@caupulican/pi-ai 0.80.19 → 0.80.21

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,
@@ -347,6 +347,23 @@ export const MODELS = {
347
347
  contextWindow: 163840,
348
348
  maxTokens: 81920,
349
349
  },
350
+ "eu.anthropic.claude-fable-5": {
351
+ id: "eu.anthropic.claude-fable-5",
352
+ name: "Claude Fable 5 (EU)",
353
+ api: "bedrock-converse-stream",
354
+ provider: "amazon-bedrock",
355
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
356
+ reasoning: true,
357
+ input: ["text", "image"],
358
+ cost: {
359
+ input: 11,
360
+ output: 55,
361
+ cacheRead: 1.1,
362
+ cacheWrite: 13.75,
363
+ },
364
+ contextWindow: 1000000,
365
+ maxTokens: 128000,
366
+ },
350
367
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
351
368
  id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
352
369
  name: "Claude Haiku 4.5 (EU)",
@@ -469,6 +486,23 @@ export const MODELS = {
469
486
  contextWindow: 1000000,
470
487
  maxTokens: 64000,
471
488
  },
489
+ "global.anthropic.claude-fable-5": {
490
+ id: "global.anthropic.claude-fable-5",
491
+ name: "Claude Fable 5 (Global)",
492
+ api: "bedrock-converse-stream",
493
+ provider: "amazon-bedrock",
494
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
495
+ reasoning: true,
496
+ input: ["text", "image"],
497
+ cost: {
498
+ input: 10,
499
+ output: 50,
500
+ cacheRead: 1,
501
+ cacheWrite: 12.5,
502
+ },
503
+ contextWindow: 1000000,
504
+ maxTokens: 128000,
505
+ },
472
506
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
473
507
  id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
474
508
  name: "Claude Haiku 4.5 (Global)",
@@ -1128,7 +1162,7 @@ export const MODELS = {
1128
1162
  api: "bedrock-converse-stream",
1129
1163
  provider: "amazon-bedrock",
1130
1164
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1131
- reasoning: false,
1165
+ reasoning: true,
1132
1166
  input: ["text"],
1133
1167
  cost: {
1134
1168
  input: 0.15,
@@ -1145,7 +1179,7 @@ export const MODELS = {
1145
1179
  api: "bedrock-converse-stream",
1146
1180
  provider: "amazon-bedrock",
1147
1181
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1148
- reasoning: false,
1182
+ reasoning: true,
1149
1183
  input: ["text"],
1150
1184
  cost: {
1151
1185
  input: 0.15,
@@ -1162,7 +1196,7 @@ export const MODELS = {
1162
1196
  api: "bedrock-converse-stream",
1163
1197
  provider: "amazon-bedrock",
1164
1198
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1165
- reasoning: false,
1199
+ reasoning: true,
1166
1200
  input: ["text"],
1167
1201
  cost: {
1168
1202
  input: 0.07,
@@ -1179,7 +1213,7 @@ export const MODELS = {
1179
1213
  api: "bedrock-converse-stream",
1180
1214
  provider: "amazon-bedrock",
1181
1215
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1182
- reasoning: false,
1216
+ reasoning: true,
1183
1217
  input: ["text"],
1184
1218
  cost: {
1185
1219
  input: 0.07,
@@ -1343,6 +1377,23 @@ export const MODELS = {
1343
1377
  contextWindow: 262000,
1344
1378
  maxTokens: 262000,
1345
1379
