@fleetagent/pi-ai 0.2.1 → 0.2.3

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.
@@ -276,6 +276,57 @@ export const MODELS = {
276
276
  contextWindow: 1000000,
277
277
  maxTokens: 128000,
278
278
  },
279
+ "apac.amazon.nova-lite-v1:0": {
280
+ id: "apac.amazon.nova-lite-v1:0",
281
+ name: "Nova Lite (APAC)",
282
+ api: "bedrock-converse-stream",
283
+ provider: "amazon-bedrock",
284
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
285
+ reasoning: false,
286
+ input: ["text", "image"],
287
+ cost: {
288
+ input: 0.063,
289
+ output: 0.252,
290
+ cacheRead: 0.01575,
291
+ cacheWrite: 0,
292
+ },
293
+ contextWindow: 300000,
294
+ maxTokens: 8192,
295
+ },
296
+ "apac.amazon.nova-micro-v1:0": {
297
+ id: "apac.amazon.nova-micro-v1:0",
298
+ name: "Nova Micro (APAC)",
299
+ api: "bedrock-converse-stream",
300
+ provider: "amazon-bedrock",
301
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
302
+ reasoning: false,
303
+ input: ["text"],
304
+ cost: {
305
+ input: 0.037,
306
+ output: 0.148,
307
+ cacheRead: 0.00925,
308
+ cacheWrite: 0,
309
+ },
310
+ contextWindow: 128000,
311
+ maxTokens: 8192,
312
+ },
313
+ "apac.amazon.nova-pro-v1:0": {
314
+ id: "apac.amazon.nova-pro-v1:0",
315
+ name: "Nova Pro (APAC)",
316
+ api: "bedrock-converse-stream",
317
+ provider: "amazon-bedrock",
318
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
319
+ reasoning: false,
320
+ input: ["text", "image"],
321
+ cost: {
322
+ input: 0.84,
323
+ output: 3.36,
324
+ cacheRead: 0.21,
325
+ cacheWrite: 0,
326
+ },
327
+ contextWindow: 300000,
328
+ maxTokens: 8192,
329
+ },
279
330
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
280
331
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
281
332
  name: "Claude Haiku 4.5 (AU)",
@@ -311,6 +362,24 @@ export const MODELS = {
311
362
  contextWindow: 1000000,
312
363
  maxTokens: 128000,
313
364
  },
365
+ "au.anthropic.claude-opus-4-7": {
366
+ id: "au.anthropic.claude-opus-4-7",
367
+ name: "Claude Opus 4.7 (AU)",
368
+ api: "bedrock-converse-stream",
369
+ provider: "amazon-bedrock",
370
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
371
+ reasoning: true,
372
+ thinkingLevelMap: { "xhigh": "xhigh" },
373
+ input: ["text", "image"],
374
+ cost: {
375
+ input: 5.5,
376
+ output: 27.5,
377
+ cacheRead: 0.55,
378
+ cacheWrite: 6.875,
379
+ },
380
+ contextWindow: 1000000,
381
+ maxTokens: 128000,
382
+ },
314
383
  "au.anthropic.claude-opus-4-8": {
315
384
  id: "au.anthropic.claude-opus-4-8",
316
385
  name: "Claude Opus 4.8 (AU)",
@@ -396,6 +465,23 @@ export const MODELS = {
396
465
  contextWindow: 1000000,
397
466
  maxTokens: 128000,
398
467
  },
468
+ "ca.amazon.nova-lite-v1:0": {
469
+ id: "ca.amazon.nova-lite-v1:0",
470
+ name: "Nova Lite (CA)",
471
+ api: "bedrock-converse-stream",
472
+ provider: "amazon-bedrock",
473
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
474
+ reasoning: false,
475
+ input: ["text", "image"],
476
+ cost: {
477
+ input: 0.064,
478
+ output: 0.256,
479
+ cacheRead: 0.016,
480
+ cacheWrite: 0,
481
+ },
482
+ contextWindow: 300000,
483
+ maxTokens: 8192,
484
+ },
399
485
  "deepseek.r1-v1:0": {
400
486
  id: "deepseek.r1-v1:0",
401
487
  name: "DeepSeek-R1",
@@ -447,6 +533,74 @@ export const MODELS = {
447
533
  contextWindow: 163840,
448
534
  maxTokens: 81920,
449
535
  },
536
+ "eu.amazon.nova-2-lite-v1:0": {
537
+ id: "eu.amazon.nova-2-lite-v1:0",
538
+ name: "Nova 2 Lite (EU)",
539
+ api: "bedrock-converse-stream",
540
+ provider: "amazon-bedrock",
541
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
542
+ reasoning: true,
543
+ input: ["text", "image"],
544
+ cost: {
545
+ input: 0.374,
546
+ output: 3.157,
547
+ cacheRead: 0.0935,
548
+ cacheWrite: 0,
549
+ },
550
+ contextWindow: 1000000,
551
+ maxTokens: 64000,
552
+ },
553
+ "eu.amazon.nova-lite-v1:0": {
554
+ id: "eu.amazon.nova-lite-v1:0",
555
+ name: "Nova Lite (EU)",
556
+ api: "bedrock-converse-stream",
557
+ provider: "amazon-bedrock",
558
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
559
+ reasoning: false,
560
+ input: ["text", "image"],
561
+ cost: {
562
+ input: 0.069,
563
+ output: 0.276,
564
+ cacheRead: 0.01725,
565
+ cacheWrite: 0,
566
+ },
567
+ contextWindow: 300000,
568
+ maxTokens: 8192,
569
+ },
570
+ "eu.amazon.nova-micro-v1:0": {
571
+ id: "eu.amazon.nova-micro-v1:0",
572
+ name: "Nova Micro (EU)",
573
+ api: "bedrock-converse-stream",
574
+ provider: "amazon-bedrock",
575
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
576
+ reasoning: false,
577
+ input: ["text"],
578
+ cost: {
579
+ input: 0.04,
580
+ output: 0.16,
581
+ cacheRead: 0.01,
582
+ cacheWrite: 0,
583
+ },
584
+ contextWindow: 128000,
585
+ maxTokens: 8192,
586
+ },
587
+ "eu.amazon.nova-pro-v1:0": {
588
+ id: "eu.amazon.nova-pro-v1:0",
589
+ name: "Nova Pro (EU)",
590
+ api: "bedrock-converse-stream",
591
+ provider: "amazon-bedrock",
592
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
593
+ reasoning: false,
594
+ input: ["text", "image"],
595
+ cost: {
596
+ input: 0.92,
597
+ output: 3.68,
598
+ cacheRead: 0.23,
599
+ cacheWrite: 0,
600
+ },
601
+ contextWindow: 300000,
602
+ maxTokens: 8192,
603
+ },
450
604
  "eu.anthropic.claude-fable-5": {
451
605
  id: "eu.anthropic.claude-fable-5",
452
606
  name: "Claude Fable 5 (EU)",
@@ -619,6 +773,40 @@ export const MODELS = {
619
773
  contextWindow: 1000000,
620
774
  maxTokens: 128000,
621
775
  },
776
+ "eu.mistral.pixtral-large-2502-v1:0": {
777
+ id: "eu.mistral.pixtral-large-2502-v1:0",
778
+ name: "Pixtral Large (25.02) (EU)",
779
+ api: "bedrock-converse-stream",
780
