ input: 0.049999999999999996,
9091
+ output: 0.15,
9074
9092
  cacheRead: 0,
9075
9093
  cacheWrite: 0,
9076
9094
  },
@@ -9358,7 +9376,7 @@ export const MODELS = {
9358
9376
  reasoning: false,
9359
9377
  input: ["text", "image"],
9360
9378
  cost: {
9361
- input: 0.08,
9379
+ input: 0.09999999999999999,
9362
9380
  output: 0.3,
9363
9381
  cacheRead: 0,
9364
9382
  cacheWrite: 0,
@@ -9427,8 +9445,8 @@ export const MODELS = {
9427
9445
  input: ["text"],
9428
9446
  cost: {
9429
9447
  input: 0.15,
9430
- output: 1.15,
9431
- cacheRead: 0,
9448
+ output: 0.8999999999999999,
9449
+ cacheRead: 0.049999999999999996,
9432
9450
  cacheWrite: 0,
9433
9451
  },
9434
9452
  contextWindow: 204800,
@@ -9443,13 +9461,13 @@ export const MODELS = {
9443
9461
  reasoning: true,
9444
9462
  input: ["text"],
9445
9463
  cost: {
9446
- input: 0.27899999999999997,
9447
- output: 1.2,
9448
- cacheRead: 0,
9464
+ input: 0.27,
9465
+ output: 1.08,
9466
+ cacheRead: 0.054,
9449
9467
  cacheWrite: 0,
9450
9468
  },
9451
9469
  contextWindow: 204800,
9452
- maxTokens: 196608,
9470
+ maxTokens: 131072,
9453
9471
  },
9454
9472
  "minimax/minimax-m3": {
9455
9473
  id: "minimax/minimax-m3",
@@ -9834,13 +9852,13 @@ export const MODELS = {
9834
9852
  reasoning: true,
9835
9853
  input: ["text", "image"],
9836
9854
  cost: {
9837
- input: 0.684,
9838
- output: 3.42,
9839
- cacheRead: 0.144,
9855
+ input: 0.6799999999999999,
9856
+ output: 3.41,
9857
+ cacheRead: 0.33999999999999997,
9840
9858
  cacheWrite: 0,
9841
9859
  },
9842
9860
  contextWindow: 262144,
9843
- maxTokens: 262144,
9861
+ maxTokens: 262142,
9844
9862
  },
9845
9863
  "moonshotai/kimi-k2.6:free": {
9846
9864
  id: "moonshotai/kimi-k2.6:free",
@@ -9859,22 +9877,22 @@ export const MODELS = {
9859
9877
  contextWindow: 262144,
9860
9878
  maxTokens: 4096,
9861
9879
  },
9862
- "nex-agi/deepseek-v3.1-nex-n1": {
9863
- id: "nex-agi/deepseek-v3.1-nex-n1",
9864
- name: "Nex AGI: DeepSeek V3.1 Nex N1",
9880
+ "nex-agi/nex-n2-pro:free": {
9881
+ id: "nex-agi/nex-n2-pro:free",
9882
+ name: "Nex AGI: Nex-N2-Pro (free)",
9865
9883
  api: "openai-completions",
9866
9884
  provider: "openrouter",
9867
9885
  baseUrl: "https://openrouter.ai/api/v1",
9868
- reasoning: false,
9869
- input: ["text"],
9886
+ reasoning: true,
9887
+ input: ["text", "image"],
9870
9888
  cost: {
9871
- input: 0.135,
9872
- output: 0.5,
9889
+ input: 0,
9890
+ output: 0,
9873
9891
  cacheRead: 0,
9874
9892
  cacheWrite: 0,
9875
9893
  },
9876
- contextWindow: 131072,
9877
- maxTokens: 163840,
9894
+ contextWindow: 262144,
9895
+ maxTokens: 262144,
9878
9896
  },
9879
9897
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
9880
9898
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -9885,7 +9903,7 @@ export const MODELS = {
9885
9903
  reasoning: true,
9886
9904
  input: ["text"],
9887
9905
  cost: {
9888
- input: 0.09999999999999999,
9906
+ input: 0.39999999999999997,
9889
9907
  output: 0.39999999999999997,
9890
9908
  cacheRead: 0,
9891
9909
  cacheWrite: 0,
@@ -11104,7 +11122,7 @@ export const MODELS = {
11104
11122
  reasoning: false,
11105
11123
  input: ["text"],
11106
11124
  cost: {
11107
- input: 0.071,
11125
+ input: 0.09,
11108
11126
  output: 0.09999999999999999,
11109
11127
  cacheRead: 0,
11110
11128
  cacheWrite: 0,
@@ -11138,8 +11156,8 @@ export const MODELS = {
11138
11156
  reasoning: true,
11139
11157
  input: ["text"],
