@dreb/ai 2.25.3 → 2.27.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -8,7 +8,7 @@ export const MODELS = {
8
8
  api: "bedrock-converse-stream",
9
9
  provider: "amazon-bedrock",
10
10
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
11
- reasoning: false,
11
+ reasoning: true,
12
12
  input: ["text", "image"],
13
13
  cost: {
14
14
  input: 0.33,
@@ -342,6 +342,23 @@ export const MODELS = {
342
342
  contextWindow: 163840,
343
343
  maxTokens: 81920,
344
344
  },
345
+ "eu.anthropic.claude-fable-5": {
346
+ id: "eu.anthropic.claude-fable-5",
347
+ name: "Claude Fable 5 (EU)",
348
+ api: "bedrock-converse-stream",
349
+ provider: "amazon-bedrock",
350
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
351
+ reasoning: true,
352
+ input: ["text", "image"],
353
+ cost: {
354
+ input: 11,
355
+ output: 55,
356
+ cacheRead: 1.1,
357
+ cacheWrite: 13.75,
358
+ },
359
+ contextWindow: 1000000,
360
+ maxTokens: 128000,
361
+ },
345
362
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
346
363
  id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
347
364
  name: "Claude Haiku 4.5 (EU)",
@@ -461,6 +478,23 @@ export const MODELS = {
461
478
  contextWindow: 1000000,
462
479
  maxTokens: 64000,
463
480
  },
481
+ "global.anthropic.claude-fable-5": {
482
+ id: "global.anthropic.claude-fable-5",
483
+ name: "Claude Fable 5 (Global)",
484
+ api: "bedrock-converse-stream",
485
+ provider: "amazon-bedrock",
486
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
487
+ reasoning: true,
488
+ input: ["text", "image"],
489
+ cost: {
490
+ input: 10,
491
+ output: 50,
492
+ cacheRead: 1,
493
+ cacheWrite: 12.5,
494
+ },
495
+ contextWindow: 1000000,
496
+ maxTokens: 128000,
497
+ },
464
498
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
465
499
  id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
466
500
  name: "Claude Haiku 4.5 (Global)",
@@ -1113,7 +1147,7 @@ export const MODELS = {
1113
1147
  api: "bedrock-converse-stream",
1114
1148
  provider: "amazon-bedrock",
1115
1149
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1116
- reasoning: false,
1150
+ reasoning: true,
1117
1151
  input: ["text"],
1118
1152
  cost: {
1119
1153
  input: 0.15,
@@ -1130,7 +1164,7 @@ export const MODELS = {
1130
1164
  api: "bedrock-converse-stream",
1131
1165
  provider: "amazon-bedrock",
1132
1166
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1133
- reasoning: false,
1167
+ reasoning: true,
1134
1168
  input: ["text"],
1135
1169
  cost: {
1136
1170
  input: 0.15,
@@ -1147,7 +1181,7 @@ export const MODELS = {
1147
1181
  api: "bedrock-converse-stream",
1148
1182
  provider: "amazon-bedrock",
1149
1183
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1150
- reasoning: false,
1184
+ reasoning: true,
1151
1185
  input: ["text"],
1152
1186
  cost: {
1153
1187
  input: 0.07,
@@ -1164,7 +1198,7 @@ export const MODELS = {
1164
1198
  api: "bedrock-converse-stream",
1165
1199
  provider: "amazon-bedrock",
1166
1200
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1167
- reasoning: false,
1201
+ reasoning: true,
1168
1202
  input: ["text"],
1169
1203
  cost: {
1170
1204
  input: 0.07,
@@ -1328,6 +1362,23 @@ export const MODELS = {
1328
1362
  contextWindow: 262000,
1329
1363
  maxTokens: 262000,
1330
1364
  },
1365
+ "us.anthropic.claude-fable-5": {
1366
+ id: "us.anthropic.claude-fable-5",
1367
+ name: "Claude Fable 5 (US)",
1368
+ api: "bedrock-converse-stream",
1369
+ provider: "amazon-bedrock",
1370
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1371
