@dreb/ai 2.25.4 → 2.27.3

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -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",
@@ -6787,6 +6702,23 @@ export const MODELS = {
6787
6702
  contextWindow: 1000000,
6788
6703
  maxTokens: 128000,
6789
6704
  },
6705
+ "north-mini-code-free": {
6706
+ id: "north-mini-code-free",
6707
+ name: "North Mini Code Free",
6708
+ api: "openai-completions",
6709
+ provider: "opencode",
6710
+ baseUrl: "https://opencode.ai/zen/v1",
6711
+ reasoning: true,
6712
+ input: ["text"],
6713
+ cost: {
6714
+ input: 0,
6715
+ output: 0,
6716
+ cacheRead: 0,
6717
+ cacheWrite: 0,
6718
+ },
6719
+ contextWindow: 256000,
6720
+ maxTokens: 64000,
6721
+ },
6790
6722
  "qwen3.5-plus": {
6791
6723
  id: "qwen3.5-plus",
6792
6724
  name: "Qwen3.5 Plus",
@@ -7199,6 +7131,23 @@ export const MODELS = {
7199
7131
  contextWindow: 200000,
7200
7132
  maxTokens: 8192,
7201
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
+ },
7202
7151
  "anthropic/claude-haiku-4.5": {
7203
7152
  id: "anthropic/claude-haiku-4.5",
7204
7153
  name: "Anthropic: Claude Haiku 4.5",
@@ -7621,7 +7570,7 @@ export const MODELS = {
7621
7570
  cacheRead: 0.135,
7622
7571
  cacheWrite: 0,
7623
7572
  },
7624
- contextWindow: 163840,
7573
+ contextWindow: 131072,
7625
7574
  maxTokens: 16384,
7626
7575
  },
7627
7576
  "deepseek/deepseek-chat-v3.1": {
@@ -8059,12 +8008,12 @@ export const MODELS = {
8059
8008
  input: ["text", "image"],
8060
8009
  cost: {
8061
8010
  input: 0.12,
8062
- output: 0.36,
8011
+ output: 0.35,
8063
8012
  cacheRead: 0.09,
8064
8013
  cacheWrite: 0,
8065
8014
  },
8066
8015
  contextWindow: 262144,
8067
- maxTokens: 8192,
8016
+ maxTokens: 262144,
8068
8017
  },
8069
8018
  "google/gemma-4-31b-it:free": {
8070
8019
  id: "google/gemma-4-31b-it:free",
@@ -8348,8 +8297,8 @@ export const MODELS = {
8348
8297
  input: ["text"],
8349
8298
  cost: {
8350
8299
  input: 0.15,
8351
- output: 1.15,
8352
- cacheRead: 0,
8300
+ output: 0.8999999999999999,
8301
+ cacheRead: 0.049999999999999996,
8353
8302
  cacheWrite: 0,
8354
8303
  },
8355
8304
  contextWindow: 204800,
@@ -8364,13 +8313,13 @@ export const MODELS = {
8364
8313
  reasoning: true,
8365
8314
  input: ["text"],
8366
8315
  cost: {
8367
- input: 0.27899999999999997,
8368
- output: 1.2,
8369
- cacheRead: 0,
8316
+ input: 0.27,
8317
+ output: 1.08,
8318
+ cacheRead: 0.054,
8370
8319
  cacheWrite: 0,
8371
8320
  },
8372
8321
  contextWindow: 204800,
8373
- maxTokens: 196608,
8322
+ maxTokens: 131072,
8374
8323
  },
8375
8324
  "minimax/minimax-m3": {
8376
8325
  id: "minimax/minimax-m3",
@@ -8763,9 +8712,9 @@ export const MODELS = {
8763
8712
  contextWindow: 262144,
8764
8713
  maxTokens: 262142,
8765
8714
  },
8766
- "moonshotai/kimi-k2.6:free": {
8767
- id: "moonshotai/kimi-k2.6:free",
8768
- 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)",
8769
8718
  api: "openai-completions",
8770
8719
  provider: "openrouter",
8771
8720
  baseUrl: "https://openrouter.ai/api/v1",
@@ -8778,7 +8727,7 @@ export const MODELS = {
8778
8727
  cacheWrite: 0,
8779
8728
  },
8780
8729
  contextWindow: 262144,
8781
- maxTokens: 4096,
8730
+ maxTokens: 262144,
8782
8731
  },
8783
8732
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