+ provider: "amazon-bedrock",
781
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
782
+ reasoning: false,
783
+ input: ["text", "image"],
784
+ cost: {
785
+ input: 2,
786
+ output: 6,
787
+ cacheRead: 0,
788
+ cacheWrite: 0,
789
+ },
790
+ contextWindow: 128000,
791
+ maxTokens: 8192,
792
+ },
793
+ "global.amazon.nova-2-lite-v1:0": {
794
+ id: "global.amazon.nova-2-lite-v1:0",
795
+ name: "Nova 2 Lite (Global)",
796
+ api: "bedrock-converse-stream",
797
+ provider: "amazon-bedrock",
798
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
799
+ reasoning: true,
800
+ input: ["text", "image"],
801
+ cost: {
802
+ input: 0.3,
803
+ output: 2.5,
804
+ cacheRead: 0.075,
805
+ cacheWrite: 0,
806
+ },
807
+ contextWindow: 1000000,
808
+ maxTokens: 64000,
809
+ },
622
810
  "global.anthropic.claude-fable-5": {
623
811
  id: "global.anthropic.claude-fable-5",
624
812
  name: "Claude Fable 5 (Global)",
@@ -862,6 +1050,24 @@ export const MODELS = {
862
1050
  contextWindow: 1050000,
863
1051
  maxTokens: 128000,
864
1052
  },
1053
+ "global.openai.gpt-6-astra": {
1054
+ id: "global.openai.gpt-6-astra",
1055
+ name: "GPT-6 Astra (Global)",
1056
+ api: "bedrock-converse-stream",
1057
+ provider: "amazon-bedrock",
1058
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1059
+ reasoning: true,
1060
+ thinkingLevelMap: { "xhigh": "xhigh" },
1061
+ input: ["text", "image"],
1062
+ cost: {
1063
+ input: 10,
1064
+ output: 50,
1065
+ cacheRead: 1,
1066
+ cacheWrite: 12.5,
1067
+ },
1068
+ contextWindow: 1050000,
1069
+ maxTokens: 128000,
1070
+ },
865
1071
  "global.xai.grok-4.6": {
866
1072
  id: "global.xai.grok-4.6",
867
1073
  name: "Grok 4.6 (Global)",
@@ -913,6 +1119,23 @@ export const MODELS = {
913
1119
  contextWindow: 128000,
914
1120
  maxTokens: 4096,
915
1121
  },
1122
+ "jp.amazon.nova-2-lite-v1:0": {
1123
+ id: "jp.amazon.nova-2-lite-v1:0",
1124
+ name: "Nova 2 Lite (JP)",
1125
+ api: "bedrock-converse-stream",
1126
+ provider: "amazon-bedrock",
1127
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1128
+ reasoning: true,
1129
+ input: ["text", "image"],
1130
+ cost: {
1131
+ input: 0.396,
1132
+ output: 3.311,
1133
+ cacheRead: 0.099,
1134
+ cacheWrite: 0,
1135
+ },
1136
+ contextWindow: 1000000,
1137
+ maxTokens: 64000,
1138
+ },
916
1139
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
917
1140
  id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
918
1141
  name: "Claude Haiku 4.5 (JP)",
@@ -1514,6 +1737,24 @@ export const MODELS = {
1514
1737
  contextWindow: 1050000,
1515
1738
  maxTokens: 128000,
1516
1739
  },
1740
+ "openai.gpt-6-astra": {
1741
+ id: "openai.gpt-6-astra",
1742
+ name: "GPT-6 Astra",
1743
+ api: "bedrock-converse-stream",
1744
+ provider: "amazon-bedrock",
1745
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1746
+ reasoning: true,
1747
+ thinkingLevelMap: { "xhigh": "xhigh" },
1748
+ input: ["text", "image"],
1749
+ cost: {
1750
+ input: 11,
1751
+ output: 55,
1752
+ cacheRead: 1.1,
1753
+ cacheWrite: 13.75,
1754
+ },
1755
+ contextWindow: 1050000,
1756
+ maxTokens: 128000,
1757
+ },
1517
1758
  "openai.gpt-oss-120b": {
1518
1759
  id: "openai.gpt-oss-120b",
1519
1760
  name: "gpt-oss-120b",
@@ -1735,6 +1976,74 @@ export const MODELS = {
1735
1976
  contextWindow: 262000,
1736
1977
  maxTokens: 262000,
1737
1978
  },
1979
+ "us.amazon.nova-2-lite-v1:0": {
1980
+ id: "us.amazon.nova-2-lite-v1:0",
1981
+ name: "Nova 2 Lite (US)",
1982
+ api: "bedrock-converse-stream",
1983
+ provider: "amazon-bedrock",
1984
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1985
+ reasoning: true,
1986
+ input: ["text", "image"],
1987
+ cost: {
1988
+ input: 0.33,
1989
+ output: 2.75,
1990
+ cacheRead: 0.0825,
1991
+ cacheWrite: 0,
1992
+ },
1993
+ contextWindow: 1000000,
1994
+ maxTokens: 64000,
1995
+ },
1996
+ "us.amazon.nova-lite-v1:0": {
1997
+ id: "us.amazon.nova-lite-v1:0",
1998
+ name: "Nova Lite (US)",
1999
+ api: "bedrock-converse-stream",
2000
+ provider: "amazon-bedrock",
2001
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2002
+ reasoning: false,
2003
+ input: ["text", "image"],
2004
+ cost: {
2005
+ input: 0.06,
2006
+ output: 0.24,
2007
+ cacheRead: 0.015,
2008
+ cacheWrite: 0,
2009
+ },
2010
+ contextWindow: 300000,
2011
+ maxTokens: 8192,
2012
+ },
2013
+ "us.amazon.nova-micro-v1:0": {
2014
+ id: "us.amazon.nova-micro-v1:0",
2015
+ name: "Nova Micro (US)",
2016
+ api: "bedrock-converse-stream",
2017
+ provider: "amazon-bedrock",
2018
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2019
+ reasoning: false,
2020
+ input: ["text"],
2021
+ cost: {
2022
+ input: 0.035,
2023
+ output: 0.14,
2024
+ cacheRead: 0.00875,
2025
+ cacheWrite: 0,
2026
+ },
2027
+ contextWindow: 128000,
2028
+ maxTokens: 8192,
2029
+ },
2030
+ "us.amazon.nova-pro-v1:0": {
2031
+ id: "us.amazon.nova-pro-v1:0",
2032
+ name: "Nova Pro (US)",
2033
+ api: "bedrock-converse-stream",
2034
+ provider: "amazon-bedrock",
2035
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2036
+ reasoning: false,
2037
+ input: ["text", "image"],
2038
+ cost: {
2039
+ input: 0.8,
2040
+ output: 3.2,
2041
+ cacheRead: 0.2,
2042
+ cacheWrite: 0,
2043
+ },
2044
+ contextWindow: 300000,
2045
+ maxTokens: 8192,
2046
+ },
1738
2047
  "us.anthropic.claude-fable-5": {
1739
2048
  id: "us.anthropic.claude-fable-5",
1740
2049
  name: "Claude Fable 5 (US)",
@@ -1958,6 +2267,57 @@ export const MODELS = {
1958
2267
  contextWindow: 128000,
1959
2268
  maxTokens: 32768,
1960
2269
  },
2270
+ "us.meta.llama3-1-70b-instruct-v1:0": {
2271
+ id: "us.meta.llama3-1-70b-instruct-v1:0",
2272
+ name: "Llama 3.1 70B Instruct (US)",