+ reasoning: true,
1372
+ input: ["text", "image"],
1373
+ cost: {
1374
+ input: 10,
1375
+ output: 50,
1376
+ cacheRead: 1,
1377
+ cacheWrite: 12.5,
1378
+ },
1379
+ contextWindow: 1000000,
1380
+ maxTokens: 128000,
1381
+ },
1331
1382
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
1332
1383
  id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
1333
1384
  name: "Claude Haiku 4.5 (US)",
@@ -1738,6 +1789,23 @@ export const MODELS = {
1738
1789
  contextWindow: 200000,
1739
1790
  maxTokens: 4096,
1740
1791
  },
1792
+ "claude-fable-5": {
1793
+ id: "claude-fable-5",
1794
+ name: "Claude Fable 5",
1795
+ api: "anthropic-messages",
1796
+ provider: "anthropic",
1797
+ baseUrl: "https://api.anthropic.com",
1798
+ reasoning: true,
1799
+ input: ["text", "image"],
1800
+ cost: {
1801
+ input: 10,
1802
+ output: 50,
1803
+ cacheRead: 1,
1804
+ cacheWrite: 12.5,
1805
+ },
1806
+ contextWindow: 1000000,
1807
+ maxTokens: 128000,
1808
+ },
1741
1809
  "claude-haiku-4-5": {
1742
1810
  id: "claude-haiku-4-5",
1743
1811
  name: "Claude Haiku 4.5 (latest)",
@@ -3907,77 +3975,9 @@ export const MODELS = {
3907
3975
  },
3908
3976
  },
3909
3977
  "groq": {
3910
- "deepseek-r1-distill-llama-70b": {
3911
- id: "deepseek-r1-distill-llama-70b",
3912
- name: "DeepSeek R1 Distill Llama 70B",
3913
- api: "openai-completions",
3914
- provider: "groq",
3915
- baseUrl: "https://api.groq.com/openai/v1",
3916
- reasoning: true,
3917
- input: ["text"],
3918
- cost: {
3919
- input: 0.75,
3920
- output: 0.99,
3921
- cacheRead: 0,
3922
- cacheWrite: 0,
3923
- },
3924
- contextWindow: 131072,
3925
- maxTokens: 8192,
3926
- },
3927
- "gemma2-9b-it": {
3928
- id: "gemma2-9b-it",
3929
- name: "Gemma 2 9B",
3930
- api: "openai-completions",
3931
- provider: "groq",
3932
- baseUrl: "https://api.groq.com/openai/v1",
3933
- reasoning: false,
3934
- input: ["text"],
3935
- cost: {
3936
- input: 0.2,
3937
- output: 0.2,
3938
- cacheRead: 0,
3939
- cacheWrite: 0,
3940
- },
3941
- contextWindow: 8192,
3942
- maxTokens: 8192,
3943
- },
3944
- "groq/compound": {
3945
- id: "groq/compound",
3946
- name: "Compound",
3947
- api: "openai-completions",
3948
- provider: "groq",
3949
- baseUrl: "https://api.groq.com/openai/v1",
3950
- reasoning: true,
3951
- input: ["text"],
3952
- cost: {
3953
- input: 0,
3954
- output: 0,
3955
- cacheRead: 0,
3956
- cacheWrite: 0,
3957
- },
3958
- contextWindow: 131072,
3959
- maxTokens: 8192,
3960
- },
3961
- "groq/compound-mini": {
3962
- id: "groq/compound-mini",
3963
- name: "Compound Mini",
3964
- api: "openai-completions",
3965
- provider: "groq",
3966
- baseUrl: "https://api.groq.com/openai/v1",
3967
- reasoning: true,
3968
- input: ["text"],
3969
- cost: {
3970
- input: 0,
3971
- output: 0,
3972
- cacheRead: 0,
3973
- cacheWrite: 0,
3974
- },
3975
- contextWindow: 131072,
3976
- maxTokens: 8192,
3977
- },
3978
3978
  "llama-3.1-8b-instant": {
3979
3979
  id: "llama-3.1-8b-instant",
3980
- name: "Llama 3.1 8B Instant",
3980
+ name: "Llama 3.1 8B",
3981
3981
  api: "openai-completions",
3982
3982
  provider: "groq",
3983
3983
  baseUrl: "https://api.groq.com/openai/v1",
@@ -3994,7 +3994,7 @@ export const MODELS = {
3994
3994
  },
3995
3995
  "llama-3.3-70b-versatile": {
3996
3996
  id: "llama-3.3-70b-versatile",
3997
- name: "Llama 3.3 70B Versatile",
3997
+ name: "Llama 3.3 70B",
3998
3998
  api: "openai-completions",
3999
3999
  provider: "groq",
4000
4000
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4009,60 +4009,9 @@ export const MODELS = {
4009
4009
