@fleetagent/pi-ai 0.0.9 → 0.0.10

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,
@@ -345,6 +345,23 @@ export const MODELS = {
345
345
  contextWindow: 163840,
346
346
  maxTokens: 81920,
347
347
  },
348
+ "eu.anthropic.claude-fable-5": {
349
+ id: "eu.anthropic.claude-fable-5",
350
+ name: "Claude Fable 5 (EU)",
351
+ api: "bedrock-converse-stream",
352
+ provider: "amazon-bedrock",
353
+ baseUrl: "https://bedrock-runtime.eu-central-1.amazonaws.com",
354
+ reasoning: true,
355
+ input: ["text", "image"],
356
+ cost: {
357
+ input: 11,
358
+ output: 55,
359
+ cacheRead: 1.1,
360
+ cacheWrite: 13.75,
361
+ },
362
+ contextWindow: 1000000,
363
+ maxTokens: 128000,
364
+ },
348
365
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {
349
366
  id: "eu.anthropic.claude-haiku-4-5-20251001-v1:0",
350
367
  name: "Claude Haiku 4.5 (EU)",
@@ -466,6 +483,23 @@ export const MODELS = {
466
483
  contextWindow: 1000000,
467
484
  maxTokens: 64000,
468
485
  },
486
+ "global.anthropic.claude-fable-5": {
487
+ id: "global.anthropic.claude-fable-5",
488
+ name: "Claude Fable 5 (Global)",
489
+ api: "bedrock-converse-stream",
490
+ provider: "amazon-bedrock",
491
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
492
+ reasoning: true,
493
+ input: ["text", "image"],
494
+ cost: {
495
+ input: 10,
496
+ output: 50,
497
+ cacheRead: 1,
498
+ cacheWrite: 12.5,
499
+ },
500
+ contextWindow: 1000000,
501
+ maxTokens: 128000,
502
+ },
469
503
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {
470
504
  id: "global.anthropic.claude-haiku-4-5-20251001-v1:0",
471
505
  name: "Claude Haiku 4.5 (Global)",
@@ -1123,7 +1157,7 @@ export const MODELS = {
1123
1157
  api: "bedrock-converse-stream",
1124
1158
  provider: "amazon-bedrock",
1125
1159
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1126
- reasoning: false,
1160
+ reasoning: true,
1127
1161
  input: ["text"],
1128
1162
  cost: {
1129
1163
  input: 0.15,
@@ -1140,7 +1174,7 @@ export const MODELS = {
1140
1174
  api: "bedrock-converse-stream",
1141
1175
  provider: "amazon-bedrock",
1142
1176
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1143
- reasoning: false,
1177
+ reasoning: true,
1144
1178
  input: ["text"],
1145
1179
  cost: {
1146
1180
  input: 0.15,
@@ -1157,7 +1191,7 @@ export const MODELS = {
1157
1191
  api: "bedrock-converse-stream",
1158
1192
  provider: "amazon-bedrock",
1159
1193
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1160
- reasoning: false,
1194
+ reasoning: true,
1161
1195
  input: ["text"],
1162
1196
  cost: {
1163
1197
  input: 0.07,
@@ -1174,7 +1208,7 @@ export const MODELS = {
1174
1208
  api: "bedrock-converse-stream",
1175
1209
  provider: "amazon-bedrock",
1176
1210
  baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1177
- reasoning: false,
1211
+ reasoning: true,
1178
1212
  input: ["text"],
1179
1213
  cost: {
1180
1214
  input: 0.07,
@@ -1338,6 +1372,23 @@ export const MODELS = {
1338
1372
  contextWindow: 262000,
1339
