@elyracode/ai 0.9.36 → 0.9.37

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -13,11 +13,11 @@ export const MODELS = {
13
13
  cost: {
14
14
  input: 0.33,
15
15
  output: 2.75,
16
- cacheRead: 0,
17
- cacheWrite: 0,
16
+ cacheRead: 0.0825,
17
+ cacheWrite: 0.33,
18
18
  },
19
- contextWindow: 128000,
20
- maxTokens: 4096,
19
+ contextWindow: 1000000,
20
+ maxTokens: 65535,
21
21
  },
22
22
  "amazon.nova-lite-v1:0": {
23
23
  id: "amazon.nova-lite-v1:0",
@@ -31,10 +31,10 @@ export const MODELS = {
31
31
  input: 0.06,
32
32
  output: 0.24,
33
33
  cacheRead: 0.015,
34
- cacheWrite: 0,
34
+ cacheWrite: 0.06,
35
35
  },
36
36
  contextWindow: 300000,
37
- maxTokens: 8192,
37
+ maxTokens: 10000,
38
38
  },
39
39
  "amazon.nova-micro-v1:0": {
40
40
  id: "amazon.nova-micro-v1:0",
@@ -48,10 +48,10 @@ export const MODELS = {
48
48
  input: 0.035,
49
49
  output: 0.14,
50
50
  cacheRead: 0.00875,
51
- cacheWrite: 0,
51
+ cacheWrite: 0.035,
52
52
  },
53
53
  contextWindow: 128000,
54
- maxTokens: 8192,
54
+ maxTokens: 10000,
55
55
  },
56
56
  "amazon.nova-pro-v1:0": {
57
57
  id: "amazon.nova-pro-v1:0",
@@ -65,10 +65,10 @@ export const MODELS = {
65
65
  input: 0.8,
66
66
  output: 3.2,
67
67
  cacheRead: 0.2,
68
- cacheWrite: 0,
68
+ cacheWrite: 0.8,
69
69
  },
70
70
  contextWindow: 300000,
71
- maxTokens: 8192,
71
+ maxTokens: 10000,
72
72
  },
73
73
  "anthropic.claude-fable-5": {
74
74
  id: "anthropic.claude-fable-5",
@@ -291,6 +291,57 @@ export const MODELS = {
291
291
  contextWindow: 1000000,
292
292
  maxTokens: 128000,
293
293
  },
294
+ "apac.amazon.nova-lite-v1:0": {
295
+ id: "apac.amazon.nova-lite-v1:0",
296
+ name: "Nova Lite (APAC)",
297
+ api: "bedrock-converse-stream",
298
+ provider: "amazon-bedrock",
299
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
300
+ reasoning: false,
301
+ input: ["text", "image"],
302
+ cost: {
303
+ input: 0.063,
304
+ output: 0.252,
305
+ cacheRead: 0.01575,
306
+ cacheWrite: 0.063,
307
+ },
308
+ contextWindow: 300000,
309
+ maxTokens: 10000,
310
+ },
311
+ "apac.amazon.nova-micro-v1:0": {
312
+ id: "apac.amazon.nova-micro-v1:0",
313
+ name: "Nova Micro (APAC)",
314
+ api: "bedrock-converse-stream",
315
+ provider: "amazon-bedrock",
316
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
317
+ reasoning: false,
318
+ input: ["text"],
319
+ cost: {
320
+ input: 0.037,
321
+ output: 0.148,
322
+ cacheRead: 0.00925,
323
+ cacheWrite: 0.037,
324
+ },
325
+ contextWindow: 128000,
326
+ maxTokens: 10000,
327
+ },
328
+ "apac.amazon.nova-pro-v1:0": {
329
+ id: "apac.amazon.nova-pro-v1:0",
330
+ name: "Nova Pro (APAC)",
331
+ api: "bedrock-converse-stream",
332
+ provider: "amazon-bedrock",
333
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
334
+ reasoning: false,
335
+ input: ["text", "image"],
336
+ cost: {
337
+ input: 0.84,
338
+ output: 3.36,
339
+ cacheRead: 0.21,
340
+ cacheWrite: 0.84,
341
+ },
342
+ contextWindow: 300000,
343
+ maxTokens: 10000,
344
+ },
294
345
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {
295
346
  id: "au.anthropic.claude-haiku-4-5-20251001-v1:0",
296
347
  name: "Claude Haiku 4.5 (AU)",
@@ -328,6 +379,25 @@ export const MODELS = {
328
379
  contextWindow: 1000000,
329
380
  maxTokens: 128000,
330
381
  },
382
+ "au.anthropic.claude-opus-4-7": {
383
+ id: "au.anthropic.claude-opus-4-7",
384
+ name: "Claude Opus 4.7 (AU)",
385
+ api: "bedrock-converse-stream",
386
+ provider: "amazon-bedrock",
387
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
388
+ reasoning: true,
389
+ thinkingType: "adaptive",
390
+ thinkingLevelMap: { "xhigh": "xhigh" },
391
+ input: ["text", "image"],
392
+ cost: {
393
+ input: 5.5,
394
+ output: 27.5,
395
+ cacheRead: 0.55,
396
+ cacheWrite: 6.875,
397
+ },
398
+ contextWindow: 1000000,
399
+ maxTokens: 128000,
400
+ },
331
401
  "au.anthropic.claude-opus-4-8": {
332
402
  id: "au.anthropic.claude-opus-4-8",
333
403
  name: "Claude Opus 4.8 (AU)",
@@ -419,6 +489,23 @@ export const MODELS = {
419
489
  contextWindow: 1000000,
420
490
  maxTokens: 128000,
421
491
  },
492
+ "ca.amazon.nova-lite-v1:0": {
493
+ id: "ca.amazon.nova-lite-v1:0",
494
+ name: "Nova Lite (CA)",
495
+ api: "bedrock-converse-stream",
496
+ provider: "amazon-bedrock",
497
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
498
+ reasoning: false,
499
+ input: ["text", "image"],
500
+ cost: {
501
+ input: 0.064,
502
+ output: 0.256,
503
+ cacheRead: 0.016,
504
+ cacheWrite: 0.064,
505
+ },
506
+ contextWindow: 300000,
507
+ maxTokens: 10000,
508
+ },
422
509
  "deepseek.r1-v1:0": {
423
510
  id: "deepseek.r1-v1:0",
424
511
  name: "DeepSeek-R1",
@@ -470,6 +557,74 @@ export const MODELS = {
470
557
  contextWindow: 163840,
471
558
  maxTokens: 81920,
472
559
  },
560
+ "eu.amazon.nova-2-lite-v1:0": {
561
+ id: "eu.amazon.nova-2-lite-v1:0",
562
+ name: "Nova 2 Lite (EU)",
563
+ api: "bedrock-converse-stream",
564
+ provider: "amazon-bedrock",
565
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
566
+ reasoning: true,
567
+ input: ["text", "image"],
568
+ cost: {
569
+ input: 0.374,
570
+ output: 3.157,
571
+ cacheRead: 0.0935,
572
+ cacheWrite: 0.374,
573
+ },
574
+ contextWindow: 1000000,
575
+ maxTokens: 65535,
576
+ },
577
+ "eu.amazon.nova-lite-v1:0": {
578
+ id: "eu.amazon.nova-lite-v1:0",
579
+ name: "Nova Lite (EU)",
580
+ api: "bedrock-converse-stream",
581
+ provider: "amazon-bedrock",
582
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
583
+ reasoning: false,
584
+ input: ["text", "image"],
585
+ cost: {
586
+ input: 0.069,
587
+ output: 0.276,
588
+ cacheRead: 0.01725,
589
+ cacheWrite: 0.069,
590
+ },
591
+ contextWindow: 300000,
592
+ maxTokens: 10000,
593
+ },
594
+ "eu.amazon.nova-micro-v1:0": {
595
+ id: "eu.amazon.nova-micro-v1:0",
596
+ name: "Nova Micro (EU)",
597
+ api: "bedrock-converse-stream",
598
+ provider: "amazon-bedrock",
599
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
600
+ reasoning: false,
601
+ input: ["text"],
602
+ cost: {
603
+ input: 0.04,
604
+ output: 0.16,
605
+ cacheRead: 0.01,
606
+ cacheWrite: 0.04,
607
+ },
608
+ contextWindow: 128000,
609
+ maxTokens: 10000,
610
+ },
611
+ "eu.amazon.nova-pro-v1:0": {
612
+ id: "eu.amazon.nova-pro-v1:0",
613
+ name: "Nova Pro (EU)",
614
+ api: "bedrock-converse-stream",
615
+ provider: "amazon-bedrock",
616
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
617
+ reasoning: false,
618
+ input: ["text", "image"],
619
+ cost: {
620
+ input: 0.92,
621
+ output: 3.68,
622
+ cacheRead: 0.23,
623
+ cacheWrite: 0.92,
624
+ },
625
+ contextWindow: 300000,
626
+ maxTokens: 10000,
627
+ },
473
628
  "eu.anthropic.claude-fable-5": {
474
629
  id: "eu.anthropic.claude-fable-5",
475
630
  name: "Claude Fable 5 (EU)",
@@ -654,6 +809,40 @@ export const MODELS = {
654
809
  contextWindow: 1000000,
655
810
  maxTokens: 128000,
656
811
  },
812
+ "eu.mistral.pixtral-large-2502-v1:0": {
813
+ id: "eu.mistral.pixtral-large-2502-v1:0",
814
+ name: "Pixtral Large (25.02) (EU)",
815
+ api: "bedrock-converse-stream",
816
+ provider: "amazon-bedrock",
817
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
818
+ reasoning: false,
819
+ input: ["text", "image"],
820
+ cost: {
821
+ input: 2,
822
+ output: 6,
823
+ cacheRead: 0,
824
+ cacheWrite: 0,
825
+ },
826
+ contextWindow: 128000,
827
+ maxTokens: 8192,
828
+ },
829
+ "global.amazon.nova-2-lite-v1:0": {
830
+ id: "global.amazon.nova-2-lite-v1:0",
831
+ name: "Nova 2 Lite (Global)",
832
+ api: "bedrock-converse-stream",
833
+ provider: "amazon-bedrock",
834
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
835
+ reasoning: true,
836
+ input: ["text", "image"],
837
+ cost: {
838
+ input: 0.3,
839
+ output: 2.5,
840
+ cacheRead: 0.075,
841
+ cacheWrite: 0.3,
842
+ },
843
+ contextWindow: 1000000,
844
+ maxTokens: 65535,
845
+ },
657
846
  "global.anthropic.claude-fable-5": {
658
847
  id: "global.anthropic.claude-fable-5",
659
848
  name: "Claude Fable 5 (Global)",
@@ -908,30 +1097,81 @@ export const MODELS = {
908
1097
  contextWindow: 1050000,
909
1098
  maxTokens: 128000,
910
1099
  },
911
- "google.gemma-3-27b-it": {
912
- id: "google.gemma-3-27b-it",
913
- name: "Google Gemma 3 27B Instruct",
1100
+ "global.openai.gpt-6-astra": {
1101
+ id: "global.openai.gpt-6-astra",
1102
+ name: "GPT-6 Astra (Global)",
914
1103
  api: "bedrock-converse-stream",
915
1104
  provider: "amazon-bedrock",
916
1105
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
917
- reasoning: false,
1106
+ reasoning: true,
918
1107
  input: ["text", "image"],
919
1108
  cost: {
920
- input: 0.12,
921
- output: 0.2,
1109
+ input: 10,
1110
+ output: 50,
1111
+ cacheRead: 1,
1112
+ cacheWrite: 12.5,
1113
+ },
1114
+ contextWindow: 1050000,
1115
+ maxTokens: 128000,
1116
+ },
1117
+ "global.xai.grok-4.6": {
1118
+ id: "global.xai.grok-4.6",
1119
+ name: "Grok 4.6 (Global)",
1120
+ api: "bedrock-converse-stream",
1121
+ provider: "amazon-bedrock",
1122
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1123
+ reasoning: true,
1124
+ input: ["text", "image"],
1125
+ cost: {
1126
+ input: 2,
1127
+ output: 6,
1128
+ cacheRead: 0.5,
1129
+ cacheWrite: 0,
1130
+ },
1131
+ contextWindow: 500000,
1132
+ maxTokens: 500000,
1133
+ },
1134
+ "google.gemma-4-26b-a4b": {
1135
+ id: "google.gemma-4-26b-a4b",
1136
+ name: "Gemma 4 26B A4B IT",
1137
+ api: "bedrock-converse-stream",
1138
+ provider: "amazon-bedrock",
1139
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1140
+ reasoning: true,
1141