2273
+ api: "bedrock-converse-stream",
2274
+ provider: "amazon-bedrock",
2275
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2276
+ reasoning: false,
2277
+ input: ["text"],
2278
+ cost: {
2279
+ input: 0.72,
2280
+ output: 0.72,
2281
+ cacheRead: 0,
2282
+ cacheWrite: 0,
2283
+ },
2284
+ contextWindow: 128000,
2285
+ maxTokens: 4096,
2286
+ },
2287
+ "us.meta.llama3-1-8b-instruct-v1:0": {
2288
+ id: "us.meta.llama3-1-8b-instruct-v1:0",
2289
+ name: "Llama 3.1 8B Instruct (US)",
2290
+ api: "bedrock-converse-stream",
2291
+ provider: "amazon-bedrock",
2292
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2293
+ reasoning: false,
2294
+ input: ["text"],
2295
+ cost: {
2296
+ input: 0.22,
2297
+ output: 0.22,
2298
+ cacheRead: 0,
2299
+ cacheWrite: 0,
2300
+ },
2301
+ contextWindow: 128000,
2302
+ maxTokens: 4096,
2303
+ },
2304
+ "us.meta.llama3-3-70b-instruct-v1:0": {
2305
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
2306
+ name: "Llama 3.3 70B Instruct (US)",
2307
+ api: "bedrock-converse-stream",
2308
+ provider: "amazon-bedrock",
2309
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2310
+ reasoning: false,
2311
+ input: ["text"],
2312
+ cost: {
2313
+ input: 0.72,
2314
+ output: 0.72,
2315
+ cacheRead: 0,
2316
+ cacheWrite: 0,
2317
+ },
2318
+ contextWindow: 128000,
2319
+ maxTokens: 4096,
2320
+ },
1961
2321
  "us.meta.llama4-maverick-17b-instruct-v1:0": {
1962
2322
  id: "us.meta.llama4-maverick-17b-instruct-v1:0",
1963
2323
  name: "Llama 4 Maverick 17B Instruct (US)",
@@ -1992,6 +2352,129 @@ export const MODELS = {
1992
2352
  contextWindow: 3500000,
1993
2353
  maxTokens: 16384,
1994
2354
  },
2355
+ "us.mistral.pixtral-large-2502-v1:0": {
2356
+ id: "us.mistral.pixtral-large-2502-v1:0",
2357
+ name: "Pixtral Large (25.02) (US)",
2358
+ api: "bedrock-converse-stream",
2359
+ provider: "amazon-bedrock",
2360
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2361
+ reasoning: false,
2362
+ input: ["text", "image"],
2363
+ cost: {
2364
+ input: 2,
2365
+ output: 6,
2366
+ cacheRead: 0,
2367
+ cacheWrite: 0,
2368
+ },
2369
+ contextWindow: 128000,
2370
+ maxTokens: 8192,
2371
+ },
2372
+ "us.openai.gpt-5.6-luna": {
2373
+ id: "us.openai.gpt-5.6-luna",
2374
+ name: "GPT-5.6 Luna (US)",
2375
+ api: "bedrock-converse-stream",
2376
+ provider: "amazon-bedrock",
2377
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2378
+ reasoning: true,
2379
+ thinkingLevelMap: { "xhigh": "xhigh" },
2380
+ input: ["text", "image"],
2381
+ cost: {
2382
+ input: 0.22,
2383
+ output: 1.32,
2384
+ cacheRead: 0.022,
2385
+ cacheWrite: 0.275,
2386
+ },
2387
+ contextWindow: 1050000,
2388
+ maxTokens: 128000,
2389
+ },
2390
+ "us.openai.gpt-5.6-sol": {
2391
+ id: "us.openai.gpt-5.6-sol",
2392
+ name: "GPT-5.6 Sol (US)",
2393
+ api: "bedrock-converse-stream",
2394
+ provider: "amazon-bedrock",
2395
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2396
+ reasoning: true,
2397
+ thinkingLevelMap: { "xhigh": "xhigh" },
2398
+ input: ["text", "image"],
2399
+ cost: {
2400
+ input: 4.4,
2401
+ output: 22,
2402
+ cacheRead: 0.44,
2403
+ cacheWrite: 5.5,
2404
+ },
2405
+ contextWindow: 1050000,
2406
+ maxTokens: 128000,
2407
+ },
2408
+ "us.openai.gpt-5.6-terra": {
2409
+ id: "us.openai.gpt-5.6-terra",
2410
+ name: "GPT-5.6 Terra (US)",
2411
+ api: "bedrock-converse-stream",
2412
+ provider: "amazon-bedrock",
2413
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2414
+ reasoning: true,
2415
+ thinkingLevelMap: { "xhigh": "xhigh" },
2416
+ input: ["text", "image"],
2417
+ cost: {
2418
+ input: 2.2,
2419
+ output: 13.2,
2420
+ cacheRead: 0.22,
2421
+ cacheWrite: 2.75,
2422
+ },
2423
+ contextWindow: 1050000,
2424
+ maxTokens: 128000,
2425
+ },
2426
+ "us.openai.gpt-6-astra": {
2427
+ id: "us.openai.gpt-6-astra",
2428
+ name: "GPT-6 Astra (US)",
2429
+ api: "bedrock-converse-stream",
2430
+ provider: "amazon-bedrock",
2431
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2432
+ reasoning: true,
2433
+ thinkingLevelMap: { "xhigh": "xhigh" },
2434
+ input: ["text", "image"],
2435
+ cost: {
2436
+ input: 11,
2437
+ output: 55,
2438
+ cacheRead: 1.1,
2439
+ cacheWrite: 13.75,
2440
+ },
2441
+ contextWindow: 1050000,
2442
+ maxTokens: 128000,
2443
+ },
2444
+ "us.writer.palmyra-x4-v1:0": {
2445
+ id: "us.writer.palmyra-x4-v1:0",
2446
+ name: "Palmyra X4 (US)",
2447
+ api: "bedrock-converse-stream",
2448
+ provider: "amazon-bedrock",
2449
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2450
+ reasoning: true,
2451
+ input: ["text"],
2452
+ cost: {
2453
+ input: 2.5,
2454
+ output: 10,
2455
+ cacheRead: 0,
2456
+ cacheWrite: 0,
2457
+ },
2458
+ contextWindow: 122880,
2459
+ maxTokens: 8192,
2460
+ },
2461
+ "us.writer.palmyra-x5-v1:0": {
2462
+ id: "us.writer.palmyra-x5-v1:0",
2463
+ name: "Palmyra X5 (US)",
2464
+ api: "bedrock-converse-stream",
2465
+ provider: "amazon-bedrock",
2466
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2467
+ reasoning: true,
2468
+ input: ["text"],
2469
+ cost: {
2470
+ input: 0.6,
2471
+ output: 6,
2472
+ cacheRead: 0,
2473
+ cacheWrite: 0,
2474
+ },
2475
+ contextWindow: 1040000,
2476
+ maxTokens: 8192,
2477
+ },
1995
2478
  "us.xai.grok-4.6": {
1996
2479
  id: "us.xai.grok-4.6",
1997
2480
  name: "Grok 4.6 (US)",
@@ -2153,7 +2636,9 @@ export const MODELS = {
2153
2636
  api: "anthropic-messages",
2154
2637
  provider: "anthropic",
2155
2638
  baseUrl: "https://api.anthropic.com",
2639
+ compat: { "supportsTemperature": false },
2156
2640
  reasoning: true,
2641
+ thinkingLevelMap: { "off": "low", "minimal": "low", "xhigh": "xhigh" },
2157