11140
11158
  cost: {
11141
- input: 0.09,
11142
- output: 0.44999999999999996,
11159
+ input: 0.12,
11160
+ output: 0.5,
11143
11161
  cacheRead: 0,
11144
11162
  cacheWrite: 0,
11145
11163
  },
@@ -11597,13 +11615,13 @@ export const MODELS = {
11597
11615
  reasoning: true,
11598
11616
  input: ["text", "image"],
11599
11617
  cost: {
11600
- input: 0.04,
11618
+ input: 0.09999999999999999,
11601
11619
  output: 0.15,
11602
11620
  cacheRead: 0,
11603
11621
  cacheWrite: 0,
11604
11622
  },
11605
11623
  contextWindow: 262144,
11606
- maxTokens: 81920,
11624
+ maxTokens: 262144,
11607
11625
  },
11608
11626
  "qwen/qwen3.5-flash-02-23": {
11609
11627
  id: "qwen/qwen3.5-flash-02-23",
@@ -11996,23 +12014,6 @@ export const MODELS = {
11996
12014
  contextWindow: 1048576,
11997
12015
  maxTokens: 131072,
11998
12016
  },
11999
- "z-ai/glm-4-32b": {
12000
- id: "z-ai/glm-4-32b",
12001
- name: "Z.ai: GLM 4 32B ",
12002
- api: "openai-completions",
12003
- provider: "openrouter",
12004
- baseUrl: "https://openrouter.ai/api/v1",
12005
- reasoning: false,
12006
- input: ["text"],
12007
- cost: {
12008
- input: 0.09999999999999999,
12009
- output: 0.09999999999999999,
12010
- cacheRead: 0,
12011
- cacheWrite: 0,
12012
- },
12013
- contextWindow: 128000,
12014
- maxTokens: 4096,
12015
- },
12016
12017
  "z-ai/glm-4.5": {
12017
12018
  id: "z-ai/glm-4.5",
12018
12019
  name: "Z.ai: GLM 4.5",
@@ -12047,23 +12048,6 @@ export const MODELS = {
12047
12048
  contextWindow: 131072,
12048
12049
  maxTokens: 131070,
12049
12050
  },
12050
- "z-ai/glm-4.5-air:free": {
12051
- id: "z-ai/glm-4.5-air:free",
12052
- name: "Z.ai: GLM 4.5 Air (free)",
12053
- api: "openai-completions",
12054
- provider: "openrouter",
12055
- baseUrl: "https://openrouter.ai/api/v1",
12056
- reasoning: true,
12057
- input: ["text"],
12058
- cost: {
12059
- input: 0,
12060
- output: 0,
12061
- cacheRead: 0,
12062
- cacheWrite: 0,
12063
- },
12064
- contextWindow: 131072,
12065
- maxTokens: 96000,
12066
- },
12067
12051
  "z-ai/glm-4.5v": {
12068
12052
  id: "z-ai/glm-4.5v",
12069
12053
  name: "Z.ai: GLM 4.5V",
@@ -12109,11 +12093,11 @@ export const MODELS = {
12109
12093
  cost: {
12110
12094
  input: 0.3,
12111
12095
  output: 0.8999999999999999,
12112
- cacheRead: 0.049999999999999996,
12096
+ cacheRead: 0.055,
12113
12097
  cacheWrite: 0,
12114
12098
  },
12115
12099
  contextWindow: 131072,
12116
- maxTokens: 24000,
12100
+ maxTokens: 32768,
12117
12101
  },
12118
12102
  "z-ai/glm-4.7": {
12119
12103
  id: "z-ai/glm-4.7",
@@ -12180,7 +12164,7 @@ export const MODELS = {
12180
12164
  cacheRead: 0.24,
12181
12165
  cacheWrite: 0,
12182
12166
  },
12183
- contextWindow: 202752,
12167
+ contextWindow: 262144,
12184
12168
  maxTokens: 131072,
12185
12169
  },
12186
12170
  "z-ai/glm-5.1": {
@@ -12200,22 +12184,23 @@ export const MODELS = {
12200
12184
  contextWindow: 202752,
12201
12185
  maxTokens: 4096,
12202
12186
  },
12203
- "z-ai/glm-5v-turbo": {
12204
- id: "z-ai/glm-5v-turbo",
12205
- name: "Z.ai: GLM 5V Turbo",
12187
+ "~anthropic/claude-fable-latest": {
12188
+ id: "~anthropic/claude-fable-latest",
12189
+ name: "Anthropic: Claude Fable Latest",
12206
12190
  api: "openai-completions",
12207
12191
  provider: "openrouter",
12208
12192
  baseUrl: "https://openrouter.ai/api/v1",
12209
12193
  reasoning: true,
12194
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
12210
12195