+ input: ["text", "image"],
1142
+ cost: {
1143
+ input: 0.13,
1144
+ output: 0.4,
922
1145
  cacheRead: 0,
923
1146
  cacheWrite: 0,
924
1147
  },
925
- contextWindow: 202752,
926
- maxTokens: 8192,
1148
+ contextWindow: 262144,
1149
+ maxTokens: 32768,
927
1150
  },
928
- "google.gemma-3-4b-it": {
929
- id: "google.gemma-3-4b-it",
930
- name: "Gemma 3 4B IT",
1151
+ "google.gemma-4-31b": {
1152
+ id: "google.gemma-4-31b",
1153
+ name: "Gemma 4 31B IT",
931
1154
  api: "bedrock-converse-stream",
932
1155
  provider: "amazon-bedrock",
933
1156
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
934
- reasoning: false,
1157
+ reasoning: true,
1158
+ input: ["text", "image"],
1159
+ cost: {
1160
+ input: 0.14,
1161
+ output: 0.4,
1162
+ cacheRead: 0,
1163
+ cacheWrite: 0,
1164
+ },
1165
+ contextWindow: 262144,
1166
+ maxTokens: 32768,
1167
+ },
1168
+ "google.gemma-4-e2b": {
1169
+ id: "google.gemma-4-e2b",
1170
+ name: "Gemma 4 E2B IT",
1171
+ api: "bedrock-converse-stream",
1172
+ provider: "amazon-bedrock",
1173
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1174
+ reasoning: true,
935
1175
  input: ["text", "image"],
936
1176
  cost: {
937
1177
  input: 0.04,
@@ -939,8 +1179,59 @@ export const MODELS = {
939
1179
  cacheRead: 0,
940
1180
  cacheWrite: 0,
941
1181
  },
942
- contextWindow: 128000,
943
- maxTokens: 4096,
1182
+ contextWindow: 131072,
1183
+ maxTokens: 8192,
1184
+ },
1185
+ "in.openai.gpt-5.6-luna": {
1186
+ id: "in.openai.gpt-5.6-luna",
1187
+ name: "GPT-5.6 Luna (India)",
1188
+ api: "bedrock-converse-stream",
1189
+ provider: "amazon-bedrock",
1190
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1191
+ reasoning: true,
1192
+ input: ["text", "image"],
1193
+ cost: {
1194
+ input: 0.22,
1195
+ output: 1.32,
1196
+ cacheRead: 0.022,
1197
+ cacheWrite: 0.275,
1198
+ },
1199
+ contextWindow: 1050000,
1200
+ maxTokens: 128000,
1201
+ },
1202
+ "in.openai.gpt-5.6-terra": {
1203
+ id: "in.openai.gpt-5.6-terra",
1204
+ name: "GPT-5.6 Terra (India)",
1205
+ api: "bedrock-converse-stream",
1206
+ provider: "amazon-bedrock",
1207
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1208
+ reasoning: true,
1209
+ input: ["text", "image"],
1210
+ cost: {
1211
+ input: 2.2,
1212
+ output: 13.2,
1213
+ cacheRead: 0.22,
1214
+ cacheWrite: 2.75,
1215
+ },
1216
+ contextWindow: 1050000,
1217
+ maxTokens: 128000,
1218
+ },
1219
+ "jp.amazon.nova-2-lite-v1:0": {
1220
+ id: "jp.amazon.nova-2-lite-v1:0",
1221
+ name: "Nova 2 Lite (JP)",
1222
+ api: "bedrock-converse-stream",
1223
+ provider: "amazon-bedrock",
1224
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1225
+ reasoning: true,
1226
+ input: ["text", "image"],
1227
+ cost: {
1228
+ input: 0.396,
1229
+ output: 3.311,
1230
+ cacheRead: 0.099,
1231
+ cacheWrite: 0.396,
1232
+ },
1233
+ contextWindow: 1000000,
1234
+ maxTokens: 65535,
944
1235
  },
945
1236
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {
946
1237
  id: "jp.anthropic.claude-haiku-4-5-20251001-v1:0",
@@ -1548,6 +1839,23 @@ export const MODELS = {
1548
1839
  contextWindow: 1050000,
1549
1840
  maxTokens: 128000,
1550
1841
  },
1842
+ "openai.gpt-6-astra": {
1843
+ id: "openai.gpt-6-astra",
1844
+ name: "GPT-6 Astra",
1845
+ api: "bedrock-converse-stream",
1846
+ provider: "amazon-bedrock",
1847
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1848
+ reasoning: true,
1849
+ input: ["text", "image"],
1850
+ cost: {
1851
+ input: 11,
1852
+ output: 55,
1853
+ cacheRead: 1.1,
1854
+ cacheWrite: 13.75,
1855
+ },
1856
+ contextWindow: 1050000,
1857
+ maxTokens: 128000,
1858
+ },
1551
1859
  "openai.gpt-oss-120b": {
1552
1860
  id: "openai.gpt-oss-120b",
1553
1861
  name: "gpt-oss-120b",
@@ -1622,7 +1930,7 @@ export const MODELS = {
1622
1930
  api: "bedrock-converse-stream",
1623
1931
  provider: "amazon-bedrock",
1624
1932
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1625
- reasoning: false,
1933
+ reasoning: true,
1626
1934
  input: ["text"],
1627
1935
  cost: {
1628
1936
  input: 0.15,
@@ -1639,7 +1947,7 @@ export const MODELS = {
1639
1947
  api: "bedrock-converse-stream",
1640
1948
  provider: "amazon-bedrock",
1641
1949
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1642
- reasoning: false,
1950
+ reasoning: true,
1643
1951
  input: ["text"],
1644
1952
  cost: {
1645
1953
  input: 0.07,
@@ -1732,42 +2040,161 @@ export const MODELS = {
1732
2040
  cacheRead: 0,
1733
2041
  cacheWrite: 0,
1734
2042
  },
1735
- contextWindow: 131072,
1736
- maxTokens: 65536,
2043
+ contextWindow: 131072,
2044
+ maxTokens: 65536,
2045
+ },
2046
+ "qwen.qwen3-next-80b-a3b": {
2047
+ id: "qwen.qwen3-next-80b-a3b",
2048
+ name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
2049
+ api: "bedrock-converse-stream",
2050
+ provider: "amazon-bedrock",
2051
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2052
+ reasoning: false,
2053
+ input: ["text"],
2054
+ cost: {
2055
+ input: 0.14,
2056
+ output: 1.4,
2057
+ cacheRead: 0,
2058
+ cacheWrite: 0,
2059
+ },
2060
+ contextWindow: 262000,
2061
+ maxTokens: 262000,
2062
+ },
2063
+ "qwen.qwen3-vl-235b-a22b": {
2064
+ id: "qwen.qwen3-vl-235b-a22b",
2065
+ name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
2066
+ api: "bedrock-converse-stream",
2067
+ provider: "amazon-bedrock",
2068
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2069
+ reasoning: false,
2070
+ input: ["text", "image"],
2071
+ cost: {
2072
+ input: 0.3,
2073
+ output: 1.5,
2074
+ cacheRead: 0,
2075
+ cacheWrite: 0,
2076
+ },
2077
+ contextWindow: 262000,
2078
+ maxTokens: 262000,
2079
+ },
2080
+ "us-gov.openai.gpt-oss-120b-1:0": {
2081
+ id: "us-gov.openai.gpt-oss-120b-1:0",
2082
+ name: "gpt-oss-120b (GovCloud)",
2083
+ api: "bedrock-converse-stream",
2084
+ provider: "amazon-bedrock",
2085
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2086
+ reasoning: true,
2087
+ input: ["text"],
2088
+ cost: {
2089
+ input: 0.18,
2090
+ output: 0.72,
2091
+ cacheRead: 0,
2092
+ cacheWrite: 0,
2093
+ },
2094
+ contextWindow: 128000,
2095
+ maxTokens: 16384,
2096
+ },
2097
+ "us-gov.openai.gpt-oss-20b-1:0": {
2098
+ id: "us-gov.openai.gpt-oss-20b-1:0",
2099
+ name: "gpt-oss-20b (GovCloud)",
2100
+ api: "bedrock-converse-stream",
2101
+ provider: "amazon-bedrock",
2102
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2103
+ reasoning: true,
2104
+ input: ["text"],
2105
+ cost: {
2106
+ input: 0.084,
2107
+ output: 0.36,
2108
+ cacheRead: 0,
2109
+ cacheWrite: 0,
2110
+ },
2111
+ contextWindow: 128000,
2112
+ maxTokens: 16384,
2113
+ },
2114
+ "us.amazon.nova-2-lite-v1:0": {
2115
+ id: "us.amazon.nova-2-lite-v1:0",
2116
+ name: "Nova 2 Lite (US)",
2117
+ api: "bedrock-converse-stream",
2118
+ provider: "amazon-bedrock",
2119
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2120
+ reasoning: true,
2121
+ input: ["text", "image"],
2122
+ cost: {
2123
+ input: 0.33,
2124
+ output: 2.75,
2125
+ cacheRead: 0.0825,
2126
+ cacheWrite: 0.33,
2127
+ },
2128
+ contextWindow: 1000000,
2129
+ maxTokens: 65535,
2130
+ },
2131
+ "us.amazon.nova-lite-v1:0": {
2132
+ id: "us.amazon.nova-lite-v1:0",
2133
+ name: "Nova Lite (US)",
2134
+ api: "bedrock-converse-stream",
2135
+ provider: "amazon-bedrock",
2136
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2137
+ reasoning: false,
2138
+ input: ["text", "image"],
2139
+ cost: {
2140
+ input: 0.06,
2141
+ output: 0.24,
2142
+ cacheRead: 0.015,
2143
+ cacheWrite: 0.06,
2144
+ },
2145
+ contextWindow: 300000,
2146
+ maxTokens: 10000,
2147
+ },
2148
+ "us.amazon.nova-micro-v1:0": {
2149
+ id: "us.amazon.nova-micro-v1:0",
2150
+ name: "Nova Micro (US)",
2151
+ api: "bedrock-converse-stream",
2152
+ provider: "amazon-bedrock",
2153
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2154
+ reasoning: false,
2155
+ input: ["text"],
2156
+ cost: {
2157
+ input: 0.035,
2158
+ output: 0.14,
2159
+ cacheRead: 0.00875,
2160
+ cacheWrite: 0.035,
2161
+ },
2162
+ contextWindow: 128000,
2163
+ maxTokens: 10000,
1737
2164
  },
1738
- "qwen.qwen3-next-80b-a3b": {
1739
- id: "qwen.qwen3-next-80b-a3b",
1740
- name: "Qwen/Qwen3-Next-80B-A3B-Instruct",
2165
+ "us.amazon.nova-premier-v1:0": {
2166
+ id: "us.amazon.nova-premier-v1:0",
2167
+ name: "Nova Premier (US)",
1741
2168
  api: "bedrock-converse-stream",
1742
2169
  provider: "amazon-bedrock",
1743
2170
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1744
2171
  reasoning: false,
1745
- input: ["text"],
2172
+ input: ["text", "image"],
1746
2173
  cost: {
1747
- input: 0.14,
1748
- output: 1.4,
1749
- cacheRead: 0,
1750
- cacheWrite: 0,
2174
+ input: 2.5,
2175
+ output: 12.5,
2176
+ cacheRead: 0.625,
2177
+ cacheWrite: 2.5,
1751
2178
  },
1752
- contextWindow: 262000,
1753
- maxTokens: 262000,
2179
+ contextWindow: 1000000,
2180
+ maxTokens: 10000,
1754
2181
  },
1755
- "qwen.qwen3-vl-235b-a22b": {
1756
- id: "qwen.qwen3-vl-235b-a22b",
1757
- name: "Qwen/Qwen3-VL-235B-A22B-Instruct",
2182
+ "us.amazon.nova-pro-v1:0": {
2183
+ id: "us.amazon.nova-pro-v1:0",
2184
+ name: "Nova Pro (US)",
1758
2185
  api: "bedrock-converse-stream",
1759
2186
  provider: "amazon-bedrock",
1760
2187
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1761
2188
  reasoning: false,
1762
2189
  input: ["text", "image"],
1763
2190
  cost: {
1764
- input: 0.3,
1765
- output: 1.5,
1766
- cacheRead: 0,
1767
- cacheWrite: 0,
2191
+ input: 0.8,
2192
+ output: 3.2,
2193
+ cacheRead: 0.2,
2194
+ cacheWrite: 0.8,
1768
2195
  },
1769
- contextWindow: 262000,
1770
- maxTokens: 262000,
2196