  },
1380
+ "us.anthropic.claude-fable-5": {
1381
+ id: "us.anthropic.claude-fable-5",
1382
+ name: "Claude Fable 5 (US)",
1383
+ api: "bedrock-converse-stream",
1384
+ provider: "amazon-bedrock",
1385
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1386
+ reasoning: true,
1387
+ input: ["text", "image"],
1388
+ cost: {
1389
+ input: 10,
1390
+ output: 50,
1391
+ cacheRead: 1,
1392
+ cacheWrite: 12.5,
1393
+ },
1394
+ contextWindow: 1000000,
1395
+ maxTokens: 128000,
1396
+ },
1346
1397
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
1347
1398
  id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
1348
1399
  name: "Claude Haiku 4.5 (US)",
@@ -1756,6 +1807,23 @@ export const MODELS = {
1756
1807
  contextWindow: 200000,
1757
1808
  maxTokens: 4096,
1758
1809
  },
1810
+ "claude-fable-5": {
1811
+ id: "claude-fable-5",
1812
+ name: "Claude Fable 5",
1813
+ api: "anthropic-messages",
1814
+ provider: "anthropic",
1815
+ baseUrl: "https://api.anthropic.com",
1816
+ reasoning: true,
1817
+ input: ["text", "image"],
1818
+ cost: {
1819
+ input: 10,
1820
+ output: 50,
1821
+ cacheRead: 1,
1822
+ cacheWrite: 12.5,
1823
+ },
1824
+ contextWindow: 1000000,
1825
+ maxTokens: 128000,
1826
+ },
1759
1827
  "claude-haiku-4-5": {
1760
1828
  id: "claude-haiku-4-5",
1761
1829
  name: "Claude Haiku 4.5 (latest)",
@@ -2932,6 +3000,23 @@ export const MODELS = {
2932
3000
  contextWindow: 200000,
2933
3001
  maxTokens: 8192,
2934
3002
  },
3003
+ "claude-fable-5": {
3004
+ id: "claude-fable-5",
3005
+ name: "Claude Fable 5",
3006
+ api: "anthropic-messages",
3007
+ provider: "cloudflare-ai-gateway",
3008
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3009
+ reasoning: true,
3010
+ input: ["text", "image"],
3011
+ cost: {
3012
+ input: 10,
3013
+ output: 50,
3014
+ cacheRead: 1,
3015
+ cacheWrite: 12.5,
3016
+ },
3017
+ contextWindow: 1000000,
3018
+ maxTokens: 128000,
3019
+ },
2935
3020
  "claude-haiku-4-5": {
2936
3021
  id: "claude-haiku-4-5",
2937
3022
  name: "Claude Haiku 4.5 (latest)",
@@ -4870,77 +4955,9 @@ export const MODELS = {
4870
4955
  },
4871
4956
  },
4872
4957
  "groq": {
4873
- "deepseek-r1-distill-llama-70b": {
4874
- id: "deepseek-r1-distill-llama-70b",
4875
- name: "DeepSeek R1 Distill Llama 70B",
4876
- api: "openai-completions",
4877
- provider: "groq",
4878
- baseUrl: "https://api.groq.com/openai/v1",
4879
- reasoning: true,
4880
- input: ["text"],
4881
- cost: {
4882
- input: 0.75,
4883
- output: 0.99,
4884
- cacheRead: 0,
4885
- cacheWrite: 0,
4886
- },
4887
- contextWindow: 131072,
4888
- maxTokens: 8192,
4889
- },
4890
- "gemma2-9b-it": {
4891
- id: "gemma2-9b-it",
4892
- name: "Gemma 2 9B",
4893
- api: "openai-completions",
4894
- provider: "groq",
4895
- baseUrl: "https://api.groq.com/openai/v1",
4896
- reasoning: false,
4897
- input: ["text"],
4898
- cost: {
4899
- input: 0.2,
4900
- output: 0.2,
4901
- cacheRead: 0,
4902
- cacheWrite: 0,
4903
- },
4904
- contextWindow: 8192,
4905
- maxTokens: 8192,
4906
- },
4907