1373
  maxTokens: 262000,
1340
1374
  },
1375
+ "us.anthropic.claude-fable-5": {
1376
+ id: "us.anthropic.claude-fable-5",
1377
+ name: "Claude Fable 5 (US)",
1378
+ api: "bedrock-converse-stream",
1379
+ provider: "amazon-bedrock",
1380
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1381
+ reasoning: true,
1382
+ input: ["text", "image"],
1383
+ cost: {
1384
+ input: 10,
1385
+ output: 50,
1386
+ cacheRead: 1,
1387
+ cacheWrite: 12.5,
1388
+ },
1389
+ contextWindow: 1000000,
1390
+ maxTokens: 128000,
1391
+ },
1341
1392
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {
1342
1393
  id: "us.anthropic.claude-haiku-4-5-20251001-v1:0",
1343
1394
  name: "Claude Haiku 4.5 (US)",
@@ -1750,6 +1801,23 @@ export const MODELS = {
1750
1801
  contextWindow: 200000,
1751
1802
  maxTokens: 4096,
1752
1803
  },
1804
+ "claude-fable-5": {
1805
+ id: "claude-fable-5",
1806
+ name: "Claude Fable 5",
1807
+ api: "anthropic-messages",
1808
+ provider: "anthropic",
1809
+ baseUrl: "https://api.anthropic.com",
1810
+ reasoning: true,
1811
+ input: ["text", "image"],
1812
+ cost: {
1813
+ input: 10,
1814
+ output: 50,
1815
+ cacheRead: 1,
1816
+ cacheWrite: 12.5,
1817
+ },
1818
+ contextWindow: 1000000,
1819
+ maxTokens: 128000,
1820
+ },
1753
1821
  "claude-haiku-4-5": {
1754
1822
  id: "claude-haiku-4-5",
1755
1823
  name: "Claude Haiku 4.5 (latest)",
@@ -2925,6 +2993,23 @@ export const MODELS = {
2925
2993
  contextWindow: 200000,
2926
2994
  maxTokens: 8192,
2927
2995
  },
2996
+ "claude-fable-5": {
2997
+ id: "claude-fable-5",
2998
+ name: "Claude Fable 5",
2999
+ api: "anthropic-messages",
3000
+ provider: "cloudflare-ai-gateway",
3001
+ baseUrl: "https://gateway.ai.cloudflare.com/v1/{CLOUDFLARE_ACCOUNT_ID}/{CLOUDFLARE_GATEWAY_ID}/anthropic",
3002
+ reasoning: true,
3003
+ input: ["text", "image"],
3004
+ cost: {
3005
+ input: 10,
3006
+ output: 50,
3007
+ cacheRead: 1,
3008
+ cacheWrite: 12.5,
3009
+ },
3010
+ contextWindow: 1000000,
3011
+ maxTokens: 128000,
3012
+ },
2928
3013
  "claude-haiku-4-5": {
2929
3014
  id: "claude-haiku-4-5",
2930
3015
  name: "Claude Haiku 4.5 (latest)",
@@ -4861,77 +4946,9 @@ export const MODELS = {
4861
4946
  },
4862
4947
  },
4863
4948
  "groq": {
4864
- "deepseek-r1-distill-llama-70b": {
4865
- id: "deepseek-r1-distill-llama-70b",
4866
- name: "DeepSeek R1 Distill Llama 70B",
4867
- api: "openai-completions",
4868
- provider: "groq",
4869
- baseUrl: "https://api.groq.com/openai/v1",
4870
- reasoning: true,
4871
- input: ["text"],
4872
- cost: {
4873
- input: 0.75,
4874
- output: 0.99,
4875
- cacheRead: 0,
4876
- cacheWrite: 0,
4877
- },
4878
- contextWindow: 131072,
4879
- maxTokens: 8192,
4880
- },
4881
- "gemma2-9b-it": {
4882
- id: "gemma2-9b-it",
4883
- name: "Gemma 2 9B",
4884
- api: "openai-completions",
4885
- provider: "groq",
4886
- baseUrl: "https://api.groq.com/openai/v1",
4887
- reasoning: false,
4888
- input: ["text"],
4889
- cost: {
4890
- input: 0.2,
4891
- output: 0.2,
4892
- cacheRead: 0,