+ contextWindow: 300000,
2197
+ maxTokens: 10000,
1771
2198
  },
1772
2199
  "us.anthropic.claude-fable-5": {
1773
2200
  id: "us.anthropic.claude-fable-5",
@@ -2007,6 +2434,57 @@ export const MODELS = {
2007
2434
  contextWindow: 128000,
2008
2435
  maxTokens: 32768,
2009
2436
  },
2437
+ "us.meta.llama3-1-70b-instruct-v1:0": {
2438
+ id: "us.meta.llama3-1-70b-instruct-v1:0",
2439
+ name: "Llama 3.1 70B Instruct (US)",
2440
+ api: "bedrock-converse-stream",
2441
+ provider: "amazon-bedrock",
2442
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2443
+ reasoning: false,
2444
+ input: ["text"],
2445
+ cost: {
2446
+ input: 0.72,
2447
+ output: 0.72,
2448
+ cacheRead: 0,
2449
+ cacheWrite: 0,
2450
+ },
2451
+ contextWindow: 128000,
2452
+ maxTokens: 4096,
2453
+ },
2454
+ "us.meta.llama3-1-8b-instruct-v1:0": {
2455
+ id: "us.meta.llama3-1-8b-instruct-v1:0",
2456
+ name: "Llama 3.1 8B Instruct (US)",
2457
+ api: "bedrock-converse-stream",
2458
+ provider: "amazon-bedrock",
2459
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2460
+ reasoning: false,
2461
+ input: ["text"],
2462
+ cost: {
2463
+ input: 0.22,
2464
+ output: 0.22,
2465
+ cacheRead: 0,
2466
+ cacheWrite: 0,
2467
+ },
2468
+ contextWindow: 128000,
2469
+ maxTokens: 4096,
2470
+ },
2471
+ "us.meta.llama3-3-70b-instruct-v1:0": {
2472
+ id: "us.meta.llama3-3-70b-instruct-v1:0",
2473
+ name: "Llama 3.3 70B Instruct (US)",
2474
+ api: "bedrock-converse-stream",
2475
+ provider: "amazon-bedrock",
2476
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2477
+ reasoning: false,
2478
+ input: ["text"],
2479
+ cost: {
2480
+ input: 0.72,
2481
+ output: 0.72,
2482
+ cacheRead: 0,
2483
+ cacheWrite: 0,
2484
+ },
2485
+ contextWindow: 128000,
2486
+ maxTokens: 4096,
2487
+ },
2010
2488
  "us.meta.llama4-maverick-17b-instruct-v1:0": {
2011
2489
  id: "us.meta.llama4-maverick-17b-instruct-v1:0",
2012
2490
  name: "Llama 4 Maverick 17B Instruct (US)",
@@ -2041,6 +2519,142 @@ export const MODELS = {
2041
2519
  contextWindow: 3500000,
2042
2520
  maxTokens: 16384,
2043
2521
  },
2522
+ "us.mistral.pixtral-large-2502-v1:0": {
2523
+ id: "us.mistral.pixtral-large-2502-v1:0",
2524
+ name: "Pixtral Large (25.02) (US)",
2525
+ api: "bedrock-converse-stream",
2526
+ provider: "amazon-bedrock",
2527
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2528
+ reasoning: false,
2529
+ input: ["text", "image"],
2530
+ cost: {
2531
+ input: 2,
2532
+ output: 6,
2533
+ cacheRead: 0,
2534
+ cacheWrite: 0,
2535
+ },
2536
+ contextWindow: 128000,
2537
+ maxTokens: 8192,
2538
+ },
2539
+ "us.openai.gpt-5.6-luna": {
2540
+ id: "us.openai.gpt-5.6-luna",
2541
+ name: "GPT-5.6 Luna (US)",
2542
+ api: "bedrock-converse-stream",
2543
+ provider: "amazon-bedrock",
2544
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2545
+ reasoning: true,
2546
+ input: ["text", "image"],
2547
+ cost: {
2548
+ input: 0.22,
2549
+ output: 1.32,
2550
+ cacheRead: 0.022,
2551
+ cacheWrite: 0.275,
2552
+ },
2553
+ contextWindow: 1050000,
2554
+ maxTokens: 128000,
2555
+ },
2556
+ "us.openai.gpt-5.6-sol": {
2557
+ id: "us.openai.gpt-5.6-sol",
2558
+ name: "GPT-5.6 Sol (US)",
2559
+ api: "bedrock-converse-stream",
2560
+ provider: "amazon-bedrock",
2561
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2562
+ reasoning: true,
2563
+ input: ["text", "image"],
2564
+ cost: {
2565
+ input: 4.4,
2566
+ output: 22,
2567
+ cacheRead: 0.44,
2568
+ cacheWrite: 5.5,
2569
+ },
2570
+ contextWindow: 1050000,
2571
+ maxTokens: 128000,
2572
+ },
2573
+ "us.openai.gpt-5.6-terra": {
2574
+ id: "us.openai.gpt-5.6-terra",
2575
+ name: "GPT-5.6 Terra (US)",
2576
+ api: "bedrock-converse-stream",
2577
+ provider: "amazon-bedrock",
2578
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2579
+ reasoning: true,
2580
+ input: ["text", "image"],
2581
+ cost: {
2582
+ input: 2.2,
2583
+ output: 13.2,
2584
+ cacheRead: 0.22,
2585
+ cacheWrite: 2.75,
2586
+ },
2587
+ contextWindow: 1050000,
2588
+ maxTokens: 128000,
2589
+ },
2590
+ "us.openai.gpt-6-astra": {
2591
+ id: "us.openai.gpt-6-astra",
2592
+ name: "GPT-6 Astra (US)",
2593
+ api: "bedrock-converse-stream",
2594
+ provider: "amazon-bedrock",
2595
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2596
+ reasoning: true,
2597
+ input: ["text", "image"],
2598
+ cost: {
2599
+ input: 11,
2600
+ output: 55,
2601
+ cacheRead: 1.1,
2602
+ cacheWrite: 13.75,
2603
+ },
2604
+ contextWindow: 1050000,
2605
+ maxTokens: 128000,
2606
+ },
2607
+ "us.writer.palmyra-x4-v1:0": {
2608
+ id: "us.writer.palmyra-x4-v1:0",
2609
+ name: "Palmyra X4 (US)",
2610
+ api: "bedrock-converse-stream",
2611
+ provider: "amazon-bedrock",
2612
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2613
+ reasoning: true,
2614
+ input: ["text"],
2615
+ cost: {
2616
+ input: 2.5,
2617
+ output: 10,
2618
+ cacheRead: 0,
2619
+ cacheWrite: 0,
2620
+ },
2621
+ contextWindow: 122880,
2622
+ maxTokens: 8192,
2623
+ },
2624
+ "us.writer.palmyra-x5-v1:0": {
2625
+ id: "us.writer.palmyra-x5-v1:0",
2626
+ name: "Palmyra X5 (US)",
2627
+ api: "bedrock-converse-stream",
2628
+ provider: "amazon-bedrock",
2629
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2630
+ reasoning: true,
2631
+ input: ["text"],
2632
+ cost: {
2633
+ input: 0.6,
2634
+ output: 6,
2635
+ cacheRead: 0,
2636
+ cacheWrite: 0,
2637
+ },
2638
+ contextWindow: 1040000,
2639
+ maxTokens: 8192,
2640
+ },
2641
+ "us.xai.grok-4.6": {
2642
+ id: "us.xai.grok-4.6",
2643
+ name: "Grok 4.6 (US)",
2644
+ api: "bedrock-converse-stream",
2645
+ provider: "amazon-bedrock",
2646
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
2647
+ reasoning: true,
2648
+ input: ["text", "image"],
2649
+ cost: {
2650
+ input: 2.2,
2651
+ output: 6.6,
2652
+ cacheRead: 0.55,
2653
+ cacheWrite: 0,
2654
+ },
2655
+ contextWindow: 500000,
2656
+ maxTokens: 500000,
2657
+ },
2044
2658
  "writer.palmyra-x4-v1:0": {
2045
2659
  id: "writer.palmyra-x4-v1:0",
2046
2660
  name: "Palmyra X4",
@@ -3093,39 +3707,39 @@ export const MODELS = {
3093
3707
  },
3094
3708
  },
3095
3709
  "cerebras": {
3096
- "gemma-4-31b": {
3097
- id: "gemma-4-31b",
3098
- name: "Gemma 4 31B IT",
3710
+ "gpt-oss-120b": {
3711
+ id: "gpt-oss-120b",
3712
+ name: "GPT OSS 120B",
3099
3713
  api: "openai-completions",
3100
3714
  provider: "cerebras",
3101
3715
  baseUrl: "https://api.cerebras.ai/v1",
3102
3716
  reasoning: true,
3103
- input: ["text", "image"],
3717
+ input: ["text"],
3104
3718
  cost: {
3105
- input: 0.99,
3106
- output: 1.49,
3719
+ input: 0.35,
3720
+ output: 0.75,
3107
3721
  cacheRead: 0,
3108
3722
  cacheWrite: 0,
3109
3723
  },
3110
3724
  contextWindow: 131072,
3111
3725
  maxTokens: 40960,
3112
3726
  },
3113
- "gpt-oss-120b": {
3114
- id: "gpt-oss-120b",
3115
- name: "GPT OSS 120B",
3727
+ "qwen-3.8-27b": {
3728
+ id: "qwen-3.8-27b",
3729
+ name: "Qwen3.8 27B",
3116
3730
  api: "openai-completions",
3117
3731
  provider: "cerebras",
3118
3732
  baseUrl: "https://api.cerebras.ai/v1",
3119
3733
  reasoning: true,
3120
- input: ["text"],
3734
+ input: ["text", "image"],
3121
3735
  cost: {
3122
- input: 0.35,
3123
- output: 0.75,
3736
+ input: 0.99,
3737
+ output: 1.49,
3124
3738
  cacheRead: 0,
3125
3739
  cacheWrite: 0,
3126
3740
  },
3127
- contextWindow: 131072,
3128
- maxTokens: 40960,
3741
+ contextWindow: 65536,
3742
+ maxTokens: 32768,
3129
3743
  },
3130
3744
  },
3131
3745
  "cloudflare-ai-gateway": {
@@ -4112,6 +4726,23 @@ export const MODELS = {
4112
4726
  contextWindow: 1000000,
4113
4727
  maxTokens: 384000,
4114
4728
  },
4729
+ "accounts/fireworks/models/deepseek-v4p1-flash": {
4730
+ id: "accounts/fireworks/models/deepseek-v4p1-flash",
4731
+ name: "DeepSeek V4.1 Flash",
4732
+ api: "anthropic-messages",
4733
+ provider: "fireworks",
4734
+ baseUrl: "https://api.fireworks.ai/inference",
4735
+ reasoning: true,
4736
+ input: ["text", "image"],
4737
+ cost: {
4738
+ input: 0.22,
4739
+ output: 0.66,
4740
+ cacheRead: 0.007,
4741
+ cacheWrite: 0,
4742
+ },
4743
+ contextWindow: 1000000,
4744
+ maxTokens: 384000,
4745
+ },
4115
4746
  "accounts/fireworks/models/glm-5p2": {
4116
4747
  id: "accounts/fireworks/models/glm-5p2",
4117
4748
  name: "GLM 5.2",
@@ -4143,8 +4774,8 @@ export const MODELS = {
4143
4774
  cacheRead: 0.26,
4144
4775
  cacheWrite: 0,
4145
4776
  },
4146
- contextWindow: 1000000,
4147
- maxTokens: 131072,
4777
+ contextWindow: 1048573,
4778
+ maxTokens: 262144,
4148
4779
  },
4149
4780
  "accounts/fireworks/models/glm-5p3-flash": {
4150
4781
  id: "accounts/fireworks/models/glm-5p3-flash",
@@ -4160,7 +4791,7 @@ export const MODELS = {
4160
4791
  cacheRead: 0.03,
4161
4792
  cacheWrite: 0,
4162
4793
  },
4163
- contextWindow: 1000000,
4794
+ contextWindow: 1048573,
4164
4795
  maxTokens: 131072,
4165
4796
  },
4166
4797
  "accounts/fireworks/models/gpt-oss-120b": {
@@ -4265,6 +4896,23 @@ export const MODELS = {
4265
4896
  contextWindow: 512000,
4266
4897
  maxTokens: 512000,
4267
4898
  },
4899
+ "accounts/fireworks/models/mistral-large-3-fp8": {
4900
+ id: "accounts/fireworks/models/mistral-large-3-fp8",
4901
+ name: "Mistral Large 3 675B Instruct 2512",
4902
+ api: "anthropic-messages",
4903
+ provider: "fireworks",
4904
+ baseUrl: "https://api.fireworks.ai/inference",
4905
+ reasoning: false,
4906
+ input: ["text", "image"],
4907
+ cost: {
4908
+ input: 0,
4909
+ output: 0,
4910
+ cacheRead: 0,
4911
+ cacheWrite: 0,
4912
+ },
4913
+ contextWindow: 262144,
4914
+ maxTokens: 262144,
4915
+ },
4268
4916
  "accounts/fireworks/models/muse-glimmer-30b": {