- "groq/compound": {
4908
- id: "groq/compound",
4909
- name: "Compound",
4910
- api: "openai-completions",
4911
- provider: "groq",
4912
- baseUrl: "https://api.groq.com/openai/v1",
4913
- reasoning: true,
4914
- input: ["text"],
4915
- cost: {
4916
- input: 0,
4917
- output: 0,
4918
- cacheRead: 0,
4919
- cacheWrite: 0,
4920
- },
4921
- contextWindow: 131072,
4922
- maxTokens: 8192,
4923
- },
4924
- "groq/compound-mini": {
4925
- id: "groq/compound-mini",
4926
- name: "Compound Mini",
4927
- api: "openai-completions",
4928
- provider: "groq",
4929
- baseUrl: "https://api.groq.com/openai/v1",
4930
- reasoning: true,
4931
- input: ["text"],
4932
- cost: {
4933
- input: 0,
4934
- output: 0,
4935
- cacheRead: 0,
4936
- cacheWrite: 0,
4937
- },
4938
- contextWindow: 131072,
4939
- maxTokens: 8192,
4940
- },
4941
4958
  "llama-3.1-8b-instant": {
4942
4959
  id: "llama-3.1-8b-instant",
4943
- name: "Llama 3.1 8B Instant",
4960
+ name: "Llama 3.1 8B",
4944
4961
  api: "openai-completions",
4945
4962
  provider: "groq",
4946
4963
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4957,7 +4974,7 @@ export const MODELS = {
4957
4974
  },
4958
4975
  "llama-3.3-70b-versatile": {
4959
4976
  id: "llama-3.3-70b-versatile",
4960
- name: "Llama 3.3 70B Versatile",
4977
+ name: "Llama 3.3 70B",
4961
4978
  api: "openai-completions",
4962
4979
  provider: "groq",
4963
4980
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4972,60 +4989,9 @@ export const MODELS = {
4972
4989
  contextWindow: 131072,
4973
4990
  maxTokens: 32768,
4974
4991
  },
4975
- "llama3-70b-8192": {
4976
- id: "llama3-70b-8192",
4977
- name: "Llama 3 70B",
4978
- api: "openai-completions",
4979
- provider: "groq",
4980
- baseUrl: "https://api.groq.com/openai/v1",
4981
- reasoning: false,
4982
- input: ["text"],
4983
- cost: {
4984
- input: 0.59,
4985
- output: 0.79,
4986
- cacheRead: 0,
4987
- cacheWrite: 0,
4988
- },
4989
- contextWindow: 8192,
4990
- maxTokens: 8192,
4991
- },
4992
- "llama3-8b-8192": {
4993
- id: "llama3-8b-8192",
4994
- name: "Llama 3 8B",
4995
- api: "openai-completions",
4996
- provider: "groq",
4997
- baseUrl: "https://api.groq.com/openai/v1",
4998
- reasoning: false,
4999
- input: ["text"],
5000
- cost: {
5001
- input: 0.05,
5002
- output: 0.08,
5003
- cacheRead: 0,
5004
- cacheWrite: 0,
5005
- },
5006
- contextWindow: 8192,
5007
- maxTokens: 8192,
5008
- },
5009
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
5010
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
5011
- name: "Llama 4 Maverick 17B",
5012
- api: "openai-completions",
5013
- provider: "groq",
5014
- baseUrl: "https://api.groq.com/openai/v1",
5015
- reasoning: false,
5016
- input: ["text", "image"],
5017
- cost: {
5018
- input: 0.2,
5019
- output: 0.6,
5020
- cacheRead: 0,
5021
- cacheWrite: 0,
5022
- },
5023
- contextWindow: 131072,
5024
- maxTokens: 8192,
5025
- },
5026
4992
  "meta-llama/llama-4-scout-17b-16e-instruct": {
5027
4993
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
5028