4893
- cacheWrite: 0,
4894
- },
4895
- contextWindow: 8192,
4896
- maxTokens: 8192,
4897
- },
4898
- "groq/compound": {
4899
- id: "groq/compound",
4900
- name: "Compound",
4901
- api: "openai-completions",
4902
- provider: "groq",
4903
- baseUrl: "https://api.groq.com/openai/v1",
4904
- reasoning: true,
4905
- input: ["text"],
4906
- cost: {
4907
- input: 0,
4908
- output: 0,
4909
- cacheRead: 0,
4910
- cacheWrite: 0,
4911
- },
4912
- contextWindow: 131072,
4913
- maxTokens: 8192,
4914
- },
4915
- "groq/compound-mini": {
4916
- id: "groq/compound-mini",
4917
- name: "Compound Mini",
4918
- api: "openai-completions",
4919
- provider: "groq",
4920
- baseUrl: "https://api.groq.com/openai/v1",
4921
- reasoning: true,
4922
- input: ["text"],
4923
- cost: {
4924
- input: 0,
4925
- output: 0,
4926
- cacheRead: 0,
4927
- cacheWrite: 0,
4928
- },
4929
- contextWindow: 131072,
4930
- maxTokens: 8192,
4931
- },
4932
4949
  "llama-3.1-8b-instant": {
4933
4950
  id: "llama-3.1-8b-instant",
4934
- name: "Llama 3.1 8B Instant",
4951
+ name: "Llama 3.1 8B",
4935
4952
  api: "openai-completions",
4936
4953
  provider: "groq",
4937
4954
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4948,7 +4965,7 @@ export const MODELS = {
4948
4965
  },
4949
4966
  "llama-3.3-70b-versatile": {
4950
4967
  id: "llama-3.3-70b-versatile",
4951
- name: "Llama 3.3 70B Versatile",
4968
+ name: "Llama 3.3 70B",
4952
4969
  api: "openai-completions",
4953
4970
  provider: "groq",
4954
4971
  baseUrl: "https://api.groq.com/openai/v1",
@@ -4963,60 +4980,9 @@ export const MODELS = {
4963
4980
  contextWindow: 131072,
4964
4981
  maxTokens: 32768,
4965
4982
  },
4966
- "llama3-70b-8192": {
4967
- id: "llama3-70b-8192",
4968
- name: "Llama 3 70B",
4969
- api: "openai-completions",
4970
- provider: "groq",
4971
- baseUrl: "https://api.groq.com/openai/v1",
4972
- reasoning: false,
4973
- input: ["text"],
4974
- cost: {
4975
- input: 0.59,
4976
- output: 0.79,
4977
- cacheRead: 0,
4978
- cacheWrite: 0,
4979
- },
4980
- contextWindow: 8192,
4981
- maxTokens: 8192,
4982
- },
4983
- "llama3-8b-8192": {
4984
- id: "llama3-8b-8192",
4985
- name: "Llama 3 8B",
4986
- api: "openai-completions",
4987
- provider: "groq",
4988
- baseUrl: "https://api.groq.com/openai/v1",
4989
- reasoning: false,
4990
- input: ["text"],
4991
- cost: {
4992
- input: 0.05,
4993
- output: 0.08,
4994
- cacheRead: 0,
4995
- cacheWrite: 0,
4996
- },
4997
- contextWindow: 8192,
4998
- maxTokens: 8192,
4999
- },
5000
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
5001
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
5002
- name: "Llama 4 Maverick 17B",
5003
- api: "openai-completions",
5004
- provider: "groq",
5005
- baseUrl: "https://api.groq.com/openai/v1",
5006
- reasoning: false,
5007
- input: ["text", "image"],
5008
- cost: {
5009
- input: 0.2,
5010
- output: 0.6,
5011
- cacheRead: 0,
5012
- cacheWrite: 0,
5013
- },
5014
- contextWindow: 131072,
5015
- maxTokens: 8192,
5016
- },