4269
4917
  id: "accounts/fireworks/models/muse-glimmer-30b",
4270
4918
  name: "Muse Glimmer 30B",
@@ -4381,8 +5029,25 @@ export const MODELS = {
4381
5029
  cacheRead: 0.21,
4382
5030
  cacheWrite: 0,
4383
5031
  },
4384
- contextWindow: 1048575,
4385
- maxTokens: 131072,
5032
+ contextWindow: 1048575,
5033
+ maxTokens: 131072,
5034
+ },
5035
+ "accounts/fireworks/routers/glm-5p3-fast": {
5036
+ id: "accounts/fireworks/routers/glm-5p3-fast",
5037
+ name: "GLM 5.3 Fast",
5038
+ api: "anthropic-messages",
5039
+ provider: "fireworks",
5040
+ baseUrl: "https://api.fireworks.ai/inference",
5041
+ reasoning: true,
5042
+ input: ["text"],
5043
+ cost: {
5044
+ input: 2.1,
5045
+ output: 6.6,
5046
+ cacheRead: 0.39,
5047
+ cacheWrite: 0,
5048
+ },
5049
+ contextWindow: 1048572,
5050
+ maxTokens: 262144,
4386
5051
  },
4387
5052
  "accounts/fireworks/routers/kimi-k3-fast": {
4388
5053
  id: "accounts/fireworks/routers/kimi-k3-fast",
@@ -6396,6 +7061,78 @@ export const MODELS = {
6396
7061
  contextWindow: 1000000,
6397
7062
  maxTokens: 384000,
6398
7063
  },
7064
+ "deepseek-ai/DeepSeek-V4.1-Flash": {
7065
+ id: "deepseek-ai/DeepSeek-V4.1-Flash",
7066
+ name: "DeepSeek V4.1 Flash",
7067
+ api: "openai-completions",
7068
+ provider: "huggingface",
7069
+ baseUrl: "https://router.huggingface.co/v1",
7070
+ compat: { "supportsDeveloperRole": false },
7071
+ reasoning: true,
7072
+ input: ["text", "image"],
7073
+ cost: {
7074
+ input: 0.3,
7075
+ output: 1.2,
7076
+ cacheRead: 0,
7077
+ cacheWrite: 0,
7078
+ },
7079
+ contextWindow: 1048576,
7080
+ maxTokens: 384000,
7081
+ },
7082
+ "google/gemma-3-12b-it": {
7083
+ id: "google/gemma-3-12b-it",
7084
+ name: "Gemma 3 12B IT",
7085
+ api: "openai-completions",
7086
+ provider: "huggingface",
7087
+ baseUrl: "https://router.huggingface.co/v1",
7088
+ compat: { "supportsDeveloperRole": false },
7089
+ reasoning: false,
7090
+ input: ["text", "image"],
7091
+ cost: {
7092
+ input: 0.05,
7093
+ output: 0.15,
7094
+ cacheRead: 0,
7095
+ cacheWrite: 0,
7096
+ },
7097
+ contextWindow: 131072,
7098
+ maxTokens: 131072,
7099
+ },
7100
+ "google/gemma-3-27b-it": {
7101
+ id: "google/gemma-3-27b-it",
7102
+ name: "Gemma 3 27B IT",
7103
+ api: "openai-completions",
7104
+ provider: "huggingface",
7105
+ baseUrl: "https://router.huggingface.co/v1",
7106
+ compat: { "supportsDeveloperRole": false },
7107
+ reasoning: false,
7108
+ input: ["text", "image"],
7109
+ cost: {
7110
+ input: 0.08,
7111
+ output: 0.16,
7112
+ cacheRead: 0,
7113
+ cacheWrite: 0,
7114
+ },
7115
+ contextWindow: 131072,
7116
+ maxTokens: 131072,
7117
+ },
7118
+ "google/gemma-3-4b-it": {
7119
+ id: "google/gemma-3-4b-it",
7120
+ name: "Gemma 3 4B IT",
7121
+ api: "openai-completions",
7122
+ provider: "huggingface",
7123
+ baseUrl: "https://router.huggingface.co/v1",
7124
+ compat: { "supportsDeveloperRole": false },
7125
+ reasoning: false,
7126
+ input: ["text", "image"],
7127
+ cost: {
7128
+ input: 0.05,
7129
+ output: 0.1,
7130
+ cacheRead: 0,
7131
+ cacheWrite: 0,
7132
+ },
7133
+ contextWindow: 131072,
7134
+ maxTokens: 131072,
7135
+ },
6399
7136
  "google/gemma-4-26B-A4B-it": {
6400
7137
  id: "google/gemma-4-26B-A4B-it",
6401
7138
  name: "Gemma 4 26B A4B IT",
@@ -7664,114 +8401,6 @@ export const MODELS = {
7664
8401
  },
7665
8402
  },
7666
8403
  "moonshotai": {
7667
- "kimi-k2-0711-preview": {
7668
- id: "kimi-k2-0711-preview",
7669
- name: "Kimi K2 0711",
7670
- api: "openai-completions",
7671
- provider: "moonshotai",
7672
- baseUrl: "https://api.moonshot.ai/v1",
7673
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7674
- reasoning: false,
7675
- input: ["text"],
7676
- cost: {
7677
- input: 0.6,
7678
- output: 2.5,
7679
- cacheRead: 0.15,
7680
- cacheWrite: 0,
7681
- },
7682
- contextWindow: 131072,
7683
- maxTokens: 16384,
7684
- },
7685
- "kimi-k2-0905-preview": {
7686
- id: "kimi-k2-0905-preview",
7687
- name: "Kimi K2 0905",
7688
- api: "openai-completions",
7689
- provider: "moonshotai",
7690
- baseUrl: "https://api.moonshot.ai/v1",
7691
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7692
- reasoning: false,
7693
- input: ["text"],
7694
- cost: {
7695
- input: 0.6,
7696
- output: 2.5,
7697
- cacheRead: 0.15,
7698
- cacheWrite: 0,
7699
- },
7700
- contextWindow: 262144,
7701
- maxTokens: 262144,
7702
- },
7703
- "kimi-k2-thinking": {
7704
- id: "kimi-k2-thinking",
7705
- name: "Kimi K2 Thinking",
7706
- api: "openai-completions",
7707
- provider: "moonshotai",
7708
- baseUrl: "https://api.moonshot.ai/v1",
7709
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7710
- reasoning: true,
7711
- input: ["text"],
7712
- cost: {
7713
- input: 0.6,
7714
- output: 2.5,
7715
- cacheRead: 0.15,
7716
- cacheWrite: 0,
7717
- },
7718
- contextWindow: 262144,
7719
- maxTokens: 262144,
7720
- },
7721
- "kimi-k2-thinking-turbo": {
7722
- id: "kimi-k2-thinking-turbo",
7723
- name: "Kimi K2 Thinking Turbo",
7724
- api: "openai-completions",
7725
- provider: "moonshotai",
7726
- baseUrl: "https://api.moonshot.ai/v1",
7727
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7728
- reasoning: true,
7729
- input: ["text"],
7730
- cost: {
7731
- input: 1.15,
7732
- output: 8,
7733
- cacheRead: 0.15,
7734
- cacheWrite: 0,
7735
- },
7736
- contextWindow: 262144,
7737
- maxTokens: 262144,
7738
- },
7739
- "kimi-k2-turbo-preview": {
7740
- id: "kimi-k2-turbo-preview",
7741
- name: "Kimi K2 Turbo",
7742
- api: "openai-completions",
7743
- provider: "moonshotai",
7744
- baseUrl: "https://api.moonshot.ai/v1",
7745
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7746
- reasoning: false,
7747
- input: ["text"],
7748
- cost: {
7749
- input: 2.4,
7750
- output: 10,
7751
- cacheRead: 0.6,
7752
- cacheWrite: 0,
7753
- },
7754
- contextWindow: 262144,
7755
- maxTokens: 262144,
7756
- },
7757
- "kimi-k2.5": {
7758
- id: "kimi-k2.5",
7759
- name: "Kimi K2.5",
7760
- api: "openai-completions",
7761
- provider: "moonshotai",
7762
- baseUrl: "https://api.moonshot.ai/v1",
7763
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7764
- reasoning: true,
7765
- input: ["text", "image"],
7766
- cost: {
7767
- input: 0.6,
7768
- output: 3,
7769
- cacheRead: 0.1,
7770
- cacheWrite: 0,
7771
- },
7772
- contextWindow: 262144,
7773
- maxTokens: 262144,
7774
- },
7775
8404
  "kimi-k2.6": {
7776
8405
  id: "kimi-k2.6",
7777
8406
  name: "Kimi K2.6",
@@ -7824,136 +8453,28 @@ export const MODELS = {
7824
8453
  cacheWrite: 0,
7825
8454
  },
7826
8455
  contextWindow: 262144,
7827
- maxTokens: 262144,
7828
- },
7829
- "kimi-k3": {
7830
- id: "kimi-k3",
7831
- name: "Kimi K3",
7832
- api: "openai-completions",
7833
- provider: "moonshotai",
7834
- baseUrl: "https://api.moonshot.ai/v1",
7835
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7836
- reasoning: true,
7837
- input: ["text", "image"],
7838
- cost: {
7839
- input: 3,
7840
- output: 15,
7841
- cacheRead: 0.3,
7842
- cacheWrite: 0,
7843
- },
7844
- contextWindow: 1048576,
7845
- maxTokens: 131072,
7846
- },
7847
- },
7848
- "moonshotai-cn": {
7849
- "kimi-k2-0711-preview": {
7850
- id: "kimi-k2-0711-preview",
7851
- name: "Kimi K2 0711",
7852
- api: "openai-completions",
7853
- provider: "moonshotai-cn",
7854
- baseUrl: "https://api.moonshot.cn/v1",
7855
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7856
- reasoning: false,
7857
- input: ["text"],
7858
- cost: {
7859
- input: 0.6,
7860
- output: 2.5,
7861
- cacheRead: 0.15,
7862
- cacheWrite: 0,
7863
- },
7864
- contextWindow: 131072,
7865
- maxTokens: 16384,
7866
- },
7867
- "kimi-k2-0905-preview": {
7868
- id: "kimi-k2-0905-preview",
7869
- name: "Kimi K2 0905",
7870
- api: "openai-completions",
7871
- provider: "moonshotai-cn",
7872
- baseUrl: "https://api.moonshot.cn/v1",
7873
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7874
- reasoning: false,
7875
- input: ["text"],
7876
- cost: {
7877
- input: 0.6,
7878
- output: 2.5,
7879
- cacheRead: 0.15,
7880
- cacheWrite: 0,
7881
- },
7882
- contextWindow: 262144,
7883
- maxTokens: 262144,
7884
- },
7885
- "kimi-k2-thinking": {
7886
- id: "kimi-k2-thinking",
7887
- name: "Kimi K2 Thinking",
7888
- api: "openai-completions",
7889
- provider: "moonshotai-cn",
7890
- baseUrl: "https://api.moonshot.cn/v1",
7891
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7892
- reasoning: true,
7893
- input: ["text"],
7894
- cost: {
7895
- input: 0.6,
7896
- output: 2.5,
7897
- cacheRead: 0.15,
7898
- cacheWrite: 0,
7899
- },
7900
- contextWindow: 262144,
7901
- maxTokens: 262144,
7902
- },
7903
- "kimi-k2-thinking-turbo": {
7904
- id: "kimi-k2-thinking-turbo",
7905
- name: "Kimi K2 Thinking Turbo",
7906
- api: "openai-completions",
7907
- provider: "moonshotai-cn",
7908
- baseUrl: "https://api.moonshot.cn/v1",
7909
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7910
- reasoning: true,
7911
- input: ["text"],
7912
- cost: {
7913
- input: 1.15,
7914
- output: 8,
7915
- cacheRead: 0.15,
7916
- cacheWrite: 0,
7917
- },
7918
- contextWindow: 262144,
7919
- maxTokens: 262144,
7920
- },
7921
- "kimi-k2-turbo-preview": {
7922
- id: "kimi-k2-turbo-preview",
7923
- name: "Kimi K2 Turbo",
7924
- api: "openai-completions",
7925
- provider: "moonshotai-cn",
7926
- baseUrl: "https://api.moonshot.cn/v1",
7927
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7928
- reasoning: false,
7929
- input: ["text"],
7930
- cost: {
7931
- input: 2.4,
7932
- output: 10,
7933
- cacheRead: 0.6,
7934
- cacheWrite: 0,
7935
- },
7936
- contextWindow: 262144,