  contextWindow: 131072,
4010
4010
  maxTokens: 32768,
4011
4011
  },
4012
- "llama3-70b-8192": {
4013
- id: "llama3-70b-8192",
4014
- name: "Llama 3 70B",
4015
- api: "openai-completions",
4016
- provider: "groq",
4017
- baseUrl: "https://api.groq.com/openai/v1",
4018
- reasoning: false,
4019
- input: ["text"],
4020
- cost: {
4021
- input: 0.59,
4022
- output: 0.79,
4023
- cacheRead: 0,
4024
- cacheWrite: 0,
4025
- },
4026
- contextWindow: 8192,
4027
- maxTokens: 8192,
4028
- },
4029
- "llama3-8b-8192": {
4030
- id: "llama3-8b-8192",
4031
- name: "Llama 3 8B",
4032
- api: "openai-completions",
4033
- provider: "groq",
4034
- baseUrl: "https://api.groq.com/openai/v1",
4035
- reasoning: false,
4036
- input: ["text"],
4037
- cost: {
4038
- input: 0.05,
4039
- output: 0.08,
4040
- cacheRead: 0,
4041
- cacheWrite: 0,
4042
- },
4043
- contextWindow: 8192,
4044
- maxTokens: 8192,
4045
- },
4046
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
4047
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
4048
- name: "Llama 4 Maverick 17B",
4049
- api: "openai-completions",
4050
- provider: "groq",
4051
- baseUrl: "https://api.groq.com/openai/v1",
4052
- reasoning: false,
4053
- input: ["text", "image"],
4054
- cost: {
4055
- input: 0.2,
4056
- output: 0.6,
4057
- cacheRead: 0,
4058
- cacheWrite: 0,
4059
- },
4060
- contextWindow: 131072,
4061
- maxTokens: 8192,
4062
- },
4063
4012
  "meta-llama/llama-4-scout-17b-16e-instruct": {
4064
4013
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
4065
- name: "Llama 4 Scout 17B",
4014
+ name: "Llama 4 Scout 17B 16E",
4066
4015
  api: "openai-completions",
4067
4016
  provider: "groq",
4068
4017
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4077,57 +4026,6 @@ export const MODELS = {
4077
4026
  contextWindow: 131072,
4078
4027
  maxTokens: 8192,
4079
4028
  },
4080
- "mistral-saba-24b": {
4081
- id: "mistral-saba-24b",
4082
- name: "Mistral Saba 24B",
4083
- api: "openai-completions",
4084
- provider: "groq",
4085
- baseUrl: "https://api.groq.com/openai/v1",
4086
- reasoning: false,
4087
- input: ["text"],
4088
- cost: {
4089
- input: 0.79,
4090
- output: 0.79,
4091
- cacheRead: 0,
4092
- cacheWrite: 0,
4093
- },
4094
- contextWindow: 32768,
4095
- maxTokens: 32768,
4096
- },
4097
- "moonshotai/kimi-k2-instruct": {
4098
- id: "moonshotai/kimi-k2-instruct",
4099
- name: "Kimi K2 Instruct",
4100
- api: "openai-completions",
4101
- provider: "groq",
4102
- baseUrl: "https://api.groq.com/openai/v1",
4103
- reasoning: false,
4104
- input: ["text"],
4105
- cost: {
4106
- input: 1,
4107
- output: 3,
4108
- cacheRead: 0,
4109
- cacheWrite: 0,
4110
- },
4111
- contextWindow: 131072,
4112
- maxTokens: 16384,
4113
- },
4114
- "moonshotai/kimi-k2-instruct-0905": {
4115
- id: "moonshotai/kimi-k2-instruct-0905",
4116
- name: "Kimi K2 Instruct 0905",
4117
- api: "openai-completions",
4118
- provider: "groq",
4119
- baseUrl: "https://api.groq.com/openai/v1",
4120
- reasoning: false,
4121
- input: ["text"],
4122
- cost: {
4123
- input: 1,
4124
- output: 3,
4125
- cacheRead: 0.5,
4126
- cacheWrite: 0,
4127
- },
4128
- contextWindow: 262144,
4129
- maxTokens: 16384,
4130
- },
4131
4029
  "openai/gpt-oss-120b": {
4132
4030
  id: "openai/gpt-oss-120b",
4133
4031
  name: "GPT OSS 120B",
@@ -4179,26 +4077,9 @@ export const MODELS = {
4179
4077
  contextWindow: 131072,
4180
4078
  maxTokens: 65536,
4181
4079
  },
4182
- "qwen-qwq-32b": {
4183
- id: "qwen-qwq-32b",
4184
- name: "Qwen QwQ 32B",
4185
- api: "openai-completions",
4186
- provider: "groq",
4187