2642
  input: ["text", "image"],
2158
2643
  cost: {
2159
2644
  input: 10,
@@ -2985,6 +3470,7 @@ export const MODELS = {
2985
3470
  provider: "azure-openai-responses",
2986
3471
  baseUrl: "",
2987
3472
  reasoning: true,
3473
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
2988
3474
  input: ["text", "image"],
2989
3475
  cost: {
2990
3476
  input: 10,
@@ -4904,12 +5390,12 @@ export const MODELS = {
4904
5390
  "gpt-6-astra": {
4905
5391
  id: "gpt-6-astra",
4906
5392
  name: "GPT-6 Astra",
4907
- api: "openai-completions",
5393
+ api: "openai-responses",
4908
5394
  provider: "github-copilot",
4909
5395
  baseUrl: "https://api.individual.githubcopilot.com",
4910
5396
  headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4911
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4912
5397
  reasoning: true,
5398
+ thinkingLevelMap: { "off": null, "minimal": "low", "xhigh": "xhigh" },
4913
5399
  input: ["text", "image"],
4914
5400
  cost: {
4915
5401
  input: 10,
@@ -8504,6 +8990,7 @@ export const MODELS = {
8504
8990
  provider: "openai",
8505
8991
  baseUrl: "https://api.openai.com/v1",
8506
8992
  reasoning: true,
8993
+ thinkingLevelMap: { "off": "none", "xhigh": "xhigh", "minimal": null },
8507
8994
  input: ["text", "image"],
8508
8995
  cost: {
8509
8996
  input: 10,
@@ -8815,6 +9302,24 @@ export const MODELS = {
8815
9302
  contextWindow: 272000,
8816
9303
  maxTokens: 128000,
8817
9304
  },
9305
+ "gpt-6-astra": {
9306
+ id: "gpt-6-astra",
9307
+ name: "GPT-6 Astra",
9308
+ api: "openai-codex-responses",
9309
+ provider: "openai-codex",
9310
+ baseUrl: "https://chatgpt.com/backend-api",
9311
+ reasoning: true,
9312
+ thinkingLevelMap: { "xhigh": "xhigh", "minimal": "low" },
9313
+ input: ["text", "image"],
9314
+ cost: {
9315
+ input: 10,
9316
+ output: 50,
9317
+ cacheRead: 1,
9318
+ cacheWrite: 12.5,
9319
+ },
9320
+ contextWindow: 272000,
9321
+ maxTokens: 128000,
9322
+ },
8818
9323
  },
8819
9324
  "opencode": {
8820
9325
  "big-pickle": {
@@ -9661,6 +10166,7 @@ export const MODELS = {
9661
10166
  provider: "opencode",
9662
10167
  baseUrl: "https://opencode.ai/zen/v1",
9663
10168
  reasoning: true,
10169
+ thinkingLevelMap: { "off": null, "xhigh": "xhigh" },
9664
10170
  input: ["text", "image"],
9665
10171
  cost: {
9666
10172
  input: 10,
@@ -10016,6 +10522,23 @@ export const MODELS = {
10016
10522
  },
10017
10523
  },
10018
10524
  "opencode-go": {
10525
+ "deepseek-flash": {
10526
+ id: "deepseek-flash",
10527
+ name: "DeepSeek V4.1 Flash",
10528
+ api: "openai-completions",
10529
+ provider: "opencode-go",
10530
+ baseUrl: "https://opencode.ai/zen/go/v1",
10531
+ reasoning: true,
10532
+ input: ["text", "image"],
10533
+ cost: {
10534
+ input: 0.15,
10535
+ output: 0.6,
10536
+ cacheRead: 0.003,
10537
+ cacheWrite: 0,
10538
+ },
10539
+ contextWindow: 1000000,
10540
+ maxTokens: 384000,
10541
+ },
10019
10542
  "deepseek-v4-flash": {
10020
10543
  id: "deepseek-v4-flash",
10021
10544
  name: "DeepSeek V4 Flash",
@@ -10027,9 +10550,9 @@ export const MODELS = {
10027
10550
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10028
10551
  input: ["text"],
10029
10552
  cost: {
10030
- input: 0.22,
10031
- output: 0.66,
10032
- cacheRead: 0.007,
10553
+ input: 0.15,
10554
+ output: 0.6,
10555
+ cacheRead: 0.003,
10033
10556
  cacheWrite: 0,
10034
10557
  },
10035
10558
  contextWindow: 1000000,
@@ -10046,9 +10569,9 @@ export const MODELS = {
10046
10569
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10047
10570
  input: ["text", "image"],
10048
10571
  cost: {
10049
- input: 0.22,
10050
- output: 0.66,
10051
- cacheRead: 0.007,
10572
+ input: 0.15,
10573
+ output: 0.6,
10574
+ cacheRead: 0.003,
10052
10575
  cacheWrite: 0,
10053
10576
  },
10054
10577
  contextWindow: 1000000,
@@ -10126,16 +10649,16 @@ export const MODELS = {
10126
10649
  },
10127
10650
  "glm-5.3-flash": {
10128
10651
  id: "glm-5.3-flash",
10129
- name: "GLM-5.3-Flash (2x usage)",
10652
+ name: "GLM-5.3-Flash",
10130
10653
  api: "openai-completions",
10131
10654
  provider: "opencode-go",
10132
10655
  baseUrl: "https://opencode.ai/zen/go/v1",
10133
10656
  reasoning: true,
10134
10657
  input: ["text", "image"],
10135
10658
  cost: {
10136
- input: 0.075,
10137
- output: 0.25,
10138
- cacheRead: 0.015,
10659
+ input: 0.15,
10660
+ output: 0.5,
10661
+ cacheRead: 0.03,
10139
10662
  cacheWrite: 0,
10140
10663
  },
10141
10664
  contextWindow: 1000000,
@@ -10382,23 +10905,6 @@ export const MODELS = {
10382
10905
  contextWindow: 1048576,
10383
10906
  maxTokens: 131072,
10384
10907
  },
10385
- "omen-alpha": {
10386
- id: "omen-alpha",
10387
- name: "Omen Alpha",
10388
- api: "openai-completions",
10389
- provider: "opencode-go",
10390
- baseUrl: "https://opencode.ai/zen/go/v1",
10391
- reasoning: true,
10392
- input: ["text", "image"],
10393
- cost: {
10394
- input: 0.2,
10395
- output: 0.66,
10396
- cacheRead: 0.04,
10397
- cacheWrite: 0,
10398
- },
10399
- contextWindow: 500000,
10400
- maxTokens: 128000,
10401
- },
10402
10908
  "qwen3.6-plus": {
10403
10909
  id: "qwen3.6-plus",
10404
10910
  name: "Qwen3.6 Plus",
@@ -11282,13 +11788,13 @@ export const MODELS = {
11282
11788
  reasoning: false,
11283
11789
  input: ["text"],
11284
11790
  cost: {
11285
- input: 0.32,
11286
- output: 0.8899999999999999,
11791
+ input: 0.2574,
11792
+ output: 1.0287,
11287
11793
  cacheRead: 0,
11288
11794
  cacheWrite: 0,
11289
11795
  },
11290
11796
  contextWindow: 163840,
11291
- maxTokens: 16384,
11797
+ maxTokens: 16000,
11292
11798
  },
11293
11799
  "deepseek/deepseek-chat-v3-0324": {
11294
11800