- name: "Llama 4 Scout 17B",
4994
+ name: "Llama 4 Scout 17B 16E",
5029
4995
  api: "openai-completions",
5030
4996
  provider: "groq",
5031
4997
  baseUrl: "https://api.groq.com/openai/v1",
@@ -5040,57 +5006,6 @@ export const MODELS = {
5040
5006
  contextWindow: 131072,
5041
5007
  maxTokens: 8192,
5042
5008
  },
5043
- "mistral-saba-24b": {
5044
- id: "mistral-saba-24b",
5045
- name: "Mistral Saba 24B",
5046
- api: "openai-completions",
5047
- provider: "groq",
5048
- baseUrl: "https://api.groq.com/openai/v1",
5049
- reasoning: false,
5050
- input: ["text"],
5051
- cost: {
5052
- input: 0.79,
5053
- output: 0.79,
5054
- cacheRead: 0,
5055
- cacheWrite: 0,
5056
- },
5057
- contextWindow: 32768,
5058
- maxTokens: 32768,
5059
- },
5060
- "moonshotai/kimi-k2-instruct": {
5061
- id: "moonshotai/kimi-k2-instruct",
5062
- name: "Kimi K2 Instruct",
5063
- api: "openai-completions",
5064
- provider: "groq",
5065
- baseUrl: "https://api.groq.com/openai/v1",
5066
- reasoning: false,
5067
- input: ["text"],
5068
- cost: {
5069
- input: 1,
5070
- output: 3,
5071
- cacheRead: 0,
5072
- cacheWrite: 0,
5073
- },
5074
- contextWindow: 131072,
5075
- maxTokens: 16384,
5076
- },
5077
- "moonshotai/kimi-k2-instruct-0905": {
5078
- id: "moonshotai/kimi-k2-instruct-0905",
5079
- name: "Kimi K2 Instruct 0905",
5080
- api: "openai-completions",
5081
- provider: "groq",
5082
- baseUrl: "https://api.groq.com/openai/v1",
5083
- reasoning: false,
5084
- input: ["text"],
5085
- cost: {
5086
- input: 1,
5087
- output: 3,
5088
- cacheRead: 0.5,
5089
- cacheWrite: 0,
5090
- },
5091
- contextWindow: 262144,
5092
- maxTokens: 16384,
5093
- },
5094
5009
  "openai/gpt-oss-120b": {
5095
5010
  id: "openai/gpt-oss-120b",
5096
5011
  name: "GPT OSS 120B",
@@ -5142,26 +5057,9 @@ export const MODELS = {
5142
5057
  contextWindow: 131072,
5143
5058
  maxTokens: 65536,
5144
5059
  },
5145
- "qwen-qwq-32b": {
5146
- id: "qwen-qwq-32b",
5147
- name: "Qwen QwQ 32B",
5148
- api: "openai-completions",
5149
- provider: "groq",
5150
- baseUrl: "https://api.groq.com/openai/v1",
5151
- reasoning: true,
5152
- input: ["text"],
5153
- cost: {
5154
- input: 0.29,
5155
- output: 0.39,
5156
- cacheRead: 0,
5157
- cacheWrite: 0,
5158
- },
5159
- contextWindow: 131072,
5160
- maxTokens: 16384,
5161
- },
5162
5060
  "qwen/qwen3-32b": {
5163
5061
  id: "qwen/qwen3-32b",
5164
- name: "Qwen3 32B",
5062
+ name: "Qwen3-32B",
5165
5063
  api: "openai-completions",
5166
5064
  provider: "groq",
5167
5065
  baseUrl: "https://api.groq.com/openai/v1",
@@ -7342,6 +7240,23 @@ export const MODELS = {
7342
7240
  contextWindow: 200000,
7343
7241
  maxTokens: 32000,
7344
7242
  },
7243
+ "claude-fable-5": {
7244
+ id: "claude-fable-5",
7245
+ name: "Claude Fable 5",
7246
+ api: "anthropic-messages",
7247
+ provider: "opencode",
7248
+ baseUrl: "https://opencode.ai/zen",
7249
+ reasoning: true,
7250
+ input: ["text", "image"],
7251
+ cost: {
7252
+ input: 10,
7253
+ output: 50,
7254
+ cacheRead: 1,
7255
+ cacheWrite: 12.5,
7256
+ },