5017
4983
  "meta-llama/llama-4-scout-17b-16e-instruct": {
5018
4984
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
5019
- name: "Llama 4 Scout 17B",
4985
+ name: "Llama 4 Scout 17B 16E",
5020
4986
  api: "openai-completions",
5021
4987
  provider: "groq",
5022
4988
  baseUrl: "https://api.groq.com/openai/v1",
@@ -5031,57 +4997,6 @@ export const MODELS = {
5031
4997
  contextWindow: 131072,
5032
4998
  maxTokens: 8192,
5033
4999
  },
5034
- "mistral-saba-24b": {
5035
- id: "mistral-saba-24b",
5036
- name: "Mistral Saba 24B",
5037
- api: "openai-completions",
5038
- provider: "groq",
5039
- baseUrl: "https://api.groq.com/openai/v1",
5040
- reasoning: false,
5041
- input: ["text"],
5042
- cost: {
5043
- input: 0.79,
5044
- output: 0.79,
5045
- cacheRead: 0,
5046
- cacheWrite: 0,
5047
- },
5048
- contextWindow: 32768,
5049
- maxTokens: 32768,
5050
- },
5051
- "moonshotai/kimi-k2-instruct": {
5052
- id: "moonshotai/kimi-k2-instruct",
5053
- name: "Kimi K2 Instruct",
5054
- api: "openai-completions",
5055
- provider: "groq",
5056
- baseUrl: "https://api.groq.com/openai/v1",
5057
- reasoning: false,
5058
- input: ["text"],
5059
- cost: {
5060
- input: 1,
5061
- output: 3,
5062
- cacheRead: 0,
5063
- cacheWrite: 0,
5064
- },
5065
- contextWindow: 131072,
5066
- maxTokens: 16384,
5067
- },
5068
- "moonshotai/kimi-k2-instruct-0905": {
5069
- id: "moonshotai/kimi-k2-instruct-0905",
5070
- name: "Kimi K2 Instruct 0905",
5071
- api: "openai-completions",
5072
- provider: "groq",
5073
- baseUrl: "https://api.groq.com/openai/v1",
5074
- reasoning: false,
5075
- input: ["text"],
5076
- cost: {
5077
- input: 1,
5078
- output: 3,
5079
- cacheRead: 0.5,
5080
- cacheWrite: 0,
5081
- },
5082
- contextWindow: 262144,
5083
- maxTokens: 16384,
5084
- },
5085
5000
  "openai/gpt-oss-120b": {
5086
5001
  id: "openai/gpt-oss-120b",
5087
5002
  name: "GPT OSS 120B",
@@ -5133,26 +5048,9 @@ export const MODELS = {
5133
5048
  contextWindow: 131072,
5134
5049
  maxTokens: 65536,
5135
5050
  },
5136
- "qwen-qwq-32b": {
5137
- id: "qwen-qwq-32b",
5138
- name: "Qwen QwQ 32B",
5139
- api: "openai-completions",
5140
- provider: "groq",
5141
- baseUrl: "https://api.groq.com/openai/v1",
5142
- reasoning: true,
5143
- input: ["text"],
5144
- cost: {
5145
- input: 0.29,
5146
- output: 0.39,
5147
- cacheRead: 0,
5148
- cacheWrite: 0,
5149
- },
5150
- contextWindow: 131072,
5151
- maxTokens: 16384,
5152
- },
5153
5051
  "qwen/qwen3-32b": {
5154
5052
  id: "qwen/qwen3-32b",
5155
- name: "Qwen3 32B",
5053
+ name: "Qwen3-32B",
5156
5054
  api: "openai-completions",
5157
5055
  provider: "groq",
5158
5056
  baseUrl: "https://api.groq.com/openai/v1",
@@ -7313,6 +7211,23 @@ export const MODELS = {
7313
7211
  contextWindow: 200000,
7314
7212
  maxTokens: 32000,
7315
7213
  },
7214
+ "claude-fable-5": {
7215
+ id: "claude-fable-5",
7216
+ name: "Claude Fable 5",
7217
+ api: "anthropic-messages",
7218
+ provider: "opencode",
7219
+ baseUrl: "https://opencode.ai/zen",
7220