7937
- maxTokens: 262144,
7938
- },
7939
- "kimi-k2.5": {
7940
- id: "kimi-k2.5",
7941
- name: "Kimi K2.5",
8456
+ maxTokens: 262144,
8457
+ },
8458
+ "kimi-k3": {
8459
+ id: "kimi-k3",
8460
+ name: "Kimi K3",
7942
8461
  api: "openai-completions",
7943
- provider: "moonshotai-cn",
7944
- baseUrl: "https://api.moonshot.cn/v1",
8462
+ provider: "moonshotai",
8463
+ baseUrl: "https://api.moonshot.ai/v1",
7945
8464
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false },
7946
8465
  reasoning: true,
7947
8466
  input: ["text", "image"],
7948
8467
  cost: {
7949
- input: 0.6,
7950
- output: 3,
7951
- cacheRead: 0.1,
8468
+ input: 3,
8469
+ output: 15,
8470
+ cacheRead: 0.3,
7952
8471
  cacheWrite: 0,
7953
8472
  },
7954
- contextWindow: 262144,
7955
- maxTokens: 262144,
8473
+ contextWindow: 1048576,
8474
+ maxTokens: 131072,
7956
8475
  },
8476
+ },
8477
+ "moonshotai-cn": {
7957
8478
  "kimi-k2.6": {
7958
8479
  id: "kimi-k2.6",
7959
8480
  name: "Kimi K2.6",
@@ -10086,9 +10607,9 @@ export const MODELS = {
10086
10607
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10087
10608
  input: ["text"],
10088
10609
  cost: {
10089
- input: 0.22,
10090
- output: 0.66,
10091
- cacheRead: 0.007,
10610
+ input: 0.15,
10611
+ output: 0.6,
10612
+ cacheRead: 0.003,
10092
10613
  cacheWrite: 0,
10093
10614
  },
10094
10615
  contextWindow: 1000000,
@@ -10105,9 +10626,9 @@ export const MODELS = {
10105
10626
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10106
10627
  input: ["text", "image"],
10107
10628
  cost: {
10108
- input: 0.22,
10109
- output: 0.66,
10110
- cacheRead: 0.007,
10629
+ input: 0.15,
10630
+ output: 0.6,
10631
+ cacheRead: 0.003,
10111
10632
  cacheWrite: 0,
10112
10633
  },
10113
10634
  contextWindow: 1000000,
@@ -10132,6 +10653,25 @@ export const MODELS = {
10132
10653
  contextWindow: 1000000,
10133
10654
  maxTokens: 384000,
10134
10655
  },
10656
+ "deepseek-v4.1-flash": {
10657
+ id: "deepseek-v4.1-flash",
10658
+ name: "DeepSeek V4.1 Flash",
10659
+ api: "openai-completions",
10660
+ provider: "opencode-go",
10661
+ baseUrl: "https://opencode.ai/zen/go/v1",
10662
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
10663
+ reasoning: true,
10664
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
10665
+ input: ["text", "image"],
10666
+ cost: {
10667
+ input: 0.15,
10668
+ output: 0.6,
10669
+ cacheRead: 0.003,
10670
+ cacheWrite: 0,
10671
+ },
10672
+ contextWindow: 1000000,
10673
+ maxTokens: 384000,
10674
+ },
10135
10675
  "glm-5.1": {
10136
10676
  id: "glm-5.1",
10137
10677
  name: "GLM-5.1",
@@ -10185,16 +10725,16 @@ export const MODELS = {
10185
10725
  },
10186
10726
  "glm-5.3-flash": {
10187
10727
  id: "glm-5.3-flash",
10188
- name: "GLM-5.3-Flash (2x usage)",
10728
+ name: "GLM-5.3-Flash",
10189
10729
  api: "openai-completions",
10190
10730
  provider: "opencode-go",
10191
10731
  baseUrl: "https://opencode.ai/zen/go/v1",
10192
10732
  reasoning: true,
10193
10733
  input: ["text", "image"],
10194
10734
  cost: {
10195
- input: 0.075,
10196
- output: 0.25,
10197
- cacheRead: 0.015,
10735
+ input: 0.15,
10736
+ output: 0.5,
10737
+ cacheRead: 0.03,
10198
10738
  cacheWrite: 0,
10199
10739
  },
10200
10740
  contextWindow: 1000000,
@@ -10439,23 +10979,6 @@ export const MODELS = {
10439
10979
  contextWindow: 1048576,
10440
10980
  maxTokens: 131072,
10441
10981
  },
10442
- "omen-alpha": {
10443
- id: "omen-alpha",
10444
- name: "Omen Alpha",
10445
- api: "openai-completions",
10446
- provider: "opencode-go",
10447
- baseUrl: "https://opencode.ai/zen/go/v1",
10448
- reasoning: true,
10449
- input: ["text", "image"],
10450
- cost: {
10451
- input: 0.2,
10452
- output: 0.66,
10453
- cacheRead: 0.04,
10454
- cacheWrite: 0,
10455
- },
10456
- contextWindow: 500000,
10457
- maxTokens: 128000,
10458
- },
10459
10982
  "qwen3.6-plus": {
10460
10983
  id: "qwen3.6-plus",
10461
10984
  name: "Qwen3.6 Plus",
@@ -11345,13 +11868,13 @@ export const MODELS = {
11345
11868
  reasoning: false,
11346
11869
  input: ["text"],
11347
11870
  cost: {
11348
- input: 0.32,
11349
- output: 0.8899999999999999,
11871
+ input: 0.2574,
11872
+ output: 1.0287,
11350
11873
  cacheRead: 0,
11351
11874
  cacheWrite: 0,
11352
11875
  },
11353
11876
  contextWindow: 163840,
11354
- maxTokens: 16384,
11877
+ maxTokens: 16000,
11355
11878
  },
11356
11879
  "deepseek/deepseek-chat-v3-0324": {
11357
11880
  id: "deepseek/deepseek-chat-v3-0324",
@@ -11379,13 +11902,13 @@ export const MODELS = {
11379
11902
  reasoning: true,
11380
11903
  input: ["text"],
11381
11904
  cost: {
11382
- input: 0.55,
11383
- output: 1.6500000000000001,
11384
- cacheRead: 0.55,
11905
+ input: 0.25,
11906
+ output: 0.95,
11907
+ cacheRead: 0.13,
11385
11908
  cacheWrite: 0,
11386
11909
  },
11387
11910
  contextWindow: 163840,
11388
- maxTokens: 144900,
11911
+ maxTokens: 32768,
11389
11912
  },
11390
11913
  "deepseek/deepseek-r1": {
11391
11914
  id: "deepseek/deepseek-r1",
@@ -11483,9 +12006,9 @@ export const MODELS = {
11483
12006
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11484
12007
  input: ["text"],
11485
12008
  cost: {
11486
- input: 0.08358,
11487
- output: 0.16716,
11488
- cacheRead: 0.016715999999999998,
12009
+ input: 0.088606,
12010
+ output: 0.177212,
12011
+ cacheRead: 0.017721200000000003,
11489
12012
  cacheWrite: 0,
11490
12013
  },
11491
12014
  contextWindow: 1048576,
@@ -11521,9 +12044,9 @@ export const MODELS = {
11521
12044
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11522
12045
  input: ["text"],
11523
12046
  cost: {
11524
- input: 0.14,
11525
- output: 0.28,
11526
- cacheRead: 0.03,
12047
+ input: 0.11,
12048
+ output: 0.33,
12049
+ cacheRead: 0.0035,
11527
12050
  cacheWrite: 0,
11528
12051
  },
11529
12052
  contextWindow: 1048576,
@@ -11546,7 +12069,26 @@ export const MODELS = {
11546
12069
  cacheWrite: 0,
11547
12070
  },
11548
12071
  contextWindow: 1048576,
11549
- maxTokens: 384000,
12072
+ maxTokens: 943718,
12073
+ },
12074
+ "deepseek/deepseek-v4-flash-vision-exp:batch": {
12075
+ id: "deepseek/deepseek-v4-flash-vision-exp:batch",
12076
+ name: "DeepSeek: DeepSeek V4 Flash Vision Exp (batch)",
12077
+ api: "openai-completions",
12078
+ provider: "openrouter",
12079
+ baseUrl: "https://openrouter.ai/api/v1",
12080
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
12081
+ reasoning: true,
12082
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12083
+ input: ["text", "image"],
12084
+ cost: {
12085
+ input: 0.11,
12086
+ output: 0.33,
12087
+ cacheRead: 0.0035,
12088
+ cacheWrite: 0,
12089
+ },
12090
+ contextWindow: 1048576,
12091
+ maxTokens: 943718,
11550
12092
  },
11551
12093
  "deepseek/deepseek-v4-pro": {
11552
12094
  id: "deepseek/deepseek-v4-pro",
@@ -11559,9 +12101,9 @@ export const MODELS = {
11559
12101
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11560
12102
  input: ["text"],
11561
12103
  cost: {
11562
- input: 0.8193659999999999,
11563
- output: 1.6387319999999999,
11564
- cacheRead: 0.0682805,
12104
+ input: 0.9552599999999999,
12105
+ output: 1.9105199999999998,
12106
+ cacheRead: 0.07960500000000001,
11565
12107
  cacheWrite: 0,
11566
12108
  },
11567
12109
  contextWindow: 1048576,
@@ -11578,13 +12120,13 @@ export const MODELS = {
11578
12120
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11579
12121
  input: ["text"],
11580
12122
  cost: {
11581
- input: 0.57948,
11582
- output: 1.73844,
11583
- cacheRead: 0.019316,
12123
+ input: 0.66,
12124
+ output: 1.9800000000000002,
12125
+ cacheRead: 0.06599999999999999,
11584
12126
  cacheWrite: 0,
11585
12127
  },
11586
12128
  contextWindow: 1048576,
11587
- maxTokens: 384000,
12129
+ maxTokens: 943718,
11588
12130
  },
11589
12131
  "deepseek/deepseek-v4-pro-0813:batch": {
11590
12132
  id: "deepseek/deepseek-v4-pro-0813:batch",
@@ -11597,14 +12139,33 @@ export const MODELS = {
11597
12139
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
11598
12140
  input: ["text"],
11599
12141
  cost: {
11600
- input: 1.32,
11601
- output: 3.9600000000000004,
11602
- cacheRead: 0.13,
12142
+ input: 0.66,
12143
+ output: 1.9800000000000002,
12144
+ cacheRead: 0.022,
11603
12145
  cacheWrite: 0,
11604
12146
  },
11605
12147
  contextWindow: 1048576,
11606
12148
  maxTokens: 943718,
11607
12149
  },
12150
+ "deepseek/deepseek-v4.1-flash": {
12151
+ id: "deepseek/deepseek-v4.1-flash",
12152
+ name: "DeepSeek: DeepSeek V4.1 Flash",
12153
+ api: "openai-completions",
12154
+ provider: "openrouter",
12155
+ baseUrl: "https://openrouter.ai/api/v1",
12156
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
12157
+ reasoning: true,
12158
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
12159
+ input: ["text", "image"],
12160
+ cost: {
12161
+ input: 0.15,
12162
+ output: 0.6,
12163
+ cacheRead: 0.003,
12164
+ cacheWrite: 0,
12165
+ },
12166
+ contextWindow: 1048576,
12167
+ maxTokens: 384000,
12168
+ },
11608
12169
  "dots-studio/dots-3-note-preview:free": {
11609
12170
  id: "dots-studio/dots-3-note-preview:free",
11610
12171
  name: "Dots Studio: Dots3-Note Preview (free)",
@@ -12124,13 +12685,13 @@ export const MODELS = {
12124
12685
  reasoning: true,
12125
12686
  input: ["text", "image"],
12126
12687
  cost: {
12127
- input: 0.07,
12128
- output: 0.33999999999999997,
12688
+ input: 0.041999999999999996,
12689
+ output: 0.22,
12129
12690
  cacheRead: 0,
12130
12691
  cacheWrite: 0,
12131
12692
  },
12132
12693
  contextWindow: 262144,
12133
- maxTokens: 16384,
12694