7257
+ contextWindow: 1000000,
7258
+ maxTokens: 128000,
7259
+ },
7345
7260
  "claude-haiku-4-5": {
7346
7261
  id: "claude-haiku-4-5",
7347
7262
  name: "Claude Haiku 4.5",
@@ -7515,7 +7430,7 @@ export const MODELS = {
7515
7430
  cost: {
7516
7431
  input: 0.14,
7517
7432
  output: 0.28,
7518
- cacheRead: 0.03,
7433
+ cacheRead: 0.028,
7519
7434
  cacheWrite: 0,
7520
7435
  },
7521
7436
  contextWindow: 1000000,
@@ -7540,6 +7455,25 @@ export const MODELS = {
7540
7455
  contextWindow: 200000,
7541
7456
  maxTokens: 128000,
7542
7457
  },
7458
+ "deepseek-v4-pro": {
7459
+ id: "deepseek-v4-pro",
7460
+ name: "DeepSeek V4 Pro",
7461
+ api: "openai-completions",
7462
+ provider: "opencode",
7463
+ baseUrl: "https://opencode.ai/zen/v1",
7464
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7465
+ reasoning: true,
7466
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7467
+ input: ["text"],
7468
+ cost: {
7469
+ input: 1.74,
7470
+ output: 3.84,
7471
+ cacheRead: 0.145,
7472
+ cacheWrite: 0,
7473
+ },
7474
+ contextWindow: 1000000,
7475
+ maxTokens: 384000,
7476
+ },
7543
7477
  "gemini-3-flash": {
7544
7478
  id: "gemini-3-flash",
7545
7479
  name: "Gemini 3 Flash",
@@ -8038,6 +7972,23 @@ export const MODELS = {
8038
7972
  contextWindow: 1000000,
8039
7973
  maxTokens: 128000,
8040
7974
  },
7975
+ "north-mini-code-free": {
7976
+ id: "north-mini-code-free",
7977
+ name: "North Mini Code Free",
7978
+ api: "openai-completions",
7979
+ provider: "opencode",
7980
+ baseUrl: "https://opencode.ai/zen/v1",
7981
+ reasoning: true,
7982
+ input: ["text"],
7983
+ cost: {
7984
+ input: 0,
7985
+ output: 0,
7986
+ cacheRead: 0,
7987
+ cacheWrite: 0,
7988
+ },
7989
+ contextWindow: 256000,
7990
+ maxTokens: 64000,
7991
+ },
8041
7992
  "qwen3.5-plus": {
8042
7993
  id: "qwen3.5-plus",
8043
7994
  name: "Qwen3.5 Plus",
@@ -8457,6 +8408,23 @@ export const MODELS = {
8457
8408
  contextWindow: 200000,
8458
8409
  maxTokens: 8192,
8459
8410
  },
8411
+ "anthropic/claude-fable-5": {
8412
+ id: "anthropic/claude-fable-5",
8413
+ name: "Anthropic: Claude Fable 5",
8414
+ api: "openai-completions",
8415
+ provider: "openrouter",
8416
+ baseUrl: "https://openrouter.ai/api/v1",
8417
+ reasoning: true,
8418
+ input: ["text", "image"],
8419
+ cost: {
8420
+ input: 10,
8421
+ output: 50,
8422
+ cacheRead: 1,
8423
+ cacheWrite: 12.5,
8424
+ },
8425
+ contextWindow: 1000000,
8426
+ maxTokens: 128000,
8427
+ },
8460
8428
  "anthropic/claude-haiku-4.5": {
8461
8429
  id: "anthropic/claude-haiku-4.5",
8462
8430
  name: "Anthropic: Claude Haiku 4.5",
@@ -8885,7 +8853,7 @@ export const MODELS = {
8885
8853
  cacheRead: 0.135,
8886
8854
  cacheWrite: 0,
8887
8855
  },
8888
- contextWindow: 163840,
8856
+ contextWindow: 131072,
8889
8857
  maxTokens: 16384,
8890
8858
  },
8891
8859
  "deepseek/deepseek-chat-v3.1": {
@@ -9617,8 +9585,8 @@ export const MODELS = {
9617
9585
  input: ["text"],
9618
9586
  cost: {
9619
9587
  input: 0.15,
9620
- output: 1.15,
9621
- cacheRead: 0,