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11458,9 +11964,9 @@ export const MODELS = {
11458
11964
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11459
11965
  input: ["text"],
11460
11966
  cost: {
11461
- input: 0.14,
11462
- output: 0.28,
11463
- cacheRead: 0.03,
11967
+ input: 0.11,
11968
+ output: 0.33,
11969
+ cacheRead: 0.0035,
11464
11970
  cacheWrite: 0,
11465
11971
  },
11466
11972
  contextWindow: 1048576,
@@ -11477,13 +11983,32 @@ export const MODELS = {
11477
11983
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11478
11984
  input: ["text", "image"],
11479
11985
  cost: {
11480
- input: 0.44,
11481
- output: 1.32,
11482
- cacheRead: 0.014,
11986
+ input: 0.22,
11987
+ output: 0.66,
11988
+ cacheRead: 0.007,
11483
11989
  cacheWrite: 0,
11484
11990
  },
11485
11991
  contextWindow: 1048576,
11486
- maxTokens: 384000,
11992
+ maxTokens: 943718,
11993
+ },
11994
+ "deepseek/deepseek-v4-flash-vision-exp:batch": {
11995
+ id: "deepseek/deepseek-v4-flash-vision-exp:batch",
11996
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp (batch)",
11997
+ api: "openai-completions",
11998
+ provider: "openrouter",
11999
+ baseUrl: "https://openrouter.ai/api/v1",
12000
+ compat: { "requiresReasoningContentOnAssistantMessages": true },
12001
+ reasoning: true,
12002
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
12003
+ input: ["text", "image"],
12004
+ cost: {
12005
+ input: 0.11,
12006
+ output: 0.33,
12007
+ cacheRead: 0.0035,
12008
+ cacheWrite: 0,
12009
+ },
12010
+ contextWindow: 1048576,
12011
+ maxTokens: 943718,
11487
12012
  },
11488
12013
  "deepseek/deepseek-v4-pro": {
11489
12014
  id: "deepseek/deepseek-v4-pro",
@@ -11534,9 +12059,9 @@ export const MODELS = {
11534
12059
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
11535
12060
  input: ["text"],
11536
12061
  cost: {
11537
- input: 1.32,
11538
- output: 3.9600000000000004,
11539
- cacheRead: 0.13,
12062
+ input: 0.66,
12063
+ output: 1.9800000000000002,
12064
+ cacheRead: 0.022,
11540
12065
  cacheWrite: 0,
11541
12066
  },
11542
12067
  contextWindow: 1048576,
@@ -12144,11 +12669,11 @@ export const MODELS = {
12144
12669
  provider: "openrouter",
12145
12670
  baseUrl: "https://openrouter.ai/api/v1",
12146
12671
  reasoning: true,
12147
- input: ["text"],
12148
- cost: {
12149
- input: 0.09999999999999999,
12150
- output: 0.15,
12151
- cacheRead: 0.049999999999999996,
12672
+ input: ["text"],
12673
+ cost: {
12674
+ input: 0.06,
12675
+ output: 0.25,
12676
+ cacheRead: 0.015,
12152
12677
  cacheWrite: 0,
12153
12678
  },
12154
12679
  contextWindow: 131072,
@@ -12172,9 +12697,9 @@ export const MODELS = {
12172
12697
  contextWindow: 128000,
12173
12698
  maxTokens: 50000,
12174
12699
  },
12175
- "inception/mercury-2.5-preview": {
12176
- id: "inception/mercury-2.5-preview",
12177
- name: "Inception: Mercury 2.5 Preview",
12700
+ "inception/mercury-2.5": {
12701
+ id: "inception/mercury-2.5",
12702
+ name: "Inception: Mercury 2.5",
12178
12703
  api: "openai-completions",
12179
12704
  provider: "openrouter",
12180
12705
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12437,9 +12962,9 @@ export const MODELS = {
12437
12962
  reasoning: true,
12438
12963
  input: ["text", "image"],
12439
12964
  cost: {
12440
- input: 0.35,
12441
- output: 1.5,
12442
- cacheRead: 0.04,
12965
+ input: 0.175,
12966
+ output: 0.75,
12967
+ cacheRead: 0.02,
12443
12968
  cacheWrite: 0,
12444
12969
  },
12445
12970
  contextWindow: 131072,
@@ -12539,7 +13064,7 @@ export const MODELS = {
12539
13064
  reasoning: true,
12540
13065
  input: ["text"],
12541
13066
  cost: {
12542
- input: 0.39999999999999997,
13067
+ input: 0.55,
12543
13068
  output: 2.2,
12544
13069
  cacheRead: 0,
12545
13070
  cacheWrite: 0,
@@ -12590,13 +13115,13 @@ export const MODELS = {
12590
13115
  reasoning: true,
12591
13116
  input: ["text"],
12592
13117
  cost: {
12593
- input: 0.27,
12594
- output: 1.08,
12595
- cacheRead: 0.027,
13118
+ input: 0.3,
13119
+ output: 1.2,
13120
+ cacheRead: 0.03,
12596
13121
  cacheWrite: 0,
12597
13122
  },
12598
13123
  contextWindow: 204800,
12599
- maxTokens: 128000,
13124
+ maxTokens: 131072,
12600
13125
  },
12601
13126
  "minimax/minimax-m2.7": {
12602
13127
  id: "minimax/minimax-m2.7",
@@ -12666,6 +13191,23 @@ export const MODELS = {
12666
13191
  contextWindow: 256000,
12667
13192
  maxTokens: 204800,
12668
13193
  },
13194
+ "mistralai/codestral-2508:batch": {
13195
+ id: "mistralai/codestral-2508:batch",
13196
+ name: "Mistral: Codestral 2508 (batch)",
13197
+ api: "openai-completions",
13198
+ provider: "openrouter",
13199
+ baseUrl: "https://openrouter.ai/api/v1",
13200
+ reasoning: false,
13201
+ input: ["text"],
13202
+ cost: {
13203
+ input: 0.15,
13204
+ output: 0.44999999999999996,
13205
+ cacheRead: 0.015,
13206
+ cacheWrite: 0,
13207
+ },
13208
+ contextWindow: 256000,
13209
+ maxTokens: 204800,
13210
+ },
12669
13211
  "mistralai/devstral-2512": {
12670
13212
  id: "mistralai/devstral-2512",
12671
13213
  name: "Mistral: Devstral 2 2512",
@@ -12734,6 +13276,23 @@ export const MODELS = {
12734
13276
  contextWindow: 262144,
12735
13277
  maxTokens: 209715,
12736
13278
  },
13279
+ "mistralai/ministral-8b-2512:batch": {
13280
+ id: "mistralai/ministral-8b-2512:batch",
13281
+ name: "Mistral: Ministral 3 8B 2512 (batch)",
13282
+ api: "openai-completions",
13283
+ provider: "openrouter",
13284
+ baseUrl: "https://openrouter.ai/api/v1",
13285
+ reasoning: false,
13286
+ input: ["text", "image"],
13287
+ cost: {
13288
+ input: 0.075,
13289
+ output: 0.075,
13290
+ cacheRead: 0.0075,
13291