+ maxTokens: 32768,
12134
12695
  },
12135
12696
  "google/gemma-4-26b-a4b-it:free": {
12136
12697
  id: "google/gemma-4-26b-a4b-it:free",
@@ -12209,9 +12770,9 @@ export const MODELS = {
12209
12770
  reasoning: true,
12210
12771
  input: ["text"],
12211
12772
  cost: {
12212
- input: 0.09999999999999999,
12213
- output: 0.15,
12214
- cacheRead: 0.049999999999999996,
12773
+ input: 0.06,
12774
+ output: 0.25,
12775
+ cacheRead: 0.015,
12215
12776
  cacheWrite: 0,
12216
12777
  },
12217
12778
  contextWindow: 131072,
@@ -12234,9 +12795,9 @@ export const MODELS = {
12234
12795
  contextWindow: 128000,
12235
12796
  maxTokens: 50000,
12236
12797
  },
12237
- "inception/mercury-2.5-preview": {
12238
- id: "inception/mercury-2.5-preview",
12239
- name: "Inception: Mercury 2.5 Preview",
12798
+ "inception/mercury-2.5": {
12799
+ id: "inception/mercury-2.5",
12800
+ name: "Inception: Mercury 2.5",
12240
12801
  api: "openai-completions",
12241
12802
  provider: "openrouter",
12242
12803
  baseUrl: "https://openrouter.ai/api/v1",
@@ -12319,6 +12880,23 @@ export const MODELS = {
12319
12880
  contextWindow: 262144,
12320
12881
  maxTokens: 32768,
12321
12882
  },
12883
+ "inclusionai/ling-3.0-flash-vl:free": {
12884
+ id: "inclusionai/ling-3.0-flash-vl:free",
12885
+ name: "inclusionAI: Ling 3.0 Flash VL (free)",
12886
+ api: "openai-completions",
12887
+ provider: "openrouter",
12888
+ baseUrl: "https://openrouter.ai/api/v1",
12889
+ reasoning: true,
12890
+ input: ["text", "image"],
12891
+ cost: {
12892
+ input: 0,
12893
+ output: 0,
12894
+ cacheRead: 0,
12895
+ cacheWrite: 0,
12896
+ },
12897
+ contextWindow: 262144,
12898
+ maxTokens: 32768,
12899
+ },
12322
12900
  "kwaipilot/kat-coder-pro-v2": {
12323
12901
  id: "kwaipilot/kat-coder-pro-v2",
12324
12902
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -12498,9 +13076,9 @@ export const MODELS = {
12498
13076
  reasoning: true,
12499
13077
  input: ["text", "image"],
12500
13078
  cost: {
12501
- input: 0.35,
12502
- output: 1.5,
12503
- cacheRead: 0.04,
13079
+ input: 0.175,
13080
+ output: 0.75,
13081
+ cacheRead: 0.02,
12504
13082
  cacheWrite: 0,
12505
13083
  },
12506
13084
  contextWindow: 131072,
@@ -12651,13 +13229,13 @@ export const MODELS = {
12651
13229
  reasoning: true,
12652
13230
  input: ["text"],
12653
13231
  cost: {
12654
- input: 0.27,
12655
- output: 1.08,
12656
- cacheRead: 0.027,
13232
+ input: 0.3,
13233
+ output: 1.2,
13234
+ cacheRead: 0.03,
12657
13235
  cacheWrite: 0,
12658
13236
  },
12659
13237
  contextWindow: 204800,
12660
- maxTokens: 128000,
13238
+ maxTokens: 131072,
12661
13239
  },
12662
13240
  "minimax/minimax-m2.7": {
12663
13241
  id: "minimax/minimax-m2.7",
@@ -12676,23 +13254,6 @@ export const MODELS = {
12676
13254
  contextWindow: 204800,
12677
13255
  maxTokens: 131072,
12678
13256
  },
12679
- "minimax/minimax-m2.7:free": {
12680
- id: "minimax/minimax-m2.7:free",
12681
- name: "MiniMax: MiniMax M2.7 (free)",
12682
- api: "openai-completions",
12683
- provider: "openrouter",
12684
- baseUrl: "https://openrouter.ai/api/v1",
12685
- reasoning: true,
12686
- input: ["text"],
12687
- cost: {
12688
- input: 0,
12689
- output: 0,
12690
- cacheRead: 0,
12691
- cacheWrite: 0,
12692
- },
12693
- contextWindow: 196608,
12694
- maxTokens: 176947,
12695
- },
12696
13257
  "minimax/minimax-m3": {
12697
13258
  id: "minimax/minimax-m3",
12698
13259
  name: "MiniMax: MiniMax M3",
@@ -12727,35 +13288,35 @@ export const MODELS = {
12727
13288
  contextWindow: 524288,
12728
13289
  maxTokens: 471859,
12729
13290
  },
12730
- "minimax/minimax-m3:free": {
12731
- id: "minimax/minimax-m3:free",
12732
- name: "MiniMax: MiniMax M3 (free)",
13291
+ "mistralai/codestral-2508": {
13292
+ id: "mistralai/codestral-2508",
13293
+ name: "Mistral: Codestral 2508",
12733
13294
  api: "openai-completions",
12734
13295
  provider: "openrouter",
12735
13296
  baseUrl: "https://openrouter.ai/api/v1",
12736
- reasoning: true,
12737
- input: ["text", "image"],
13297
+ reasoning: false,
13298
+ input: ["text"],
12738
13299
  cost: {
12739
- input: 0,
12740
- output: 0,
12741
- cacheRead: 0,
13300
+ input: 0.3,
13301
+ output: 0.8999999999999999,
13302
+ cacheRead: 0.03,
12742
13303
  cacheWrite: 0,
12743
13304
  },
12744
- contextWindow: 1048576,
12745
- maxTokens: 943718,
13305
+ contextWindow: 256000,
13306
+ maxTokens: 204800,
12746
13307
  },
12747
- "mistralai/codestral-2508": {
12748
- id: "mistralai/codestral-2508",
12749
- name: "Mistral: Codestral 2508",
13308
+ "mistralai/codestral-2508:batch": {
13309
+ id: "mistralai/codestral-2508:batch",
13310
+ name: "Mistral: Codestral 2508 (batch)",
12750
13311
  api: "openai-completions",
12751
13312
  provider: "openrouter",
12752
13313
  baseUrl: "https://openrouter.ai/api/v1",
12753
13314
  reasoning: false,
12754
13315
  input: ["text"],
12755
13316
  cost: {
12756
- input: 0.3,
12757
- output: 0.8999999999999999,
12758
- cacheRead: 0.03,
13317
+ input: 0.15,
13318
+ output: 0.44999999999999996,
13319
+ cacheRead: 0.015,
12759
13320
  cacheWrite: 0,
12760
13321
  },
12761
13322
  contextWindow: 256000,
@@ -12829,6 +13390,23 @@ export const MODELS = {
12829
13390
  contextWindow: 262144,
12830
13391
  maxTokens: 209715,
12831
13392
  },
13393
+ "mistralai/ministral-8b-2512:batch": {
13394
+ id: "mistralai/ministral-8b-2512:batch",
13395
+ name: "Mistral: Ministral 3 8B 2512 (batch)",
13396
+ api: "openai-completions",
13397
+ provider: "openrouter",
13398
+ baseUrl: "https://openrouter.ai/api/v1",
13399
+ reasoning: false,
13400
+ input: ["text", "image"],
13401
+ cost: {
13402
+ input: 0.075,
13403
+ output: 0.075,
13404
+ cacheRead: 0.0075,
13405
+ cacheWrite: 0,
13406
+ },
13407
+ contextWindow: 262144,
13408
+ maxTokens: 209715,
13409
+ },
12832
13410
  "mistralai/mistral-large": {
12833
13411
  id: "mistralai/mistral-large",
12834
13412
  name: "Mistral Large",
@@ -12880,6 +13458,23 @@ export const MODELS = {
12880
13458
  contextWindow: 262144,
12881
13459
  maxTokens: 209715,
12882
13460
  },
13461
+ "mistralai/mistral-large-2512:batch": {
13462
+ id: "mistralai/mistral-large-2512:batch",
13463
+ name: "Mistral: Mistral Large 3 2512 (batch)",
13464
+ api: "openai-completions",
13465
+ provider: "openrouter",
13466
+ baseUrl: "https://openrouter.ai/api/v1",
13467
+ reasoning: false,
13468
+ input: ["text", "image"],
13469
+ cost: {
13470
+ input: 0.25,
13471
+ output: 0.75,
13472
+ cacheRead: 0.024999999999999998,
13473
+ cacheWrite: 0,
13474
+ },
13475
+ contextWindow: 262144,
13476
+ maxTokens: 209715,
13477
+ },
12883
13478
  "mistralai/mistral-medium-3": {
12884
13479
  id: "mistralai/mistral-medium-3",
12885
13480
  name: "Mistral: Mistral Medium 3",
@@ -12928,8 +13523,8 @@ export const MODELS = {
12928
13523
  cacheRead: 0,
12929
13524
  cacheWrite: 0,
12930
13525
  },
12931
- contextWindow: 32768,
12932
- maxTokens: 26214,
13526
+ contextWindow: 262144,
13527
+ maxTokens: 209715,
12933
13528
  },
12934
13529
  "mistralai/mistral-medium-3.1": {
12935
13530
  id: "mistralai/mistral-medium-3.1",
@@ -12948,6 +13543,23 @@ export const MODELS = {
12948
13543
  contextWindow: 131072,
12949
13544
  maxTokens: 104857,
12950
13545
  },
13546
+ "mistralai/mistral-medium-3.1:batch": {
13547
+ id: "mistralai/mistral-medium-3.1:batch",
13548
+ name: "Mistral: Mistral Medium 3.1 (batch)",
13549
+ api: "openai-completions",
13550
+ provider: "openrouter",
13551
+ baseUrl: "https://openrouter.ai/api/v1",
13552
+ reasoning: false,
13553
+ input: ["text", "image"],
13554
+ cost: {
13555
+ input: 0.19999999999999998,
13556
+ output: 1,
13557
+ cacheRead: 0.02,
13558
+ cacheWrite: 0,
13559
+ },
13560
+ contextWindow: 131072,
13561
+ maxTokens: 104857,
13562
+ },
12951
13563
  "mistralai/mistral-nemo": {
12952
13564
  id: "mistralai/mistral-nemo",
12953
13565
  name: "Mistral: Mistral Nemo",
@@ -12999,6 +13611,23 @@ export const MODELS = {
12999
13611
  contextWindow: 262144,
13000
13612
  maxTokens: 209715,
13001
13613
  },
13614
+ "mistralai/mistral-small-2603:batch": {
13615
+ id: "mistralai/mistral-small-2603:batch",
13616
+ name: "Mistral: Mistral Small 4 (batch)",
13617
+ api: "openai-completions",
13618
+ provider: "openrouter",
13619
+ baseUrl: "https://openrouter.ai/api/v1",
13620
+ reasoning: true,
13621
+ input: ["text", "image"],
13622
+ cost: {
13623
+ input: 0.075,
13624
+ output: 0.3,
13625
+ cacheRead: 0.0075,
13626
+ cacheWrite: 0,
13627
+ },
13628
+ contextWindow: 262144,
13629
+ maxTokens: 209715,
13630
+ },
13002
13631
  "mistralai/mistral-small-3.2-24b-instruct": {
13003
13632
  id: "mistralai/mistral-small-3.2-24b-instruct",
13004
13633
  name: "Mistral: Mistral Small 3.2 24B",
@@ -13143,10 +13772,10 @@ export const MODELS = {
13143
13772
  baseUrl: "https://openrouter.ai/api/v1",
13144
13773
  reasoning: true,
13145
13774
  input: ["text", "image"],
13146
- cost: {
13147
- input: 0.66,
13148
- output: 3.4,
13149
- cacheRead: 0.18,
13775
+ cost: {
13776
+ input: 0.71,
13777
+ output: 3.5,
13778
+ cacheRead: 0.15,
13150
13779
  cacheWrite: 0,
13151
13780
  },
13152
13781
  contextWindow: 262144,
@@ -13161,9 +13790,9 @@ export const MODELS = {
13161
13790
  reasoning: true,
13162
13791
  input: ["text", "image"],
13163
13792
  cost: {
13164
- input: 3,
13165
- output: 15,
13166
- cacheRead: 0.3,
13793
+ input: 2.34,
13794
+ output: 11.7,
13795
+ cacheRead: 0.261,
13167
13796
  cacheWrite: 0,
13168
13797
  },
13169
13798
  contextWindow: 1048576,
@@ -13186,35 +13815,35 @@ export const MODELS = {