9588
+ output: 0.8999999999999999,
9589
+ cacheRead: 0.049999999999999996,
9622
9590
  cacheWrite: 0,
9623
9591
  },
9624
9592
  contextWindow: 204800,
@@ -9633,13 +9601,13 @@ export const MODELS = {
9633
9601
  reasoning: true,
9634
9602
  input: ["text"],
9635
9603
  cost: {
9636
- input: 0.27899999999999997,
9637
- output: 1.2,
9638
- cacheRead: 0,
9604
+ input: 0.27,
9605
+ output: 1.08,
9606
+ cacheRead: 0.054,
9639
9607
  cacheWrite: 0,
9640
9608
  },
9641
9609
  contextWindow: 204800,
9642
- maxTokens: 196608,
9610
+ maxTokens: 131072,
9643
9611
  },
9644
9612
  "minimax/minimax-m3": {
9645
9613
  id: "minimax/minimax-m3",
@@ -10033,24 +10001,6 @@ export const MODELS = {
10033
10001
  contextWindow: 262144,
10034
10002
  maxTokens: 262142,
10035
10003
  },
10036
- "moonshotai/kimi-k2.6:free": {
10037
- id: "moonshotai/kimi-k2.6:free",
10038
- name: "MoonshotAI: Kimi K2.6 (free)",
10039
- api: "openai-completions",
10040
- provider: "openrouter",
10041
- baseUrl: "https://openrouter.ai/api/v1",
10042
- compat: { "supportsDeveloperRole": false },
10043
- reasoning: true,
10044
- input: ["text", "image"],
10045
- cost: {
10046
- input: 0,
10047
- output: 0,
10048
- cacheRead: 0,
10049
- cacheWrite: 0,
10050
- },
10051
- contextWindow: 262144,
10052
- maxTokens: 4096,
10053
- },
10054
10004
  "nex-agi/nex-n2-pro:free": {
10055
10005
  id: "nex-agi/nex-n2-pro:free",
10056
10006
  name: "Nex AGI: Nex-N2-Pro (free)",
@@ -12324,23 +12274,6 @@ export const MODELS = {
12324
12274
  contextWindow: 1048576,
12325
12275
  maxTokens: 131072,
12326
12276
  },
12327
- "z-ai/glm-4-32b": {
12328
- id: "z-ai/glm-4-32b",
12329
- name: "Z.ai: GLM 4 32B ",
12330
- api: "openai-completions",
12331
- provider: "openrouter",
12332
- baseUrl: "https://openrouter.ai/api/v1",
12333
- reasoning: false,
12334
- input: ["text"],
12335
- cost: {
12336
- input: 0.09999999999999999,
12337
- output: 0.09999999999999999,
12338
- cacheRead: 0,
12339
- cacheWrite: 0,
12340
- },
12341
- contextWindow: 128000,
12342
- maxTokens: 4096,
12343
- },
12344
12277
  "z-ai/glm-4.5": {
12345
12278
  id: "z-ai/glm-4.5",
12346
12279
  name: "Z.ai: GLM 4.5",
@@ -12375,23 +12308,6 @@ export const MODELS = {
12375
12308
  contextWindow: 131072,
12376
12309
  maxTokens: 131070,
12377
12310
  },
12378
- "z-ai/glm-4.5-air:free": {
12379
- id: "z-ai/glm-4.5-air:free",
12380
- name: "Z.ai: GLM 4.5 Air (free)",
12381
- api: "openai-completions",
12382
- provider: "openrouter",
12383
- baseUrl: "https://openrouter.ai/api/v1",
12384
- reasoning: true,
12385
- input: ["text"],
12386
- cost: {
12387
- input: 0,
12388
- output: 0,
12389
- cacheRead: 0,
12390
- cacheWrite: 0,
12391
- },
12392
- contextWindow: 131072,
12393
- maxTokens: 96000,
12394
- },
12395
12311
  "z-ai/glm-4.5v": {
12396
12312
  id: "z-ai/glm-4.5v",
12397
12313
  name: "Z.ai: GLM 4.5V",
@@ -12437,11 +12353,11 @@ export const MODELS = {
12437
12353
  cost: {
12438
12354
  input: 0.3,
12439
12355
  output: 0.8999999999999999,
12440