- baseUrl: "https://api.groq.com/openai/v1",
4188
- reasoning: true,
4189
- input: ["text"],
4190
- cost: {
4191
- input: 0.29,
4192
- output: 0.39,
4193
- cacheRead: 0,
4194
- cacheWrite: 0,
4195
- },
4196
- contextWindow: 131072,
4197
- maxTokens: 16384,
4198
- },
4199
4080
  "qwen/qwen3-32b": {
4200
4081
  id: "qwen/qwen3-32b",
4201
- name: "Qwen3 32B",
4082
+ name: "Qwen3-32B",
4202
4083
  api: "openai-completions",
4203
4084
  provider: "groq",
4204
4085
  baseUrl: "https://api.groq.com/openai/v1",
@@ -6124,6 +6005,23 @@ export const MODELS = {
6124
6005
  contextWindow: 200000,
6125
6006
  maxTokens: 32000,
6126
6007
  },
6008
+ "claude-fable-5": {
6009
+ id: "claude-fable-5",
6010
+ name: "Claude Fable 5",
6011
+ api: "anthropic-messages",
6012
+ provider: "opencode",
6013
+ baseUrl: "https://opencode.ai/zen",
6014
+ reasoning: true,
6015
+ input: ["text", "image"],
6016
+ cost: {
6017
+ input: 10,
6018
+ output: 50,
6019
+ cacheRead: 1,
6020
+ cacheWrite: 12.5,
6021
+ },
6022
+ contextWindow: 1000000,
6023
+ maxTokens: 128000,
6024
+ },
6127
6025
  "claude-haiku-4-5": {
6128
6026
  id: "claude-haiku-4-5",
6129
6027
  name: "Claude Haiku 4.5",
@@ -6288,7 +6186,7 @@ export const MODELS = {
6288
6186
  cost: {
6289
6187
  input: 0.14,
6290
6188
  output: 0.28,
6291
- cacheRead: 0.03,
6189
+ cacheRead: 0.028,
6292
6190
  cacheWrite: 0,
6293
6191
  },
6294
6192
  contextWindow: 1000000,
@@ -6311,6 +6209,23 @@ export const MODELS = {
6311
6209
  contextWindow: 200000,
6312
6210
  maxTokens: 128000,
6313
6211
  },
6212
+ "deepseek-v4-pro": {
6213
+ id: "deepseek-v4-pro",
6214
+ name: "DeepSeek V4 Pro",
6215
+ api: "openai-completions",
6216
+ provider: "opencode",
6217
+ baseUrl: "https://opencode.ai/zen/v1",
6218
+ reasoning: true,
6219
+ input: ["text"],
6220
+ cost: {
6221
+ input: 1.74,
6222
+ output: 3.84,
6223
+ cacheRead: 0.145,
6224
+ cacheWrite: 0,
6225
+ },
6226
+ contextWindow: 1000000,
6227
+ maxTokens: 384000,
6228
+ },
6314
6229
  "gemini-3-flash": {
6315
6230
  id: "gemini-3-flash",
6316
6231
  name: "Gemini 3 Flash",
@@ -6770,26 +6685,26 @@ export const MODELS = {
6770
6685
  contextWindow: 204800,
6771
6686
  maxTokens: 131072,
6772
6687
  },
6773
- "minimax-m3-free": {
6774
- id: "minimax-m3-free",
6775
- name: "MiniMax M3 Free",
6776
- api: "anthropic-messages",
6688
+ "nemotron-3-ultra-free": {
6689
+ id: "nemotron-3-ultra-free",
6690
+ name: "Nemotron 3 Ultra Free",
6691
+ api: "openai-completions",
6777
6692
  provider: "opencode",
6778
- baseUrl: "https://opencode.ai/zen",
6693
+ baseUrl: "https://opencode.ai/zen/v1",
6779
6694
  reasoning: true,
6780
- input: ["text", "image"],
6695
+ input: ["text"],
6781
6696
  cost: {
6782
6697
  input: 0,
6783
6698
  output: 0,
6784
6699
  cacheRead: 0,
6785
6700
  cacheWrite: 0,
6786
6701
  },
6787
- contextWindow: 200000,
6788
- maxTokens: 32000,
6702
+ contextWindow: 1000000,
6703
+ maxTokens: 128000,
6789
6704
  },
6790
- "nemotron-3-ultra-free": {
6791
- id: "nemotron-3-ultra-free",
6792
- name: "Nemotron 3 Ultra Free",
6705
+ "north-mini-code-free": {
6706
+ id: "north-mini-code-free",
6707
+ name: "North Mini Code Free",
6793
6708
  api: "openai-completions",
6794
6709
  provider: "opencode",
6795
6710
  baseUrl: "https://opencode.ai/zen/v1",
@@ -6801,8 +6716,8 @@ export const MODELS = {
6801
6716
  cacheRead: 0,
6802
6717
  cacheWrite: 0,
6803
6718
  },
6804
- contextWindow: 1000000,
6805
- maxTokens: 128000,
6719
+ contextWindow: 256000,
6720
+ maxTokens: 64000,
6806
6721
  },
6807
6722
  "qwen3.5-plus": {