13186
13815
  contextWindow: 1048576,
13187
13816
  maxTokens: 943718,
13188
13817
  },
13189
- "nex-agi/nex-n2-mini": {
13190
- id: "nex-agi/nex-n2-mini",
13191
- name: "Nex AGI: Nex-N2-Mini",
13818
+ "nex-agi/nex-n2.5-mini:free": {
13819
+ id: "nex-agi/nex-n2.5-mini:free",
13820
+ name: "Nex AGI: Nex-N2.5-Mini (free)",
13192
13821
  api: "openai-completions",
13193
13822
  provider: "openrouter",
13194
13823
  baseUrl: "https://openrouter.ai/api/v1",
13195
13824
  reasoning: true,
13196
13825
  input: ["text", "image"],
13197
13826
  cost: {
13198
- input: 0.024999999999999998,
13199
- output: 0.09999999999999999,
13200
- cacheRead: 0.0025,
13827
+ input: 0,
13828
+ output: 0,
13829
+ cacheRead: 0,
13201
13830
  cacheWrite: 0,
13202
13831
  },
13203
13832
  contextWindow: 262144,
13204
13833
  maxTokens: 235929,
13205
13834
  },
13206
- "nex-agi/nex-n2-pro": {
13207
- id: "nex-agi/nex-n2-pro",
13208
- name: "Nex AGI: Nex-N2-Pro",
13835
+ "nex-agi/nex-n2.5-pro:free": {
13836
+ id: "nex-agi/nex-n2.5-pro:free",
13837
+ name: "Nex AGI: Nex-N2.5-Pro (free)",
13209
13838
  api: "openai-completions",
13210
13839
  provider: "openrouter",
13211
13840
  baseUrl: "https://openrouter.ai/api/v1",
13212
13841
  reasoning: true,
13213
13842
  input: ["text", "image"],
13214
13843
  cost: {
13215
- input: 0.25,
13216
- output: 1,
13217
- cacheRead: 0.024999999999999998,
13844
+ input: 0,
13845
+ output: 0,
13846
+ cacheRead: 0,
13218
13847
  cacheWrite: 0,
13219
13848
  },
13220
13849
  contextWindow: 262144,
@@ -13268,7 +13897,7 @@ export const MODELS = {
13268
13897
  cacheRead: 0,
13269
13898
  cacheWrite: 0,
13270
13899
  },
13271
- contextWindow: 1000000,
13900
+ contextWindow: 262144,
13272
13901
  maxTokens: 16384,
13273
13902
  },
13274
13903
  "nvidia/nemotron-3-super-120b-a12b:free": {
@@ -14948,13 +15577,13 @@ export const MODELS = {
14948
15577
  reasoning: true,
14949
15578
  input: ["text"],
14950
15579
  cost: {
14951
- input: 0.12,
14952
- output: 0.24,
15580
+ input: 0.22749999999999998,
15581
+ output: 0.9099999999999999,
14953
15582
  cacheRead: 0,
14954
15583
  cacheWrite: 0,
14955
15584
  },
14956
15585
  contextWindow: 131072,
14957
- maxTokens: 16384,
15586
+ maxTokens: 8192,
14958
15587
  },
14959
15588
  "qwen/qwen3-235b-a22b": {
14960
15589
  id: "qwen/qwen3-235b-a22b",
@@ -14982,8 +15611,8 @@ export const MODELS = {
14982
15611
  reasoning: false,
14983
15612
  input: ["text"],
14984
15613
  cost: {
14985
- input: 0.09,
14986
- output: 0.55,
15614
+ input: 0.22,
15615
+ output: 0.88,
14987
15616
  cacheRead: 0,
14988
15617
  cacheWrite: 0,
14989
15618
  },
@@ -15033,13 +15662,13 @@ export const MODELS = {
15033
15662
  reasoning: false,
15034
15663
  input: ["text"],
15035
15664
  cost: {
15036
- input: 0.04815,
15037
- output: 0.19305,
15665
+ input: 0.09,
15666
+ output: 0.3,
15038
15667
  cacheRead: 0,
15039
15668
  cacheWrite: 0,
15040
15669
  },
15041
15670
  contextWindow: 262144,
15042
- maxTokens: 32000,
15671
+ maxTokens: 235929,
15043
15672
  },
15044
15673
  "qwen/qwen3-30b-a3b-thinking-2507": {
15045
15674
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -15220,13 +15849,13 @@ export const MODELS = {
15220
15849
  reasoning: false,
15221
15850
  input: ["text"],
15222
15851
  cost: {
15223
- input: 0.09999999999999999,
15852
+ input: 0.09,
15224
15853
  output: 1.1,
15225
- cacheRead: 0.07,
15854
+ cacheRead: 0,
15226
15855
  cacheWrite: 0,
15227
15856
  },
15228
15857
  contextWindow: 262144,
15229
- maxTokens: 235929,
15858
+ maxTokens: 16384,
15230
15859
  },
15231
15860
  "qwen/qwen3-next-80b-a3b-thinking": {
15232
15861
  id: "qwen/qwen3-next-80b-a3b-thinking",
@@ -15243,7 +15872,7 @@ export const MODELS = {
15243
15872
  cacheWrite: 0,
15244
15873
  },
15245
15874
  contextWindow: 262144,
15246
- maxTokens: 32768,
15875
+ maxTokens: 235929,
15247
15876
  },
15248
15877
  "qwen/qwen3-vl-235b-a22b-instruct": {
15249
15878
  id: "qwen/qwen3-vl-235b-a22b-instruct",
@@ -15373,13 +16002,13 @@ export const MODELS = {
15373
16002
  reasoning: true,
15374
16003
  input: ["text", "image"],
15375
16004
  cost: {
15376
- input: 0.29,
15377
- output: 2.4,
16005
+ input: 0.26,
16006
+ output: 2.08,
15378
16007
  cacheRead: 0,
15379
16008
  cacheWrite: 0,
15380
16009
  },
15381
16010
  contextWindow: 262144,
15382
- maxTokens: 81920,
16011
+ maxTokens: 65536,
15383
16012
  },
15384
16013
  "qwen/qwen3.5-27b": {
15385
16014
  id: "qwen/qwen3.5-27b",
@@ -15668,7 +16297,7 @@ export const MODELS = {
15668
16297
  cacheWrite: 0,
15669
16298
  },
15670
16299
  contextWindow: 1048576,
15671
- maxTokens: 262144,
16300
+ maxTokens: 131072,
15672
16301
  },
15673
16302
  "qwen/qwen3.8-2.4t-a95b:batch": {
15674
16303
  id: "qwen/qwen3.8-2.4t-a95b:batch",
@@ -15772,6 +16401,23 @@ export const MODELS = {
15772
16401
  contextWindow: 256000,
15773
16402
  maxTokens: 128000,
15774
16403
  },
16404
+ "sakana/fugu-max": {
16405
+ id: "sakana/fugu-max",
16406
+ name: "Sakana: Fugu Max",
16407
+ api: "openai-completions",
16408
+ provider: "openrouter",
16409
+ baseUrl: "https://openrouter.ai/api/v1",
16410
+ reasoning: true,
16411
+ input: ["text", "image"],
16412
+ cost: {
16413
+ input: 2,
16414
+ output: 6,
16415
+ cacheRead: 0.25,
16416
+ cacheWrite: 0,
16417
+ },
16418
+ contextWindow: 1000000,
16419
+ maxTokens: 128000,
16420
+ },
15775
16421
  "sakana/fugu-ultra": {
15776
16422
  id: "sakana/fugu-ultra",
15777
16423
  name: "Sakana: Fugu Ultra",
@@ -15789,6 +16435,23 @@ export const MODELS = {
15789
16435
  contextWindow: 1000000,
15790
16436
  maxTokens: 128000,
15791
16437
  },
16438
+ "sakana/fugu-ultra-v2": {
16439
+ id: "sakana/fugu-ultra-v2",
16440
+ name: "Sakana: Fugu Ultra v2",
16441
+ api: "openai-completions",
16442
+ provider: "openrouter",
16443
+ baseUrl: "https://openrouter.ai/api/v1",
16444
+ reasoning: true,
16445
+ input: ["text", "image"],
16446
+ cost: {
16447
+ input: 5,
16448
+ output: 30,
16449
+ cacheRead: 0.5,
16450
+ cacheWrite: 0,
16451
+ },
16452
+ contextWindow: 1000000,
16453
+ maxTokens: 128000,
16454
+ },
15792
16455
  "sakana/sakana-namazu": {
15793
16456
  id: "sakana/sakana-namazu",
15794
16457
  name: "Sakana: Sakana Namazu",
@@ -15866,9 +16529,9 @@ export const MODELS = {
15866
16529
  reasoning: true,
15867
16530
  input: ["text"],
15868
16531
  cost: {
15869
- input: 0.0825,
15870
- output: 0.33,
15871
- cacheRead: 0.020625,
16532
+ input: 0.13199999999999998,
16533
+ output: 0.5279999999999999,
16534
+ cacheRead: 0.032999999999999995,
15872
16535
  cacheWrite: 0,
15873
16536
  },
15874
16537
  contextWindow: 262144,
@@ -15906,7 +16569,7 @@ export const MODELS = {
15906
16569
  cacheWrite: 0,
15907
16570
  },
15908
16571
  contextWindow: 1024000,
15909
- maxTokens: 26214,
16572
+ maxTokens: 819200,
15910
16573
  },
15911
16574
  "thinkingmachines/inkling": {
15912
16575
  id: "thinkingmachines/inkling",
@@ -15923,7 +16586,7 @@ export const MODELS = {
15923
16586
  cacheWrite: 0,
15924
16587
  },
15925
16588
  contextWindow: 1048576,
15926
- maxTokens: 471859,
16589
+ maxTokens: 32768,
15927
16590
  },
15928
16591
  "thinkingmachines/inkling-small": {
15929
16592
  id: "thinkingmachines/inkling-small",
@@ -16036,9 +16699,9 @@ export const MODELS = {
16036
16699
  reasoning: true,
16037
16700
  input: ["text"],
16038
16701
  cost: {
16039
- input: 0.03,
16040
- output: 0.12,
16041
- cacheRead: 0.006,
16702
+ input: 0.09,
16703
+ output: 0.36,
16704
+ cacheRead: 0.018,
16042
16705
  cacheWrite: 0,
16043
16706
  },
16044
16707
  contextWindow: 524288,
@@ -16240,13 +16903,13 @@ export const MODELS = {
16240
16903
  reasoning: true,
16241
16904
  input: ["text"],
16242
16905
  cost: {
16243
- input: 0.55,
16244
- output: 2.2,
16245
- cacheRead: 0.11,
16906
+ input: 0.43,
16907
+ output: 1.75,
16908
+ cacheRead: 0.08,
16246
16909
  cacheWrite: 0,
16247
16910
  },
16248
16911
  contextWindow: 204800,
16249
- maxTokens: 131072,
16912
+ maxTokens: 16384,
16250
16913
  },
16251
16914
  "z-ai/glm-4.6v": {
16252
16915
  id: "z-ai/glm-4.6v",
@@ -16291,13 +16954,13 @@ export const MODELS = {
16291
16954
  reasoning: true,
16292
16955
  input: ["text"],
16293
16956
  cost: {
16294
- input: 0.06,
16957
+ input: 0.060500000000000005,
16295
16958
  output: 0.39999999999999997,
16296
- cacheRead: 0.01,
16959
+ cacheRead: 0,
16297
16960
  cacheWrite: 0,
16298
16961
  },
16299
- contextWindow: 202752,
16300
- maxTokens: 16384,
16962
+ contextWindow: 200000,
16963
+ maxTokens: 117964,
16301
16964
  },
16302
16965
  "z-ai/glm-5": {
16303
16966
  id: "z-ai/glm-5",
@@ -16367,22 +17030,22 @@ export const MODELS = {
16367
17030
  contextWindow: 1048576,
16368
17031
  maxTokens: 131072,
16369
17032
  },
16370
- "z-ai/glm-5.2:free": {
16371
- id: "z-ai/glm-5.2:free",
16372
- name: "Z.ai: GLM 5.2 (free)",
17033
+ "z-ai/glm-5.2:batch": {
17034
+ id: "z-ai/glm-5.2:batch",
17035
+ name: "Z.ai: GLM 5.2 (batch)",
16373
17036
  api: "openai-completions",
16374
17037
  provider: "openrouter",
16375
17038
  baseUrl: "https://openrouter.ai/api/v1",
16376
17039
  reasoning: true,
16377
17040
  input: ["text"],
16378
17041
  cost: {
16379
- input: 0,
16380
- output: 0,
16381
- cacheRead: 0,
17042
+ input: 0.7,
17043
+ output: 2.2,
17044
+ cacheRead: 0.07,
16382
17045
  cacheWrite: 0,
16383
17046
  },
16384
- contextWindow: 256000,
16385
- maxTokens: 230400,
17047
+ contextWindow: 1048576,
17048
+ maxTokens: 943718,
16386
17049
  },
16387
17050
  "z-ai/glm-5.3": {
16388
17051