+ reasoning: true,
7221
+ input: ["text", "image"],
7222
+ cost: {
7223
+ input: 10,
7224
+ output: 50,
7225
+ cacheRead: 1,
7226
+ cacheWrite: 12.5,
7227
+ },
7228
+ contextWindow: 1000000,
7229
+ maxTokens: 128000,
7230
+ },
7316
7231
  "claude-haiku-4-5": {
7317
7232
  id: "claude-haiku-4-5",
7318
7233
  name: "Claude Haiku 4.5",
@@ -7485,7 +7400,7 @@ export const MODELS = {
7485
7400
  cost: {
7486
7401
  input: 0.14,
7487
7402
  output: 0.28,
7488
- cacheRead: 0.03,
7403
+ cacheRead: 0.028,
7489
7404
  cacheWrite: 0,
7490
7405
  },
7491
7406
  contextWindow: 1000000,
@@ -7510,6 +7425,25 @@ export const MODELS = {
7510
7425
  contextWindow: 200000,
7511
7426
  maxTokens: 128000,
7512
7427
  },
7428
+ "deepseek-v4-pro": {
7429
+ id: "deepseek-v4-pro",
7430
+ name: "DeepSeek V4 Pro",
7431
+ api: "openai-completions",
7432
+ provider: "opencode",
7433
+ baseUrl: "https://opencode.ai/zen/v1",
7434
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7435
+ reasoning: true,
7436
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7437
+ input: ["text"],
7438
+ cost: {
7439
+ input: 1.74,
7440
+ output: 3.84,
7441
+ cacheRead: 0.145,
7442
+ cacheWrite: 0,
7443
+ },
7444
+ contextWindow: 1000000,
7445
+ maxTokens: 384000,
7446
+ },
7513
7447
  "gemini-3-flash": {
7514
7448
  id: "gemini-3-flash",
7515
7449
  name: "Gemini 3 Flash",
@@ -8008,6 +7942,23 @@ export const MODELS = {
8008
7942
  contextWindow: 1000000,
8009
7943
  maxTokens: 128000,
8010
7944
  },
7945
+ "north-mini-code-free": {
7946
+ id: "north-mini-code-free",
7947
+ name: "North Mini Code Free",
7948
+ api: "openai-completions",
7949
+ provider: "opencode",
7950
+ baseUrl: "https://opencode.ai/zen/v1",
7951
+ reasoning: true,
7952
+ input: ["text"],
7953
+ cost: {
7954
+ input: 0,
7955
+ output: 0,
7956
+ cacheRead: 0,
7957
+ cacheWrite: 0,
7958
+ },
7959
+ contextWindow: 256000,
7960
+ maxTokens: 64000,
7961
+ },
8011
7962
  "qwen3.5-plus": {
8012
7963
  id: "qwen3.5-plus",
8013
7964
  name: "Qwen3.5 Plus",
@@ -8229,9 +8180,9 @@ export const MODELS = {
8229
8180
  reasoning: true,
8230
8181
  input: ["text", "image"],
8231
8182
  cost: {
8232
- input: 0.6,
8233
- output: 2.4,
8234
- cacheRead: 0.12,
8183
+ input: 0.3,
8184
+ output: 1.2,
8185
+ cacheRead: 0.06,
8235
8186
  cacheWrite: 0,
8236
8187
  },
8237
8188
  contextWindow: 512000,
@@ -8427,6 +8378,23 @@ export const MODELS = {
8427
8378
  contextWindow: 200000,
8428
8379
  maxTokens: 8192,
8429
8380
  },
8381
+ "anthropic/claude-fable-5": {
8382
+ id: "anthropic/claude-fable-5",
8383
+ name: "Anthropic: Claude Fable 5",
8384
+ api: "openai-completions",
8385
+ provider: "openrouter",
8386
+ baseUrl: "https://openrouter.ai/api/v1",
8387
+ reasoning: true,
8388
+ input: ["text", "image"],
8389
+ cost: {
8390
+ input: 10,
8391
+ output: 50,
8392
+ cacheRead: 1,
8393
+ cacheWrite: 12.5,
8394
+ },
8395
+ contextWindow: 1000000,
8396
+ maxTokens: 128000,
8397
+ },
8430
8398
  "anthropic/claude-haiku-4.5": {
8431