6808
6723
  id: "qwen3.5-plus",
@@ -7019,9 +6934,9 @@ export const MODELS = {
7019
6934
  reasoning: true,
7020
6935
  input: ["text", "image"],
7021
6936
  cost: {
7022
- input: 0.6,
7023
- output: 2.4,
7024
- cacheRead: 0.12,
6937
+ input: 0.3,
6938
+ output: 1.2,
6939
+ cacheRead: 0.06,
7025
6940
  cacheWrite: 0,
7026
6941
  },
7027
6942
  contextWindow: 512000,
@@ -7216,6 +7131,23 @@ export const MODELS = {
7216
7131
  contextWindow: 200000,
7217
7132
  maxTokens: 8192,
7218
7133
  },
7134
+ "anthropic/claude-fable-5": {
7135
+ id: "anthropic/claude-fable-5",
7136
+ name: "Anthropic: Claude Fable 5",
7137
+ api: "openai-completions",
7138
+ provider: "openrouter",
7139
+ baseUrl: "https://openrouter.ai/api/v1",
7140
+ reasoning: true,
7141
+ input: ["text", "image"],
7142
+ cost: {
7143
+ input: 10,
7144
+ output: 50,
7145
+ cacheRead: 1,
7146
+ cacheWrite: 12.5,
7147
+ },
7148
+ contextWindow: 1000000,
7149
+ maxTokens: 128000,
7150
+ },
7219
7151
  "anthropic/claude-haiku-4.5": {
7220
7152
  id: "anthropic/claude-haiku-4.5",
7221
7153
  name: "Anthropic: Claude Haiku 4.5",
@@ -7505,23 +7437,6 @@ export const MODELS = {
7505
7437
  contextWindow: 2000000,
7506
7438
  maxTokens: 30000,
7507
7439
  },
7508
- "baidu/ernie-4.5-vl-28b-a3b": {
7509
- id: "baidu/ernie-4.5-vl-28b-a3b",
7510
- name: "Baidu: ERNIE 4.5 VL 28B A3B",
7511
- api: "openai-completions",
7512
- provider: "openrouter",
7513
- baseUrl: "https://openrouter.ai/api/v1",
7514
- reasoning: true,
7515
- input: ["text", "image"],
7516
- cost: {
7517
- input: 0.14,
7518
- output: 0.56,
7519
- cacheRead: 0,
7520
- cacheWrite: 0,
7521
- },
7522
- contextWindow: 131072,
7523
- maxTokens: 8000,
7524
- },
7525
7440
  "bytedance-seed/seed-1.6": {
7526
7441
  id: "bytedance-seed/seed-1.6",
7527
7442
  name: "ByteDance Seed: Seed 1.6",
@@ -7655,7 +7570,7 @@ export const MODELS = {
7655
7570
  cacheRead: 0.135,
7656
7571
  cacheWrite: 0,
7657
7572
  },
7658
- contextWindow: 163840,
7573
+ contextWindow: 131072,
7659
7574
  maxTokens: 16384,
7660
7575
  },
7661
7576
  "deepseek/deepseek-chat-v3.1": {
@@ -8024,8 +7939,8 @@ export const MODELS = {
8024
7939
  reasoning: false,
8025
7940
  input: ["text", "image"],
8026
7941
  cost: {
8027
- input: 0.04,
8028
- output: 0.13,
7942
+ input: 0.049999999999999996,
7943
+ output: 0.15,
8029
7944
  cacheRead: 0,
8030
7945
  cacheWrite: 0,
8031
7946
  },
@@ -8313,7 +8228,7 @@ export const MODELS = {
8313
8228
  reasoning: false,
8314
8229
  input: ["text", "image"],
8315
8230
  cost: {
8316
- input: 0.08,
8231
+ input: 0.09999999999999999,
8317
8232
  output: 0.3,
8318
8233
  cacheRead: 0,
8319
8234
  cacheWrite: 0,
@@ -8382,8 +8297,8 @@ export const MODELS = {
8382
8297
  input: ["text"],
8383
8298
  cost: {
8384
8299
  input: 0.15,
8385
- output: 1.15,
8386
- cacheRead: 0,
8300
+ output: 0.8999999999999999,
8301
+ cacheRead: 0.049999999999999996,
8387
8302
  cacheWrite: 0,
8388
8303
  },
8389
8304
  contextWindow: 204800,
@@ -8398,13 +8313,13 @@ export const MODELS = {
8398
8313
  reasoning: true,
8399
8314
  input: ["text"],
8400
8315
  cost: {
8401
- input: 0.27899999999999997,
8402
- output: 1.2,
8403
- cacheRead: 0,
8316
+ input: 0.27,
8317
+ output: 1.08,
8318
+ cacheRead: 0.054,
8404
8319
  cacheWrite: 0,
8405
8320
  },
8406
8321
  contextWindow: 204800,
8407
- maxTokens: 196608,
8322
+ maxTokens: 131072,
8408
8323
  },
8409
8324
  "minimax/minimax-m3": {
8410
8325
  id: "minimax/minimax-m3",
@@ -8789,17 +8704,17 @@ export const MODELS = {
8789
8704
  reasoning: true,