  input: ["text", "image"],
12211
12196
  cost: {
12212
- input: 1.2,
12213
- output: 4,
12214
- cacheRead: 0.24,
12215
- cacheWrite: 0,
12197
+ input: 10,
12198
+ output: 50,
12199
+ cacheRead: 1,
12200
+ cacheWrite: 12.5,
12216
12201
  },
12217
- contextWindow: 202752,
12218
- maxTokens: 131072,
12202
+ contextWindow: 1000000,
12203
+ maxTokens: 128000,
12219
12204
  },
12220
12205
  "~anthropic/claude-haiku-latest": {
12221
12206
  id: "~anthropic/claude-haiku-latest",
@@ -12311,13 +12296,13 @@ export const MODELS = {
12311
12296
  reasoning: true,
12312
12297
  input: ["text", "image"],
12313
12298
  cost: {
12314
- input: 0.684,
12315
- output: 3.42,
12316
- cacheRead: 0.144,
12299
+ input: 0.6799999999999999,
12300
+ output: 3.41,
12301
+ cacheRead: 0.33999999999999997,
12317
12302
  cacheWrite: 0,
12318
12303
  },
12319
12304
  contextWindow: 262144,
12320
- maxTokens: 262144,
12305
+ maxTokens: 262142,
12321
12306
  },
12322
12307
  "~openai/gpt-latest": {
12323
12308
  id: "~openai/gpt-latest",
@@ -12758,8 +12743,8 @@ export const MODELS = {
12758
12743
  reasoning: true,
12759
12744
  input: ["text"],
12760
12745
  cost: {
12761
- input: 0.08,
12762
- output: 0.29,
12746
+ input: 0.12,
12747
+ output: 0.5,
12763
12748
  cacheRead: 0,
12764
12749
  cacheWrite: 0,
12765
12750
  },
@@ -13106,6 +13091,24 @@ export const MODELS = {
13106
13091
  contextWindow: 200000,
13107
13092
  maxTokens: 4096,
13108
13093
  },
13094
+ "anthropic/claude-fable-5": {
13095
+ id: "anthropic/claude-fable-5",
13096
+ name: "Claude Fable 5",
13097
+ api: "anthropic-messages",
13098
+ provider: "vercel-ai-gateway",
13099
+ baseUrl: "https://ai-gateway.vercel.sh",
13100
+ reasoning: true,
13101
+ thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "xhigh", "xhigh": "max" },
13102
+ input: ["text", "image"],
13103
+ cost: {
13104
+ input: 10,
13105
+ output: 50,
13106
+ cacheRead: 1,
13107
+ cacheWrite: 12.5,
13108
+ },
13109
+ contextWindow: 1000000,
13110
+ maxTokens: 128000,
13111
+ },
13109
13112
  "anthropic/claude-haiku-4.5": {
13110
13113
  id: "anthropic/claude-haiku-4.5",
13111
13114
  name: "Claude Haiku 4.5",
@@ -13483,40 +13486,6 @@ export const MODELS = {
13483
13486
  contextWindow: 1000000,
13484
13487
  maxTokens: 384000,
13485
13488
  },
13486
- "google/gemini-2.0-flash": {
13487
- id: "google/gemini-2.0-flash",
13488
- name: "Gemini 2.0 Flash",
13489
- api: "anthropic-messages",
13490
- provider: "vercel-ai-gateway",
13491
- baseUrl: "https://ai-gateway.vercel.sh",
13492
- reasoning: false,
13493
- input: ["text", "image"],
13494
- cost: {
13495
- input: 0.15,
13496
- output: 0.6,
13497
- cacheRead: 0.024999999999999998,
13498
- cacheWrite: 0,
13499
- },
13500
- contextWindow: 1048576,
13501
- maxTokens: 8192,
13502
- },
13503
- "google/gemini-2.0-flash-lite": {
13504
- id: "google/gemini-2.0-flash-lite",
13505
- name: "Gemini 2.0 Flash Lite",
13506
- api: "anthropic-messages",
13507
- provider: "vercel-ai-gateway",
13508
- baseUrl: "https://ai-gateway.vercel.sh",
13509
- reasoning: false,
13510
- input: ["text", "image"],
13511
- cost: {
13512
- input: 0.075,
13513
- output: 0.3,
13514
- cacheRead: 0.02,
13515
- cacheWrite: 0,
13516
- },
13517
- contextWindow: 1048576,
13518
- maxTokens: 8192,
13519
- },
13520
13489
  "google/gemini-2.5-flash": {
13521
13490
  id: "google/gemini-2.5-flash",
13522
13491
  name: "Gemini 2.5 Flash",