+ cacheWrite: 0,
13292
+ },
13293
+ contextWindow: 262144,
13294
+ maxTokens: 209715,
13295
+ },
12737
13296
  "mistralai/mistral-large": {
12738
13297
  id: "mistralai/mistral-large",
12739
13298
  name: "Mistral Large",
@@ -12785,6 +13344,23 @@ export const MODELS = {
12785
13344
  contextWindow: 262144,
12786
13345
  maxTokens: 209715,
12787
13346
  },
13347
+ "mistralai/mistral-large-2512:batch": {
13348
+ id: "mistralai/mistral-large-2512:batch",
13349
+ name: "Mistral: Mistral Large 3 2512 (batch)",
13350
+ api: "openai-completions",
13351
+ provider: "openrouter",
13352
+ baseUrl: "https://openrouter.ai/api/v1",
13353
+ reasoning: false,
13354
+ input: ["text", "image"],
13355
+ cost: {
13356
+ input: 0.25,
13357
+ output: 0.75,
13358
+ cacheRead: 0.024999999999999998,
13359
+ cacheWrite: 0,
13360
+ },
13361
+ contextWindow: 262144,
13362
+ maxTokens: 209715,
13363
+ },
12788
13364
  "mistralai/mistral-medium-3": {
12789
13365
  id: "mistralai/mistral-medium-3",
12790
13366
  name: "Mistral: Mistral Medium 3",
@@ -12853,6 +13429,23 @@ export const MODELS = {
12853
13429
  contextWindow: 131072,
12854
13430
  maxTokens: 104857,
12855
13431
  },
13432
+ "mistralai/mistral-medium-3.1:batch": {
13433
+ id: "mistralai/mistral-medium-3.1:batch",
13434
+ name: "Mistral: Mistral Medium 3.1 (batch)",
13435
+ api: "openai-completions",
13436
+ provider: "openrouter",
13437
+ baseUrl: "https://openrouter.ai/api/v1",
13438
+ reasoning: false,
13439
+ input: ["text", "image"],
13440
+ cost: {
13441
+ input: 0.19999999999999998,
13442
+ output: 1,
13443
+ cacheRead: 0.02,
13444
+ cacheWrite: 0,
13445
+ },
13446
+ contextWindow: 131072,
13447
+ maxTokens: 104857,
13448
+ },
12856
13449
  "mistralai/mistral-nemo": {
12857
13450
  id: "mistralai/mistral-nemo",
12858
13451
  name: "Mistral: Mistral Nemo",
@@ -12904,6 +13497,23 @@ export const MODELS = {
12904
13497
  contextWindow: 262144,
12905
13498
  maxTokens: 209715,
12906
13499
  },
13500
+ "mistralai/mistral-small-2603:batch": {
13501
+ id: "mistralai/mistral-small-2603:batch",
13502
+ name: "Mistral: Mistral Small 4 (batch)",
13503
+ api: "openai-completions",
13504
+ provider: "openrouter",
13505
+ baseUrl: "https://openrouter.ai/api/v1",
13506
+ reasoning: true,
13507
+ input: ["text", "image"],
13508
+ cost: {
13509
+ input: 0.075,
13510
+ output: 0.3,
13511
+ cacheRead: 0.0075,
13512
+ cacheWrite: 0,
13513
+ },
13514
+ contextWindow: 262144,
13515
+ maxTokens: 209715,
13516
+ },
12907
13517
  "mistralai/mistral-small-3.2-24b-instruct": {
12908
13518
  id: "mistralai/mistral-small-3.2-24b-instruct",
12909
13519
  name: "Mistral: Mistral Small 3.2 24B",
@@ -13000,11 +13610,11 @@ export const MODELS = {
13000
13610
  cost: {
13001
13611
  input: 0.6,
13002
13612
  output: 2.5,
13003
- cacheRead: 0,
13613
+ cacheRead: 0.15,
13004
13614
  cacheWrite: 0,
13005
13615
  },
13006
13616
  contextWindow: 262144,
13007
- maxTokens: 235929,
13617
+ maxTokens: 100352,
13008
13618
  },
13009
13619
  "moonshotai/kimi-k2.5": {
13010
13620
  id: "moonshotai/kimi-k2.5",
@@ -13092,35 +13702,35 @@ export const MODELS = {
13092
13702
  contextWindow: 1048576,
13093
13703
  maxTokens: 943718,
13094
13704
  },
13095
- "nex-agi/nex-n2-mini": {
13096
- id: "nex-agi/nex-n2-mini",
13097
- name: "Nex AGI: Nex-N2-Mini",
13705
+ "nex-agi/nex-n2.5-mini:free": {
13706
+ id: "nex-agi/nex-n2.5-mini:free",
13707
+ name: "Nex AGI: Nex-N2.5-Mini (free)",
13098
13708
  api: "openai-completions",
13099
13709
  provider: "openrouter",
13100
13710
  baseUrl: "https://openrouter.ai/api/v1",
13101
13711
  reasoning: true,
13102
13712
  input: ["text", "image"],
13103
13713
  cost: {
13104
- input: 0.024999999999999998,
13105
- output: 0.09999999999999999,
13106
- cacheRead: 0.0025,
13714
+ input: 0,
13715
+ output: 0,
13716
+ cacheRead: 0,
13107
13717
  cacheWrite: 0,
13108
13718
  },
13109
13719
  contextWindow: 262144,
13110
13720
  maxTokens: 235929,
13111
13721
  },
13112
- "nex-agi/nex-n2-pro": {
13113
- id: "nex-agi/nex-n2-pro",
13114
- name: "Nex AGI: Nex-N2-Pro",
13722
+ "nex-agi/nex-n2.5-pro:free": {
13723
+ id: "nex-agi/nex-n2.5-pro:free",
13724
+ name: "Nex AGI: Nex-N2.5-Pro (free)",
13115
13725
  api: "openai-completions",
13116
13726
  provider: "openrouter",
13117
13727
  baseUrl: "https://openrouter.ai/api/v1",
13118
13728
  reasoning: true,
13119
13729
  input: ["text", "image"],
13120
13730
  cost: {
13121
- input: 0.25,
13122
- output: 1,
13123
- cacheRead: 0.024999999999999998,
13731
+ input: 0,
13732
+ output: 0,
13733
+ cacheRead: 0,
13124
13734
  cacheWrite: 0,
13125
13735
  },
13126
13736
  contextWindow: 262144,
@@ -13174,7 +13784,7 @@ export const MODELS = {
13174
13784
  cacheRead: 0,
13175
13785
  cacheWrite: 0,
13176
13786
  },
13177
- contextWindow: 1000000,
13787
+ contextWindow: 262144,
13178
13788
  maxTokens: 16384,
13179
13789
  },
13180
13790
  "nvidia/nemotron-3-super-120b-a12b:free": {
@@ -14422,6 +15032,7 @@ export const MODELS = {
14422
15032
  provider: "openrouter",
14423
15033
  baseUrl: "https://openrouter.ai/api/v1",
14424
15034
  reasoning: true,
15035
+ thinkingLevelMap: { "xhigh": "xhigh" },
14425
15036
  input: ["text", "image"],
14426
15037
  cost: {
14427
15038
  input: 10,
@@ -14439,6 +15050,7 @@ export const MODELS = {
14439
15050
  provider: "openrouter",
14440
15051
  baseUrl: "https://openrouter.ai/api/v1",
14441
15052
  reasoning: true,
15053
+ thinkingLevelMap: { "xhigh": "xhigh" },
14442
15054
  input: ["text", "image"],
14443
15055
  cost: {
14444
15056
  input: 10,
@@ -14456,6 +15068,7 @@ export const MODELS = {
14456
15068