8790
8705
  input: ["text", "image"],
8791
8706
  cost: {
8792
- input: 0.684,
8793
- output: 3.42,
8794
- cacheRead: 0.144,
8707
+ input: 0.6799999999999999,
8708
+ output: 3.41,
8709
+ cacheRead: 0.33999999999999997,
8795
8710
  cacheWrite: 0,
8796
8711
  },
8797
8712
  contextWindow: 262144,
8798
- maxTokens: 262144,
8713
+ maxTokens: 262142,
8799
8714
  },
8800
- "moonshotai/kimi-k2.6:free": {
8801
- id: "moonshotai/kimi-k2.6:free",
8802
- name: "MoonshotAI: Kimi K2.6 (free)",
8715
+ "nex-agi/nex-n2-pro:free": {
8716
+ id: "nex-agi/nex-n2-pro:free",
8717
+ name: "Nex AGI: Nex-N2-Pro (free)",
8803
8718
  api: "openai-completions",
8804
8719
  provider: "openrouter",
8805
8720
  baseUrl: "https://openrouter.ai/api/v1",
@@ -8812,24 +8727,7 @@ export const MODELS = {
8812
8727
  cacheWrite: 0,
8813
8728
  },
8814
8729
  contextWindow: 262144,
8815
- maxTokens: 4096,
8816
- },
8817
- "nex-agi/deepseek-v3.1-nex-n1": {
8818
- id: "nex-agi/deepseek-v3.1-nex-n1",
8819
- name: "Nex AGI: DeepSeek V3.1 Nex N1",
8820
- api: "openai-completions",
8821
- provider: "openrouter",
8822
- baseUrl: "https://openrouter.ai/api/v1",
8823
- reasoning: false,
8824
- input: ["text"],
8825
- cost: {
8826
- input: 0.135,
8827
- output: 0.5,
8828
- cacheRead: 0,
8829
- cacheWrite: 0,
8830
- },
8831
- contextWindow: 131072,
8832
- maxTokens: 163840,
8730
+ maxTokens: 262144,
8833
8731
  },
8834
8732
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
8835
8733
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -8840,7 +8738,7 @@ export const MODELS = {
8840
8738
  reasoning: true,
8841
8739
  input: ["text"],
8842
8740
  cost: {
8843
- input: 0.09999999999999999,
8741
+ input: 0.39999999999999997,
8844
8742
  output: 0.39999999999999997,
8845
8743
  cacheRead: 0,
8846
8744
  cacheWrite: 0,
@@ -9086,23 +8984,6 @@ export const MODELS = {
9086
8984
  contextWindow: 8191,
9087
8985
  maxTokens: 4096,
9088
8986
  },
9089
- "openai/gpt-4-1106-preview": {
9090
- id: "openai/gpt-4-1106-preview",
9091
- name: "OpenAI: GPT-4 Turbo (older v1106)",
9092
- api: "openai-completions",
9093
- provider: "openrouter",
9094
- baseUrl: "https://openrouter.ai/api/v1",
9095
- reasoning: false,
9096
- input: ["text"],
9097
- cost: {
9098
- input: 10,
9099
- output: 30,
9100
- cacheRead: 0,
9101
- cacheWrite: 0,
9102
- },
9103
- contextWindow: 128000,
9104
- maxTokens: 4096,
9105
- },
9106
8987
  "openai/gpt-4-turbo": {
9107
8988
  id: "openai/gpt-4-turbo",
9108
8989
  name: "OpenAI: GPT-4 Turbo",
@@ -10166,7 +10047,7 @@ export const MODELS = {
10166
10047
  reasoning: false,
10167
10048
  input: ["text"],
10168
10049
  cost: {
10169
- input: 0.071,
10050
+ input: 0.09,
10170
10051
  output: 0.09999999999999999,
10171
10052
  cacheRead: 0,
10172
10053
  cacheWrite: 0,
@@ -10200,8 +10081,8 @@ export const MODELS = {
10200
10081
  reasoning: true,
10201
10082
  input: ["text"],
10202
10083
  cost: {
10203
- input: 0.09,
10204
- output: 0.44999999999999996,
10084
+ input: 0.12,
10085
+ output: 0.5,
10205
10086
  cacheRead: 0,
10206
10087
  cacheWrite: 0,
10207
10088
  },
@@ -10659,13 +10540,13 @@ export const MODELS = {
10659
10540
  reasoning: true,
10660
10541
  input: ["text", "image"],
10661
10542
  cost: {
10662
- input: 0.04,
10543
+ input: 0.09999999999999999,
10663
10544
  output: 0.15,
10664
10545
  cacheRead: 0,
10665
10546
  cacheWrite: 0,
10666
10547
  },
10667
10548
  contextWindow: 262144,
10668
- maxTokens: 81920,
10549
+ maxTokens: 262144,
10669
10550
  },
10670
10551
  "qwen/qwen3.5-flash-02-23": {