@@ -15551,7 +15520,7 @@ export const MODELS = {
15551
15520
  cacheRead: 0.2,
15552
15521
  cacheWrite: 0,
15553
15522
  },
15554
- contextWindow: 2000000,
15523
+ contextWindow: 1000000,
15555
15524
  maxTokens: 30000,
15556
15525
  },
15557
15526
  "grok-4.20-0309-reasoning": {
@@ -15568,7 +15537,7 @@ export const MODELS = {
15568
15537
  cacheRead: 0.2,
15569
15538
  cacheWrite: 0,
15570
15539
  },
15571
- contextWindow: 2000000,
15540
+ contextWindow: 1000000,
15572
15541
  maxTokens: 30000,
15573
15542
  },
15574
15543
  "grok-4.3": {
@@ -15658,6 +15627,23 @@ export const MODELS = {
15658
15627
  contextWindow: 1048576,
15659
15628
  maxTokens: 131072,
15660
15629
  },
15630
+ "mimo-v2.5-pro-ultraspeed": {
15631
+ id: "mimo-v2.5-pro-ultraspeed",
15632
+ name: "MiMo-V2.5-Pro-UltraSpeed",
15633
+ api: "anthropic-messages",
15634
+ provider: "xiaomi",
15635
+ baseUrl: "https://api.xiaomimimo.com/anthropic",
15636
+ reasoning: true,
15637
+ input: ["text"],
15638
+ cost: {
15639
+ input: 1.305,
15640
+ output: 2.61,
15641
+ cacheRead: 0.0108,
15642
+ cacheWrite: 0,
15643
+ },
15644
+ contextWindow: 1048576,
15645
+ maxTokens: 131072,
15646
+ },
15661
15647
  },
15662
15648
  "xiaomi-token-plan-ams": {
15663
15649
  "mimo-v2.5": {
@@ -15694,6 +15680,23 @@ export const MODELS = {
15694
15680
  contextWindow: 1048576,
15695
15681
  maxTokens: 131072,
15696
15682
  },
15683
+ "mimo-v2.5-pro-ultraspeed": {
15684
+ id: "mimo-v2.5-pro-ultraspeed",
15685
+ name: "MiMo-V2.5-Pro-UltraSpeed",
15686
+ api: "anthropic-messages",
15687
+ provider: "xiaomi-token-plan-ams",
15688
+ baseUrl: "https://token-plan-ams.xiaomimimo.com/anthropic",
15689
+ reasoning: true,
15690
+ input: ["text"],
15691
+ cost: {
15692
+ input: 1.305,
15693
+ output: 2.61,
15694
+ cacheRead: 0.0108,
15695
+ cacheWrite: 0,
15696
+ },
15697
+ contextWindow: 1048576,
15698
+ maxTokens: 131072,
15699
+ },
15697
15700
  },
15698
15701
  "xiaomi-token-plan-cn": {
15699
15702
  "mimo-v2.5": {
@@ -15730,6 +15733,23 @@ export const MODELS = {
15730
15733
  contextWindow: 1048576,
15731
15734
  maxTokens: 131072,
15732
15735
  },
15736
+ "mimo-v2.5-pro-ultraspeed": {
15737
+ id: "mimo-v2.5-pro-ultraspeed",
15738
+ name: "MiMo-V2.5-Pro-UltraSpeed",
15739
+ api: "anthropic-messages",
15740
+ provider: "xiaomi-token-plan-cn",
15741
+ baseUrl: "https://token-plan-cn.xiaomimimo.com/anthropic",
15742
+ reasoning: true,
15743
+ input: ["text"],
15744
+ cost: {
15745
+ input: 1.305,
15746
+ output: 2.61,
15747
+ cacheRead: 0.0108,
15748
+ cacheWrite: 0,
15749
+ },
15750
+ contextWindow: 1048576,
15751
+ maxTokens: 131072,
15752
+ },
15733
15753
  },
15734
15754
  "xiaomi-token-plan-sgp": {
15735
15755
  "mimo-v2.5": {
@@ -15766,6 +15786,23 @@ export const MODELS = {
15766
15786
  contextWindow: 1048576,
15767
15787
  maxTokens: 131072,
15768
15788
  },
15789
+ "mimo-v2.5-pro-ultraspeed": {
15790
+ id: "mimo-v2.5-pro-ultraspeed",
15791
+ name: "MiMo-V2.5-Pro-UltraSpeed",
15792
+ api: "anthropic-messages",
15793
+ provider: "xiaomi-token-plan-sgp",
15794
+ baseUrl: "https://token-plan-sgp.xiaomimimo.com/anthropic",
15795
+ reasoning: true,
15796
+ input: ["text"],
15797
+ cost: {
15798
+ input: 1.305,
15799
+ output: 2.61,
15800
+ cacheRead: 0.0108,
15801
+ cacheWrite: 0,
15802
+ },
15803
+ contextWindow: 1048576,
15804
+ maxTokens: 131072,
15805
+ },
15769
15806
  },
15770
15807
  "zai": {
15771
15808
  "glm-4.5-air": {