8784
8733
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -8933,23 +8882,6 @@ export const MODELS = {
8933
8882
  contextWindow: 128000,
8934
8883
  maxTokens: 128000,
8935
8884
  },
8936
- "nvidia/nemotron-nano-9b-v2": {
8937
- id: "nvidia/nemotron-nano-9b-v2",
8938
- name: "NVIDIA: Nemotron Nano 9B V2",
8939
- api: "openai-completions",
8940
- provider: "openrouter",
8941
- baseUrl: "https://openrouter.ai/api/v1",
8942
- reasoning: true,
8943
- input: ["text"],
8944
- cost: {
8945
- input: 0.04,
8946
- output: 0.16,
8947
- cacheRead: 0,
8948
- cacheWrite: 0,
8949
- },
8950
- contextWindow: 131072,
8951
- maxTokens: 16384,
8952
- },
8953
8885
  "nvidia/nemotron-nano-9b-v2:free": {
8954
8886
  id: "nvidia/nemotron-nano-9b-v2:free",
8955
8887
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -10676,13 +10608,13 @@ export const MODELS = {
10676
10608
  reasoning: true,
10677
10609
  input: ["text", "image"],
10678
10610
  cost: {
10679
- input: 0.14,
10611
+ input: 0.15,
10680
10612
  output: 1,
10681
- cacheRead: 0,
10613
+ cacheRead: 0.049999999999999996,
10682
10614
  cacheWrite: 0,
10683
10615
  },
10684
10616
  contextWindow: 262144,
10685
- maxTokens: 262140,
10617
+ maxTokens: 262144,
10686
10618
  },
10687
10619
  "qwen/qwen3.6-flash": {
10688
10620
  id: "qwen/qwen3.6-flash",
@@ -10761,10 +10693,10 @@ export const MODELS = {
10761
10693
  reasoning: true,
10762
10694
  input: ["text", "image"],
10763
10695
  cost: {
10764
- input: 0.39999999999999997,
10765
- output: 1.5999999999999999,
10766
- cacheRead: 0.08,
10767
- cacheWrite: 0.5,
10696
+ input: 0.32,
10697
+ output: 1.28,
10698
+ cacheRead: 0.064,
10699
+ cacheWrite: 0.39999999999999997,
10768
10700
  },
10769
10701
  contextWindow: 1000000,
10770
10702
  maxTokens: 65536,
@@ -11024,23 +10956,6 @@ export const MODELS = {
11024
10956
  contextWindow: 1048576,
11025
10957
  maxTokens: 131072,
11026
10958
  },
11027
- "z-ai/glm-4-32b": {
11028
- id: "z-ai/glm-4-32b",
11029
- name: "Z.ai: GLM 4 32B ",
11030
- api: "openai-completions",
11031
- provider: "openrouter",
11032
- baseUrl: "https://openrouter.ai/api/v1",
11033
- reasoning: false,
11034
- input: ["text"],
11035
- cost: {
11036
- input: 0.09999999999999999,
11037
- output: 0.09999999999999999,
11038
- cacheRead: 0,
11039
- cacheWrite: 0,
11040
- },
11041
- contextWindow: 128000,
11042
- maxTokens: 4096,
11043
- },
11044
10959
  "z-ai/glm-4.5": {
11045
10960
  id: "z-ai/glm-4.5",
11046
10961
  name: "Z.ai: GLM 4.5",
@@ -11075,23 +10990,6 @@ export const MODELS = {
11075
10990
  contextWindow: 131072,
11076
10991
  maxTokens: 131070,
11077
10992
  },
11078
- "z-ai/glm-4.5-air:free": {
11079
- id: "z-ai/glm-4.5-air:free",
11080
- name: "Z.ai: GLM 4.5 Air (free)",
11081
- api: "openai-completions",
11082
- provider: "openrouter",
11083
- baseUrl: "https://openrouter.ai/api/v1",
11084
- reasoning: true,
11085
- input: ["text"],
11086
- cost: {
11087
- input: 0,
11088
- output: 0,
11089
- cacheRead: 0,
11090
- cacheWrite: 0,
11091
- },
11092
- contextWindow: 131072,
11093
- maxTokens: 96000,
11094
- },
11095
10993
  "z-ai/glm-4.5v": {
11096
10994
  id: "z-ai/glm-4.5v",
11097
10995
  name: "Z.ai: GLM 4.5V",
@@ -11137,11 +11035,11 @@ export const MODELS = {
11137