  provider: "openrouter",
14457
15069
  baseUrl: "https://openrouter.ai/api/v1",
14458
15070
  reasoning: true,
15071
+ thinkingLevelMap: { "xhigh": "xhigh" },
14459
15072
  input: ["text", "image"],
14460
15073
  cost: {
14461
15074
  input: 5,
@@ -14473,6 +15086,7 @@ export const MODELS = {
14473
15086
  provider: "openrouter",
14474
15087
  baseUrl: "https://openrouter.ai/api/v1",
14475
15088
  reasoning: true,
15089
+ thinkingLevelMap: { "xhigh": "xhigh" },
14476
15090
  input: ["text", "image"],
14477
15091
  cost: {
14478
15092
  input: 5,
@@ -15002,8 +15616,8 @@ export const MODELS = {
15002
15616
  reasoning: false,
15003
15617
  input: ["text"],
15004
15618
  cost: {
15005
- input: 0.09,
15006
- output: 0.55,
15619
+ input: 0.22,
15620
+ output: 0.88,
15007
15621
  cacheRead: 0,
15008
15622
  cacheWrite: 0,
15009
15623
  },
@@ -15053,13 +15667,13 @@ export const MODELS = {
15053
15667
  reasoning: false,
15054
15668
  input: ["text"],
15055
15669
  cost: {
15056
- input: 0.04815,
15057
- output: 0.19305,
15670
+ input: 0.09,
15671
+ output: 0.3,
15058
15672
  cacheRead: 0,
15059
15673
  cacheWrite: 0,
15060
15674
  },
15061
15675
  contextWindow: 262144,
15062
- maxTokens: 32000,
15676
+ maxTokens: 235929,
15063
15677
  },
15064
15678
  "qwen/qwen3-30b-a3b-thinking-2507": {
15065
15679
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15240,13 +15854,13 @@ export const MODELS = {
15240
15854
  reasoning: false,
15241
15855
  input: ["text"],
15242
15856
  cost: {
15243
- input: 0.09999999999999999,
15857
+ input: 0.09,
15244
15858
  output: 1.1,
15245
- cacheRead: 0.07,
15859
+ cacheRead: 0,
15246
15860
  cacheWrite: 0,
15247
15861
  },
15248
15862
  contextWindow: 262144,
15249
- maxTokens: 235929,
15863
+ maxTokens: 16384,
15250
15864
  },
15251
15865
  "qwen/qwen3-next-80b-a3b-thinking": {
15252
15866
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -15393,13 +16007,13 @@ export const MODELS = {
15393
16007
  reasoning: true,
15394
16008
  input: ["text", "image"],
15395
16009
  cost: {
15396
- input: 0.29,
15397
- output: 2.4,
16010
+ input: 0.26,
16011
+ output: 2.08,
15398
16012
  cacheRead: 0,
15399
16013
  cacheWrite: 0,
15400
16014
  },
15401
16015
  contextWindow: 262144,
15402
- maxTokens: 81920,
16016
+ maxTokens: 65536,
15403
16017
  },
15404
16018
  "qwen/qwen3.5-27b": {
15405
16019
  id: "qwen/qwen3.5-27b",
@@ -15444,13 +16058,13 @@ export const MODELS = {
15444
16058
  reasoning: true,
15445
16059
  input: ["text", "image"],
15446
16060
  cost: {
15447
- input: 0.39,
15448
- output: 2.34,
15449
- cacheRead: 0,
16061
+ input: 0.55,
16062
+ output: 3.5,
16063
+ cacheRead: 0.22499999999999998,
15450
16064
  cacheWrite: 0,
15451
16065
  },
15452
16066
  contextWindow: 262144,
15453
- maxTokens: 65536,
16067
+ maxTokens: 235929,
15454
16068
  },
15455
16069
  "qwen/qwen3.5-9b": {
15456
16070
  id: "qwen/qwen3.5-9b",
@@ -15688,7 +16302,7 @@ export const MODELS = {
15688
16302
  cacheWrite: 0,
15689
16303
  },
15690
16304
  contextWindow: 1048576,
15691
- maxTokens: 262144,
16305
+ maxTokens: 131072,
15692
16306
  },
15693
16307
  "qwen/qwen3.8-2.4t-a95b:batch": {
15694
16308
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15960,7 +16574,7 @@ export const MODELS = {
15960
16574
  cacheWrite: 0,
15961
16575
  },
15962
16576
  contextWindow: 1048576,
15963
- maxTokens: 471859,
16577
+ maxTokens: 32768,
15964
16578
  },
15965
16579
  "thinkingmachines/inkling-small": {
15966
16580
  id: "thinkingmachines/inkling-small",
@@ -16277,13 +16891,13 @@ export const MODELS = {
16277
16891
  reasoning: true,
16278
16892
  input: ["text"],
16279
16893
  cost: {
16280
- input: 0.55,
16281
- output: 2.2,
16282
- cacheRead: 0.11,
16894
+ input: 0.43,
16895
+ output: 1.75,
16896
+ cacheRead: 0.08,
16283
16897
  cacheWrite: 0,
16284
16898
  },
16285
16899
  contextWindow: 204800,
16286
- maxTokens: 131072,
16900
+ maxTokens: 16384,
16287
16901
  },
16288
16902
  "z-ai/glm-4.6v": {
16289
16903
  id: "z-ai/glm-4.6v",
@@ -16404,6 +17018,23 @@ export const MODELS = {
16404
17018
  contextWindow: 1048576,
16405
17019
  maxTokens: 131072,
16406
17020
  },
17021
+ "z-ai/glm-5.2:batch": {
17022
+ id: "z-ai/glm-5.2:batch",
17023
+ name: "Z.ai: GLM 5.2 (batch)",
17024
+ api: "openai-completions",
17025
+ provider: "openrouter",
17026
+ baseUrl: "https://openrouter.ai/api/v1",
17027
+ reasoning: true,
17028
+ input: ["text"],
17029
+ cost: {
17030
+ input: 0.7,
17031
+ output: 2.2,
17032
+ cacheRead: 0.07,
17033
+ cacheWrite: 0,
17034
+ },
17035
+ contextWindow: 1048576,
17036
+ maxTokens: 943718,
17037
+ },
16407
17038
  "z-ai/glm-5.3": {
16408
17039
  id: "z-ai/glm-5.3",
16409
17040
  name: "Z.ai: GLM 5.3",
@@ -16447,13 +17078,30 @@ export const MODELS = {
16447
17078
  reasoning: true,
16448
17079
  input: ["text", "image"],
16449
17080
  cost: {
16450
- input: 0.15,
16451
- output: 0.5,
16452
- cacheRead: 0.03,
17081
+ input: 0.075,
17082
+ output: 0.25,
17083
+ cacheRead: 0.015,
16453
17084
  cacheWrite: 0,
16454
17085
  },
16455
- contextWindow: 1048575,
16456
- maxTokens: 943717,
17086
+ contextWindow: 1048576,
17087
+ maxTokens: 943718,
17088
+ },
17089
+ "z-ai/glm-5.3:batch": {
17090
+ id: "z-ai/glm-5.3:batch",
17091
+ name: "Z.ai: GLM 5.3 (batch)",
17092
+ api: "openai-completions",
17093
+ provider: "openrouter",
17094
+ baseUrl: "https://openrouter.ai/api/v1",
17095
+ reasoning: true,
17096
+ input: ["text"],
17097
+ cost: {
17098
+ input: 0.7,
17099
+ output: 2.2,
17100
+ cacheRead: 0.13,
17101
+ cacheWrite: 0,
17102
+ },
17103
+ contextWindow: 1048576,
17104
+ maxTokens: 943718,
16457
17105
  },