- cacheRead: 0.049999999999999996,
12356
+ cacheRead: 0.055,
12441
12357
  cacheWrite: 0,
12442
12358
  },
12443
12359
  contextWindow: 131072,
12444
- maxTokens: 24000,
12360
+ maxTokens: 32768,
12445
12361
  },
12446
12362
  "z-ai/glm-4.7": {
12447
12363
  id: "z-ai/glm-4.7",
@@ -12508,7 +12424,7 @@ export const MODELS = {
12508
12424
  cacheRead: 0.24,
12509
12425
  cacheWrite: 0,
12510
12426
  },
12511
- contextWindow: 202752,
12427
+ contextWindow: 262144,
12512
12428
  maxTokens: 131072,
12513
12429
  },
12514
12430
  "z-ai/glm-5.1": {
@@ -12528,22 +12444,22 @@ export const MODELS = {
12528
12444
  contextWindow: 202752,
12529
12445
  maxTokens: 4096,
12530
12446
  },
12531
- "z-ai/glm-5v-turbo": {
12532
- id: "z-ai/glm-5v-turbo",
12533
- name: "Z.ai: GLM 5V Turbo",
12447
+ "~anthropic/claude-fable-latest": {
12448
+ id: "~anthropic/claude-fable-latest",
12449
+ name: "Anthropic: Claude Fable Latest",
12534
12450
  api: "openai-completions",
12535
12451
  provider: "openrouter",
12536
12452
  baseUrl: "https://openrouter.ai/api/v1",
12537
12453
  reasoning: true,
12538
12454
  input: ["text", "image"],
12539
12455
  cost: {
12540
- input: 1.2,
12541
- output: 4,
12542
- cacheRead: 0.24,
12543
- cacheWrite: 0,
12456
+ input: 10,
12457
+ output: 50,
12458
+ cacheRead: 1,
12459
+ cacheWrite: 12.5,
12544
12460
  },
12545
- contextWindow: 202752,
12546
- maxTokens: 131072,
12461
+ contextWindow: 1000000,
12462
+ maxTokens: 128000,
12547
12463
  },
12548
12464
  "~anthropic/claude-haiku-latest": {
12549
12465
  id: "~anthropic/claude-haiku-latest",
@@ -13451,6 +13367,23 @@ export const MODELS = {
13451
13367
  contextWindow: 200000,
13452
13368
  maxTokens: 8192,
13453
13369
  },
13370
+ "anthropic/claude-fable-5": {
13371
+ id: "anthropic/claude-fable-5",
13372
+ name: "Claude Fable 5",
13373
+ api: "anthropic-messages",
13374
+ provider: "vercel-ai-gateway",
13375
+ baseUrl: "https://ai-gateway.vercel.sh",
13376
+ reasoning: true,
13377
+ input: ["text", "image"],
13378
+ cost: {
13379
+ input: 10,
13380
+ output: 50,
13381
+ cacheRead: 1,
13382
+ cacheWrite: 12.5,
13383
+ },
13384
+ contextWindow: 1000000,
13385
+ maxTokens: 128000,
13386
+ },
13454
13387
  "anthropic/claude-haiku-4.5": {
13455
13388
  id: "anthropic/claude-haiku-4.5",
13456
13389
  name: "Claude Haiku 4.5",
@@ -15951,7 +15884,7 @@ export const MODELS = {
15951
15884
  cacheRead: 0.2,
15952
15885
  cacheWrite: 0,
15953
15886
  },
15954
- contextWindow: 2000000,
15887
+ contextWindow: 1000000,
15955
15888
  maxTokens: 30000,
15956
15889
  },
15957
15890
  "grok-4.20-0309-reasoning": {
@@ -15968,7 +15901,7 @@ export const MODELS = {
15968
15901
  cacheRead: 0.2,
15969
15902
  cacheWrite: 0,
15970
15903
  },
15971
- contextWindow: 2000000,
15904
+ contextWindow: 1000000,
15972
15905
  maxTokens: 30000,
15973
15906
  },
15974
15907
  "grok-4.3": {
@@ -16114,6 +16047,24 @@ export const MODELS = {
16114
16047
  contextWindow: 1048576,
16115
16048
  maxTokens: 131072,
16116
16049
  },
16050