  id: "z-ai/glm-5.3",
@@ -16395,11 +17058,11 @@ export const MODELS = {
16395
17058
  cost: {
16396
17059
  input: 1.4,
16397
17060
  output: 4.4,
16398
- cacheRead: 0.14,
17061
+ cacheRead: 0.26,
16399
17062
  cacheWrite: 0,
16400
17063
  },
16401
17064
  contextWindow: 1310720,
16402
- maxTokens: 262144,
17065
+ maxTokens: 943718,
16403
17066
  },
16404
17067
  "z-ai/glm-5.3-flash": {
16405
17068
  id: "z-ai/glm-5.3-flash",
@@ -16410,9 +17073,9 @@ export const MODELS = {
16410
17073
  reasoning: true,
16411
17074
  input: ["text", "image"],
16412
17075
  cost: {
16413
- input: 0.075,
16414
- output: 0.25,
16415
- cacheRead: 0.015,
17076
+ input: 0.15,
17077
+ output: 0.5,
17078
+ cacheRead: 0.03,
16416
17079
  cacheWrite: 0,
16417
17080
  },
16418
17081
  contextWindow: 1310720,
@@ -16427,13 +17090,30 @@ export const MODELS = {
16427
17090
  reasoning: true,
16428
17091
  input: ["text", "image"],
16429
17092
  cost: {
16430
- input: 0.15,
16431
- output: 0.5,
16432
- cacheRead: 0.03,
17093
+ input: 0.075,
17094
+ output: 0.25,
17095
+ cacheRead: 0.015,
16433
17096
  cacheWrite: 0,
16434
17097
  },
16435
- contextWindow: 1048575,
16436
- maxTokens: 943717,
17098
+ contextWindow: 1048576,
17099
+ maxTokens: 943718,
17100
+ },
17101
+ "z-ai/glm-5.3:batch": {
17102
+ id: "z-ai/glm-5.3:batch",
17103
+ name: "Z.ai: GLM 5.3 (batch)",
17104
+ api: "openai-completions",
17105
+ provider: "openrouter",
17106
+ baseUrl: "https://openrouter.ai/api/v1",
17107
+ reasoning: true,
17108
+ input: ["text"],
17109
+ cost: {
17110
+ input: 0.7,
17111
+ output: 2.2,
17112
+ cacheRead: 0.13,
17113
+ cacheWrite: 0,
17114
+ },
17115
+ contextWindow: 1048576,
17116
+ maxTokens: 943718,
16437
17117
  },
16438
17118
  "z-ai/glm-5v-turbo": {
16439
17119
  id: "z-ai/glm-5v-turbo",
@@ -16532,13 +17212,13 @@ export const MODELS = {
16532
17212
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
16533
17213
  input: ["text"],
16534
17214
  cost: {
16535
- input: 0.049980000000000004,
16536
- output: 0.09996000000000001,
16537
- cacheRead: 0.009996000000000001,
17215
+ input: 0.049999999999999996,
17216
+ output: 0.16,
17217
+ cacheRead: 0.013000000000000001,
16538
17218
  cacheWrite: 0,
16539
17219
  },
16540
17220
  contextWindow: 1310720,
16541
- maxTokens: 131072,
17221
+ maxTokens: 393216,
16542
17222
  },
16543
17223
  "~google/gemini-flash-latest": {
16544
17224
  id: "~google/gemini-flash-latest",
@@ -16583,9 +17263,9 @@ export const MODELS = {
16583
17263
  reasoning: true,
16584
17264
  input: ["text", "image"],
16585
17265
  cost: {
16586
- input: 2.5500000000000003,
16587
- output: 12.75,
16588
- cacheRead: 0.25599998999999996,
17266
+ input: 2.34,
17267
+ output: 11.7,
17268
+ cacheRead: 0.261,
16589
17269
  cacheWrite: 0,
16590
17270
  },
16591
17271
  contextWindow: 1048576,
@@ -16657,7 +17337,7 @@ export const MODELS = {
16657
17337
  cacheWrite: 0,
16658
17338
  },
16659
17339
  contextWindow: 1310720,
16660
- maxTokens: 943718,
17340
+ maxTokens: 131072,
16661
17341
  },
16662
17342
  "~z-ai/glm-latest": {
16663
17343
  id: "~z-ai/glm-latest",
@@ -16668,13 +17348,13 @@ export const MODELS = {
16668
17348
  reasoning: true,
16669
17349
  input: ["text"],
16670
17350
  cost: {
16671
- input: 1.15,
16672
- output: 3.5,
16673
- cacheRead: 0.09999999999999999,
17351
+ input: 0.9700000000000001,
17352
+ output: 3.3077,
17353
+ cacheRead: 0.19399999999999998,
16674
17354
  cacheWrite: 0,
16675
17355
  },
16676
17356
  contextWindow: 1310720,
16677
- maxTokens: 235929,
17357
+ maxTokens: 943718,
16678
17358
  },
16679
17359
  },
16680
17360
  "together": {
@@ -17552,23 +18232,6 @@ export const MODELS = {
17552
18232
  contextWindow: 991000,
17553
18233
  maxTokens: 128000,
17554
18234
  },
17555
- "alibaba/qwen3.8-flash-next": {
17556
- id: "alibaba/qwen3.8-flash-next",
17557
- name: "Qwen 3.8 Flash Next",
17558
- api: "anthropic-messages",
17559
- provider: "vercel-ai-gateway",
17560
- baseUrl: "https://ai-gateway.vercel.sh",
17561
- reasoning: true,
17562
- input: ["text", "image"],
17563
- cost: {
17564
- input: 0.12,
17565
- output: 0.39999999999999997,
17566
- cacheRead: 0.01,
17567
- cacheWrite: 0,
17568
- },
17569
- contextWindow: 1048576,
17570
- maxTokens: 1048576,
17571
- },
17572
18235
  "alibaba/qwen3.8-max": {
17573
18236
  id: "alibaba/qwen3.8-max",
17574
18237
  name: "Qwen 3.8 Max",
@@ -18077,9 +18740,9 @@ export const MODELS = {
18077
18740
  reasoning: false,
18078
18741
  input: ["text"],
18079
18742
  cost: {
18080
- input: 0.28,
18081
- output: 0.42,
18082
- cacheRead: 0.028,
18743
+ input: 0.62,
18744
+ output: 1.85,
18745
+ cacheRead: 0,
18083
18746
  cacheWrite: 0,
18084
18747
  },
18085
18748
  contextWindow: 128000,
@@ -18187,6 +18850,23 @@ export const MODELS = {
18187
18850
  contextWindow: 1000000,
18188
18851
  maxTokens: 384000,
18189
18852
  },
18853
+ "deepseek/deepseek-v4.1-flash": {
18854
+ id: "deepseek/deepseek-v4.1-flash",
18855
+ name: "DeepSeek V4.1 Flash",
18856
+ api: "anthropic-messages",
18857
+ provider: "vercel-ai-gateway",
18858
+ baseUrl: "https://ai-gateway.vercel.sh",
18859
+ reasoning: true,
18860
+ input: ["text", "image"],
18861
+ cost: {
18862
+ input: 0.3,
18863
+ output: 1.2,
18864
+ cacheRead: 0.03,
18865
+ cacheWrite: 0,
18866
+ },
18867
+ contextWindow: 1048576,
18868
+ maxTokens: 32768,
18869
+ },
18190
18870
  "google/gemini-2.5-flash": {
18191
18871
  id: "google/gemini-2.5-flash",
18192
18872
  name: "Gemini 2.5 Flash",
@@ -18425,6 +19105,23 @@ export const MODELS = {
18425
19105
  contextWindow: 128000,
18426
19106
  maxTokens: 128000,
18427
19107
  },
19108
+ "inception/mercury-2.5": {
19109
+ id: "inception/mercury-2.5",
19110
+ name: "Mercury 2.5",
19111
+ api: "anthropic-messages",
19112
+ provider: "vercel-ai-gateway",
19113
+ baseUrl: "https://ai-gateway.vercel.sh",
19114
+ reasoning: true,
19115
+ input: ["text"],
19116
+ cost: {
19117
+ input: 0.04,
19118
+ output: 0.15,
19119
+ cacheRead: 0.004,
19120
+ cacheWrite: 0,
19121
+ },
19122
+ contextWindow: 260000,
19123
+ maxTokens: 65536,
19124
+ },
18428
19125
  "inception/mercury-coder-small": {
18429
19126
  id: "inception/mercury-coder-small",
18430
19127
  name: "Mercury Coder Small Beta",
@@ -18901,23 +19598,6 @@ export const MODELS = {
18901
19598
  contextWindow: 204800,
18902
19599
  maxTokens: 131000,
18903
19600
  },
18904
- "minimax/minimax-m2.7-free": {
18905
- id: "minimax/minimax-m2.7-free",
18906
- name: "MiniMax M2.7 (Free)",
18907
- api: "anthropic-messages",
18908
- provider: "vercel-ai-gateway",
18909
- baseUrl: "https://ai-gateway.vercel.sh",
18910
- reasoning: true,
18911
- input: ["text"],
18912
- cost: {
18913
- input: 0,
18914
- output: 0,
18915
- cacheRead: 0,
18916
- cacheWrite: 0,
18917
- },
18918
- contextWindow: 196608,
18919
- maxTokens: 196608,
18920
- },
18921
19601
  "minimax/minimax-m2.7-highspeed": {
18922
19602
  id: "minimax/minimax-m2.7-highspeed",
18923
19603
  name: "MiniMax M2.7 High Speed",
@@ -18952,23 +19632,6 @@ export const MODELS = {
18952
19632
  contextWindow: 512000,
18953
19633
  maxTokens: 512000,
18954
19634
  },
18955
- "minimax/minimax-m3-free": {
18956
- id: "minimax/minimax-m3-free",
18957
- name: "MiniMax M3 (Free)",
18958
- api: "anthropic-messages",
18959
- provider: "vercel-ai-gateway",
18960
- baseUrl: "https://ai-gateway.vercel.sh",
18961
- reasoning: true,
18962
- input: ["text", "image"],
18963
- cost: {
18964
- input: 0,
18965
- output: 0,
18966
- cacheRead: 0,
18967
- cacheWrite: 0,
18968
- },
18969
- contextWindow: 1048576,
18970
- maxTokens: 1048576,
18971
- },
18972
19635
  "mistral/codestral": {
18973
19636
  id: "mistral/codestral",
18974
19637
  name: "Mistral Codestral",
@@ -20803,23 +21466,6 @@ export const MODELS = {
20803
21466
  contextWindow: 1050000,
20804
21467
  maxTokens: 131000,
20805
21468
  },
20806
- "xiaomi/mimo-v2.5-pro-ultraspeed": {
20807
- id: "xiaomi/mimo-v2.5-pro-ultraspeed",
20808
- name: "MiMo V2.5 Pro UltraSpeed",
20809
- api: "anthropic-messages",
20810
- provider: "vercel-ai-gateway",
20811
- baseUrl: "https://ai-gateway.vercel.sh",
20812
- reasoning: true,
20813
- input: ["text"],
20814
- cost: {
20815
- input: 1.305,
20816
- output: 2.61,
20817
- cacheRead: 0.0108,
20818
- cacheWrite: 0,
20819
- },
20820
- contextWindow: 1048576,
20821
- maxTokens: 131072,
20822
- },
20823
21469
  "zai/glm-4.5": {
20824
21470
  id: "zai/glm-4.5",
20825
21471
  name: "GLM 4.5",
@@ -21033,9 +21679,9 @@ export const MODELS = {
21033
21679
  reasoning: true,
21034
21680
  input: ["text"],
21035
21681
  cost: {
21036
- input: 0.7,
21037
- output: 2.2,
21038
- cacheRead: 0.13,
21682
+ input: 1.4,
21683
+ output: 4.4,
21684
+ cacheRead: 0.14,
21039
21685
  cacheWrite: 0,
21040
21686
  },
21041
21687
  contextWindow: 1000000,
@@ -21075,23 +21721,6 @@ export const MODELS = {
21075
21721
  contextWindow: 1000000,
21076
21722
  maxTokens: 131000,
21077
21723
  },
21078
- "zai/glm-5.3-promo-50": {
21079
- id: "zai/glm-5.3-promo-50",
21080
- name: "GLM 5.3 (50% off)",
21081
- api: "anthropic-messages",
21082
- provider: "vercel-ai-gateway",
21083
- baseUrl: "https://ai-gateway.vercel.sh",
21084
- reasoning: true,
21085
- input: ["text"],
21086
- cost: {
21087
- input: 0.7,
21088
- output: 2.2,
21089
- cacheRead: 0.13,
21090
- cacheWrite: 0,
21091
- },
21092
- contextWindow: 1048576,
21093
- maxTokens: 1048576,
21094
- },
21095
21724
  "zai/glm-5v-turbo": {
21096
21725
  id: "zai/glm-5v-turbo",
21097
21726
  name: "GLM 5V Turbo",