16458
17106
  "z-ai/glm-5v-turbo": {
16459
17107
  id: "z-ai/glm-5v-turbo",
@@ -16602,9 +17250,9 @@ export const MODELS = {
16602
17250
  reasoning: true,
16603
17251
  input: ["text", "image"],
16604
17252
  cost: {
16605
- input: 2.5,
16606
- output: 14,
16607
- cacheRead: 0.29,
17253
+ input: 2.4,
17254
+ output: 12,
17255
+ cacheRead: 0.24,
16608
17256
  cacheWrite: 0,
16609
17257
  },
16610
17258
  contextWindow: 1048576,
@@ -16670,9 +17318,9 @@ export const MODELS = {
16670
17318
  reasoning: true,
16671
17319
  input: ["text", "image"],
16672
17320
  cost: {
16673
- input: 0.07125000000000001,
16674
- output: 0.2375,
16675
- cacheRead: 0.01425,
17321
+ input: 0.075,
17322
+ output: 0.25,
17323
+ cacheRead: 0.015,
16676
17324
  cacheWrite: 0,
16677
17325
  },
16678
17326
  contextWindow: 1310720,
@@ -16687,13 +17335,13 @@ export const MODELS = {
16687
17335
  reasoning: true,
16688
17336
  input: ["text"],
16689
17337
  cost: {
16690
- input: 1.12,
16691
- output: 3.52,
16692
- cacheRead: 0.20800000000000002,
17338
+ input: 1.092,
17339
+ output: 3.432,
17340
+ cacheRead: 0.20279999999999998,
16693
17341
  cacheWrite: 0,
16694
17342
  },
16695
17343
  contextWindow: 1310720,
16696
- maxTokens: 943718,
17344
+ maxTokens: 131072,
16697
17345
  },
16698
17346
  },
16699
17347
  "together": {
@@ -18097,9 +18745,9 @@ export const MODELS = {
18097
18745
  reasoning: false,
18098
18746
  input: ["text"],
18099
18747
  cost: {
18100
- input: 0.28,
18101
- output: 0.42,
18102
- cacheRead: 0.028,
18748
+ input: 0.62,
18749
+ output: 1.85,
18750
+ cacheRead: 0,
18103
18751
  cacheWrite: 0,
18104
18752
  },
18105
18753
  contextWindow: 128000,
@@ -18207,6 +18855,23 @@ export const MODELS = {
18207
18855
  contextWindow: 1000000,
18208
18856
  maxTokens: 384000,
18209
18857
  },
18858
+ "deepseek/deepseek-v4.1-flash": {
18859
+ id: "deepseek/deepseek-v4.1-flash",
18860
+ name: "DeepSeek V4.1 Flash",
18861
+ api: "anthropic-messages",
18862
+ provider: "vercel-ai-gateway",
18863
+ baseUrl: "https://ai-gateway.vercel.sh",
18864
+ reasoning: true,
18865
+ input: ["text", "image"],
18866
+ cost: {
18867
+ input: 0.15,
18868
+ output: 0.6,
18869
+ cacheRead: 0.003,
18870
+ cacheWrite: 0,
18871
+ },
18872
+ contextWindow: 1000000,
18873
+ maxTokens: 384000,
18874
+ },
18210
18875
  "google/gemini-2.5-flash": {
18211
18876
  id: "google/gemini-2.5-flash",
18212
18877
  name: "Gemini 2.5 Flash",
@@ -18445,6 +19110,23 @@ export const MODELS = {
18445
19110
  contextWindow: 128000,
18446
19111
  maxTokens: 128000,
18447
19112
  },
19113
+ "inception/mercury-2.5": {
19114
+ id: "inception/mercury-2.5",
19115
+ name: "Mercury 2.5",
19116
+ api: "anthropic-messages",
19117
+ provider: "vercel-ai-gateway",
19118
+ baseUrl: "https://ai-gateway.vercel.sh",
19119
+ reasoning: true,
19120
+ input: ["text"],
19121
+ cost: {
19122
+ input: 0.04,
19123
+ output: 0.15,
19124
+ cacheRead: 0.004,
19125
+ cacheWrite: 0,
19126
+ },
19127
+ contextWindow: 260000,
19128
+ maxTokens: 65536,
19129
+ },
18448
19130
  "inception/mercury-coder-small": {
18449
19131
  id: "inception/mercury-coder-small",
18450
19132
  name: "Mercury Coder Small Beta",
@@ -20190,6 +20872,7 @@ export const MODELS = {
20190
20872
  provider: "vercel-ai-gateway",
20191
20873
  baseUrl: "https://ai-gateway.vercel.sh",
20192
20874
  reasoning: true,
20875
+ thinkingLevelMap: { "xhigh": "xhigh" },
20193
20876
  input: ["text", "image"],
20194
20877
  cost: {
20195
20878
  input: 10,
@@ -20207,6 +20890,7 @@ export const MODELS = {
20207
20890
  provider: "vercel-ai-gateway",
20208
20891
  baseUrl: "https://ai-gateway.vercel.sh",
20209
20892
  reasoning: true,
20893
+ thinkingLevelMap: { "xhigh": "xhigh" },
20210
20894
  input: ["text", "image"],
20211
20895
  cost: {
20212
20896
  input: 20,
@@ -20812,23 +21496,6 @@ export const MODELS = {
20812
21496
  contextWindow: 1050000,
20813
21497
  maxTokens: 131000,
20814
21498
  },
20815
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
20816
- id: "xiaomi/mimo-v2.5-pro-ultraspeed",
20817
- name: "MiMo V2.5 Pro UltraSpeed",
20818
- api: "anthropic-messages",
20819
- provider: "vercel-ai-gateway",
20820
- baseUrl: "https://ai-gateway.vercel.sh",
20821
- reasoning: true,
20822
- input: ["text"],
20823
- cost: {
20824
- input: 1.305,
20825
- output: 2.61,
20826
- cacheRead: 0.0108,
20827
- cacheWrite: 0,
20828
- },
20829
- contextWindow: 1048576,
20830
- maxTokens: 131072,
20831
- },
20832
21499
  "zai/glm-4.5": {
20833
21500
  id: "zai/glm-4.5",
20834
21501
  name: "GLM 4.5",
@@ -21042,9 +21709,9 @@ export const MODELS = {
21042
21709
  reasoning: true,
21043
21710
  input: ["text"],
21044
21711
  cost: {
21045
- input: 0.7,
21046
- output: 2.2,
21047
- cacheRead: 0.13,
21712
+ input: 1.4,
21713
+ output: 4.4,
21714
+ cacheRead: 0.14,
21048
21715
  cacheWrite: 0,
21049
21716
  },
21050
21717
  contextWindow: 1000000,
@@ -21084,23 +21751,6 @@ export const MODELS = {
21084
21751
  contextWindow: 1000000,
21085
21752
  maxTokens: 131000,
21086
21753
  },
21087
- "zai/glm-5.3-promo-50": {
21088
- id: "zai/glm-5.3-promo-50",
21089
- name: "GLM 5.3 (50% off)",
21090
- api: "anthropic-messages",
21091
- provider: "vercel-ai-gateway",
21092
- baseUrl: "https://ai-gateway.vercel.sh",
21093
- reasoning: true,
21094
- input: ["text"],
21095
- cost: {
21096
- input: 0.7,
21097
- output: 2.2,
21098
- cacheRead: 0.13,
21099
- cacheWrite: 0,
21100
- },
21101
- contextWindow: 1048576,
21102
- maxTokens: 1048576,
21103
- },
21104
21754
  "zai/glm-5v-turbo": {
21105
21755
  id: "zai/glm-5v-turbo",
21106
21756
  name: "GLM 5V Turbo",