+ "mimo-v2.5-pro-ultraspeed": {
16051
+ id: "mimo-v2.5-pro-ultraspeed",
16052
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16053
+ api: "openai-completions",
16054
+ provider: "xiaomi",
16055
+ baseUrl: "https://api.xiaomimimo.com/v1",
16056
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16057
+ reasoning: true,
16058
+ input: ["text"],
16059
+ cost: {
16060
+ input: 1.305,
16061
+ output: 2.61,
16062
+ cacheRead: 0.0108,
16063
+ cacheWrite: 0,
16064
+ },
16065
+ contextWindow: 1048576,
16066
+ maxTokens: 131072,
16067
+ },
16117
16068
  },
16118
16069
  "xiaomi-token-plan-ams": {
16119
16070
  "mimo-v2-omni": {
@@ -16188,6 +16139,24 @@ export const MODELS = {
16188
16139
  contextWindow: 1048576,
16189
16140
  maxTokens: 131072,
16190
16141
  },
16142
+ "mimo-v2.5-pro-ultraspeed": {
16143
+ id: "mimo-v2.5-pro-ultraspeed",
16144
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16145
+ api: "openai-completions",
16146
+ provider: "xiaomi-token-plan-ams",
16147
+ baseUrl: "https://token-plan-ams.xiaomimimo.com/v1",
16148
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16149
+ reasoning: true,
16150
+ input: ["text"],
16151
+ cost: {
16152
+ input: 1.305,
16153
+ output: 2.61,
16154
+ cacheRead: 0.0108,
16155
+ cacheWrite: 0,
16156
+ },
16157
+ contextWindow: 1048576,
16158
+ maxTokens: 131072,
16159
+ },
16191
16160
  },
16192
16161
  "xiaomi-token-plan-cn": {
16193
16162
  "mimo-v2-omni": {
@@ -16262,6 +16231,24 @@ export const MODELS = {
16262
16231
  contextWindow: 1048576,
16263
16232
  maxTokens: 131072,
16264
16233
  },
16234
+ "mimo-v2.5-pro-ultraspeed": {
16235
+ id: "mimo-v2.5-pro-ultraspeed",
16236
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16237
+ api: "openai-completions",
16238
+ provider: "xiaomi-token-plan-cn",
16239
+ baseUrl: "https://token-plan-cn.xiaomimimo.com/v1",
16240
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16241
+ reasoning: true,
16242
+ input: ["text"],
16243
+ cost: {
16244
+ input: 1.305,
16245
+ output: 2.61,
16246
+ cacheRead: 0.0108,
16247
+ cacheWrite: 0,
16248
+ },
16249
+ contextWindow: 1048576,
16250
+ maxTokens: 131072,
16251
+ },
16265
16252
  },
16266
16253
  "xiaomi-token-plan-sgp": {
16267
16254
  "mimo-v2-omni": {
@@ -16336,6 +16323,24 @@ export const MODELS = {
16336
16323
  contextWindow: 1048576,
16337
16324
  maxTokens: 131072,
16338
16325
  },
16326
+ "mimo-v2.5-pro-ultraspeed": {
16327
+ id: "mimo-v2.5-pro-ultraspeed",
16328
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16329
+ api: "openai-completions",
16330
+ provider: "xiaomi-token-plan-sgp",
16331
+ baseUrl: "https://token-plan-sgp.xiaomimimo.com/v1",
16332
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16333
+ reasoning: true,
16334
+ input: ["text"],
16335
+ cost: {
16336
+ input: 1.305,
16337
+ output: 2.61,
16338
+ cacheRead: 0.0108,
16339
+ cacheWrite: 0,
16340
+ },
16341
+ contextWindow: 1048576,
16342
+ maxTokens: 131072,
16343
+ },
16339
16344
  },
16340
16345
  "zai": {
16341
16346
  "glm-4.5-air": {