10671
10552
  id: "qwen/qwen3.5-flash-02-23",
@@ -10727,13 +10608,13 @@ export const MODELS = {
10727
10608
  reasoning: true,
10728
10609
  input: ["text", "image"],
10729
10610
  cost: {
10730
- input: 0.29,
10731
- output: 3.1999999999999997,
10611
+ input: 0.28900000000000003,
10612
+ output: 2.4,
10732
10613
  cacheRead: 0,
10733
10614
  cacheWrite: 0,
10734
10615
  },
10735
10616
  contextWindow: 262144,
10736
- maxTokens: 262140,
10617
+ maxTokens: 131072,
10737
10618
  },
10738
10619
  "qwen/qwen3.6-35b-a3b": {
10739
10620
  id: "qwen/qwen3.6-35b-a3b",
@@ -11092,23 +10973,6 @@ export const MODELS = {
11092
10973
  contextWindow: 1048576,
11093
10974
  maxTokens: 131072,
11094
10975
  },
11095
- "z-ai/glm-4-32b": {
11096
- id: "z-ai/glm-4-32b",
11097
- name: "Z.ai: GLM 4 32B ",
11098
- api: "openai-completions",
11099
- provider: "openrouter",
11100
- baseUrl: "https://openrouter.ai/api/v1",
11101
- reasoning: false,
11102
- input: ["text"],
11103
- cost: {
11104
- input: 0.09999999999999999,
11105
- output: 0.09999999999999999,
11106
- cacheRead: 0,
11107
- cacheWrite: 0,
11108
- },
11109
- contextWindow: 128000,
11110
- maxTokens: 4096,
11111
- },
11112
10976
  "z-ai/glm-4.5": {
11113
10977
  id: "z-ai/glm-4.5",
11114
10978
  name: "Z.ai: GLM 4.5",
@@ -11143,23 +11007,6 @@ export const MODELS = {
11143
11007
  contextWindow: 131072,
11144
11008
  maxTokens: 131070,
11145
11009
  },
11146
- "z-ai/glm-4.5-air:free": {
11147
- id: "z-ai/glm-4.5-air:free",
11148
- name: "Z.ai: GLM 4.5 Air (free)",
11149
- api: "openai-completions",
11150
- provider: "openrouter",
11151
- baseUrl: "https://openrouter.ai/api/v1",
11152
- reasoning: true,
11153
- input: ["text"],
11154
- cost: {
11155
- input: 0,
11156
- output: 0,
11157
- cacheRead: 0,
11158
- cacheWrite: 0,
11159
- },
11160
- contextWindow: 131072,
11161
- maxTokens: 96000,
11162
- },
11163
11010
  "z-ai/glm-4.5v": {
11164
11011
  id: "z-ai/glm-4.5v",
11165
11012
  name: "Z.ai: GLM 4.5V",
@@ -11205,11 +11052,11 @@ export const MODELS = {
11205
11052
  cost: {
11206
11053
  input: 0.3,
11207
11054
  output: 0.8999999999999999,
11208
- cacheRead: 0.049999999999999996,
11055
+ cacheRead: 0.055,
11209
11056
  cacheWrite: 0,
11210
11057
  },
11211
11058
  contextWindow: 131072,
11212
- maxTokens: 24000,
11059
+ maxTokens: 32768,
11213
11060
  },
11214
11061
  "z-ai/glm-4.7": {
11215
11062
  id: "z-ai/glm-4.7",
@@ -11276,7 +11123,7 @@ export const MODELS = {
11276
11123
  cacheRead: 0.24,
11277
11124
  cacheWrite: 0,
11278
11125
  },
11279
- contextWindow: 202752,
11126
+ contextWindow: 262144,
11280
11127
  maxTokens: 131072,
11281
11128
  },
11282
11129
  "z-ai/glm-5.1": {
@@ -11296,22 +11143,22 @@ export const MODELS = {
11296
11143
  contextWindow: 202752,
11297
11144
  maxTokens: 4096,
11298
11145
  },
11299
- "z-ai/glm-5v-turbo": {
11300
- id: "z-ai/glm-5v-turbo",
11301
- name: "Z.ai: GLM 5V Turbo",
11146
+ "~anthropic/claude-fable-latest": {
11147
+ id: "~anthropic/claude-fable-latest",
11148
+ name: "Anthropic: Claude Fable Latest",
11302
11149
  api: "openai-completions",
11303
11150
  provider: "openrouter",
11304
11151
  baseUrl: "https://openrouter.ai/api/v1",
11305
11152
  reasoning: true,
11306
11153
  input: ["text", "image"],
11307
11154
  cost: {
11308
- input: 1.2,
11309
- output: 4,
11310
- cacheRead: 0.24,
11311
- cacheWrite: 0,
11155
+ input: 10,
11156
+ output: 50,
11157
+ cacheRead: 1,
11158
+ cacheWrite: 12.5,
11312
11159
  },
11313
- contextWindow: 202752,
11314
- maxTokens: 131072,
11160