11035
  cost: {
11138
11036
  input: 0.3,
11139
11037
  output: 0.8999999999999999,
11140
- cacheRead: 0.049999999999999996,
11038
+ cacheRead: 0.055,
11141
11039
  cacheWrite: 0,
11142
11040
  },
11143
11041
  contextWindow: 131072,
11144
- maxTokens: 24000,
11042
+ maxTokens: 32768,
11145
11043
  },
11146
11044
  "z-ai/glm-4.7": {
11147
11045
  id: "z-ai/glm-4.7",
@@ -11208,7 +11106,7 @@ export const MODELS = {
11208
11106
  cacheRead: 0.24,
11209
11107
  cacheWrite: 0,
11210
11108
  },
11211
- contextWindow: 202752,
11109
+ contextWindow: 262144,
11212
11110
  maxTokens: 131072,
11213
11111
  },
11214
11112
  "z-ai/glm-5.1": {
@@ -11228,22 +11126,22 @@ export const MODELS = {
11228
11126
  contextWindow: 202752,
11229
11127
  maxTokens: 4096,
11230
11128
  },
11231
- "z-ai/glm-5v-turbo": {
11232
- id: "z-ai/glm-5v-turbo",
11233
- name: "Z.ai: GLM 5V Turbo",
11129
+ "~anthropic/claude-fable-latest": {
11130
+ id: "~anthropic/claude-fable-latest",
11131
+ name: "Anthropic: Claude Fable Latest",
11234
11132
  api: "openai-completions",
11235
11133
  provider: "openrouter",
11236
11134
  baseUrl: "https://openrouter.ai/api/v1",
11237
11135
  reasoning: true,
11238
11136
  input: ["text", "image"],
11239
11137
  cost: {
11240
- input: 1.2,
11241
- output: 4,
11242
- cacheRead: 0.24,
11243
- cacheWrite: 0,
11138
+ input: 10,
11139
+ output: 50,
11140
+ cacheRead: 1,
11141
+ cacheWrite: 12.5,
11244
11142
  },
11245
- contextWindow: 202752,
11246
- maxTokens: 131072,
11143
+ contextWindow: 1000000,
11144
+ maxTokens: 128000,
11247
11145
  },
11248
11146
  "~anthropic/claude-haiku-latest": {
11249
11147
  id: "~anthropic/claude-haiku-latest",
@@ -11426,8 +11324,8 @@ export const MODELS = {
11426
11324
  reasoning: true,
11427
11325
  input: ["text"],
11428
11326
  cost: {
11429
- input: 0.08,
11430
- output: 0.29,
11327
+ input: 0.12,
11328
+ output: 0.5,
11431
11329
  cacheRead: 0,
11432
11330
  cacheWrite: 0,
11433
11331
  },
@@ -11791,6 +11689,23 @@ export const MODELS = {
11791
11689
  contextWindow: 200000,
11792
11690
  maxTokens: 8192,
11793
11691
  },
11692
+ "anthropic/claude-fable-5": {
11693
+ id: "anthropic/claude-fable-5",
11694
+ name: "Claude Fable 5",
11695
+ api: "anthropic-messages",
11696
+ provider: "vercel-ai-gateway",
11697
+ baseUrl: "https://ai-gateway.vercel.sh",
11698
+ reasoning: true,
11699
+ input: ["text", "image"],
11700
+ cost: {
11701
+ input: 10,
11702
+ output: 50,
11703
+ cacheRead: 1,
11704
+ cacheWrite: 12.5,
11705
+ },
11706
+ contextWindow: 1000000,
11707
+ maxTokens: 128000,
11708
+ },
11794
11709
  "anthropic/claude-haiku-4.5": {
11795
11710
  id: "anthropic/claude-haiku-4.5",
11796
11711
  name: "Claude Haiku 4.5",
@@ -14238,7 +14153,7 @@ export const MODELS = {
14238
14153
  cacheRead: 0.2,
14239
14154
  cacheWrite: 0,
14240
14155
  },
14241
- contextWindow: 2000000,
14156
+ contextWindow: 1000000,
14242
14157
  maxTokens: 30000,
14243
14158
  },
14244
14159
  "grok-4.20-0309-reasoning": {
@@ -14255,7 +14170,7 @@ export const MODELS = {
14255
14170
  cacheRead: 0.2,
14256
14171
  cacheWrite: 0,
14257
14172
  },
14258
- contextWindow: 2000000,
14173
+ contextWindow: 1000000,
14259
14174
  maxTokens: 30000,
14260
14175
  },
14261
14176
  "grok-4.3": {