+ contextWindow: 1000000,
11161
+ maxTokens: 128000,
11315
11162
  },
11316
11163
  "~anthropic/claude-haiku-latest": {
11317
11164
  id: "~anthropic/claude-haiku-latest",
@@ -11407,13 +11254,13 @@ export const MODELS = {
11407
11254
  reasoning: true,
11408
11255
  input: ["text", "image"],
11409
11256
  cost: {
11410
- input: 0.684,
11411
- output: 3.42,
11412
- cacheRead: 0.144,
11257
+ input: 0.6799999999999999,
11258
+ output: 3.41,
11259
+ cacheRead: 0.33999999999999997,
11413
11260
  cacheWrite: 0,
11414
11261
  },
11415
11262
  contextWindow: 262144,
11416
- maxTokens: 262144,
11263
+ maxTokens: 262142,
11417
11264
  },
11418
11265
  "~openai/gpt-latest": {
11419
11266
  id: "~openai/gpt-latest",
@@ -11494,8 +11341,8 @@ export const MODELS = {
11494
11341
  reasoning: true,
11495
11342
  input: ["text"],
11496
11343
  cost: {
11497
- input: 0.08,
11498
- output: 0.29,
11344
+ input: 0.12,
11345
+ output: 0.5,
11499
11346
  cacheRead: 0,
11500
11347
  cacheWrite: 0,
11501
11348
  },
@@ -11859,6 +11706,23 @@ export const MODELS = {
11859
11706
  contextWindow: 200000,
11860
11707
  maxTokens: 8192,
11861
11708
  },
11709
+ "anthropic/claude-fable-5": {
11710
+ id: "anthropic/claude-fable-5",
11711
+ name: "Claude Fable 5",
11712
+ api: "anthropic-messages",
11713
+ provider: "vercel-ai-gateway",
11714
+ baseUrl: "https://ai-gateway.vercel.sh",
11715
+ reasoning: true,
11716
+ input: ["text", "image"],
11717
+ cost: {
11718
+ input: 10,
11719
+ output: 50,
11720
+ cacheRead: 1,
11721
+ cacheWrite: 12.5,
11722
+ },
11723
+ contextWindow: 1000000,
11724
+ maxTokens: 128000,
11725
+ },
11862
11726
  "anthropic/claude-haiku-4.5": {
11863
11727
  id: "anthropic/claude-haiku-4.5",
11864
11728
  name: "Claude Haiku 4.5",
@@ -12233,40 +12097,6 @@ export const MODELS = {
12233
12097
  contextWindow: 1000000,
12234
12098
  maxTokens: 384000,
12235
12099
  },
12236
- "google/gemini-2.0-flash": {
12237
- id: "google/gemini-2.0-flash",
12238
- name: "Gemini 2.0 Flash",
12239
- api: "anthropic-messages",
12240
- provider: "vercel-ai-gateway",
12241
- baseUrl: "https://ai-gateway.vercel.sh",
12242
- reasoning: false,
12243
- input: ["text", "image"],
12244
- cost: {
12245
- input: 0.15,
12246
- output: 0.6,
12247
- cacheRead: 0.024999999999999998,
12248
- cacheWrite: 0,
12249
- },
12250
- contextWindow: 1048576,
12251
- maxTokens: 8192,
12252
- },
12253
- "google/gemini-2.0-flash-lite": {
12254
- id: "google/gemini-2.0-flash-lite",
12255
- name: "Gemini 2.0 Flash Lite",
12256
- api: "anthropic-messages",
12257
- provider: "vercel-ai-gateway",
12258
- baseUrl: "https://ai-gateway.vercel.sh",
12259
- reasoning: false,
12260
- input: ["text", "image"],
12261
- cost: {
12262
- input: 0.075,
12263
- output: 0.3,
12264
- cacheRead: 0.02,
12265
- cacheWrite: 0,
12266
- },
12267
- contextWindow: 1048576,
12268
- maxTokens: 8192,
12269
- },
12270
12100
  "google/gemini-2.5-flash": {
12271
12101
  id: "google/gemini-2.5-flash",
12272
12102
  name: "Gemini 2.5 Flash",
@@ -14340,7 +14170,7 @@ export const MODELS = {
14340
14170
  cacheRead: 0.2,
14341
14171
  cacheWrite: 0,
14342
14172
  },
14343
- contextWindow: 2000000,
14173
+ contextWindow: 1000000,
14344
14174
  maxTokens: 30000,
14345
14175
  },
14346
14176
  "grok-4.20-0309-reasoning": {
@@ -14357,7 +14187,7 @@ export const MODELS = {
14357
14187
  cacheRead: 0.2,
14358
14188
  cacheWrite: 0,
14359
14189
  },
14360
- contextWindow: 2000000,
14190
+ contextWindow: 1000000,
14361
14191
  maxTokens: 30000,
14362
14192
  },
14363
14193
  "grok-4.3": {