8399
  id: "anthropic/claude-haiku-4.5",
8432
8400
  name: "Anthropic: Claude Haiku 4.5",
@@ -8853,7 +8821,7 @@ export const MODELS = {
8853
8821
  cacheRead: 0.135,
8854
8822
  cacheWrite: 0,
8855
8823
  },
8856
- contextWindow: 163840,
8824
+ contextWindow: 131072,
8857
8825
  maxTokens: 16384,
8858
8826
  },
8859
8827
  "deepseek/deepseek-chat-v3.1": {
@@ -9585,8 +9553,8 @@ export const MODELS = {
9585
9553
  input: ["text"],
9586
9554
  cost: {
9587
9555
  input: 0.15,
9588
- output: 1.15,
9589
- cacheRead: 0,
9556
+ output: 0.8999999999999999,
9557
+ cacheRead: 0.049999999999999996,
9590
9558
  cacheWrite: 0,
9591
9559
  },
9592
9560
  contextWindow: 204800,
@@ -9601,13 +9569,13 @@ export const MODELS = {
9601
9569
  reasoning: true,
9602
9570
  input: ["text"],
9603
9571
  cost: {
9604
- input: 0.27899999999999997,
9605
- output: 1.2,
9606
- cacheRead: 0,
9572
+ input: 0.27,
9573
+ output: 1.08,
9574
+ cacheRead: 0.054,
9607
9575
  cacheWrite: 0,
9608
9576
  },
9609
9577
  contextWindow: 204800,
9610
- maxTokens: 196608,
9578
+ maxTokens: 131072,
9611
9579
  },
9612
9580
  "minimax/minimax-m3": {
9613
9581
  id: "minimax/minimax-m3",
@@ -9993,21 +9961,20 @@ export const MODELS = {
9993
9961
  reasoning: true,
9994
9962
  input: ["text", "image"],
9995
9963
  cost: {
9996
- input: 0.684,
9997
- output: 3.42,
9998
- cacheRead: 0.144,
9964
+ input: 0.6799999999999999,
9965
+ output: 3.41,
9966
+ cacheRead: 0.33999999999999997,
9999
9967
  cacheWrite: 0,
10000
9968
  },
10001
9969
  contextWindow: 262144,
10002
- maxTokens: 262144,
9970
+ maxTokens: 262142,
10003
9971
  },
10004
- "moonshotai/kimi-k2.6:free": {
10005
- id: "moonshotai/kimi-k2.6:free",
10006
- name: "MoonshotAI: Kimi K2.6 (free)",
9972
+ "nex-agi/nex-n2-pro:free": {
9973
+ id: "nex-agi/nex-n2-pro:free",
9974
+ name: "Nex AGI: Nex-N2-Pro (free)",
10007
9975
  api: "openai-completions",
10008
9976
  provider: "openrouter",
10009
9977
  baseUrl: "https://openrouter.ai/api/v1",
10010
- compat: { "supportsDeveloperRole": false },
10011
9978
  reasoning: true,
10012
9979
  input: ["text", "image"],
10013
9980
  cost: {
@@ -10017,24 +9984,7 @@ export const MODELS = {
10017
9984
  cacheWrite: 0,
10018
9985
  },
10019
9986
  contextWindow: 262144,
10020
- maxTokens: 4096,
10021
- },
10022
- "nex-agi/deepseek-v3.1-nex-n1": {
10023
- id: "nex-agi/deepseek-v3.1-nex-n1",
10024
- name: "Nex AGI: DeepSeek V3.1 Nex N1",
10025
- api: "openai-completions",
10026
- provider: "openrouter",
10027
- baseUrl: "https://openrouter.ai/api/v1",
10028
- reasoning: false,
10029
- input: ["text"],
10030
- cost: {
10031
- input: 0.135,
10032
- output: 0.5,
10033
- cacheRead: 0,
10034
- cacheWrite: 0,
10035
- },
10036
- contextWindow: 131072,
10037
- maxTokens: 163840,
9987
+ maxTokens: 262144,
10038
9988
  },
10039
9989
  "nvidia/llama-3.3-nemotron-super-49b-v1.5": {
10040
9990
  id: "nvidia/llama-3.3-nemotron-super-49b-v1.5",
@@ -11944,13 +11894,13 @@ export const MODELS = {
11944
11894
  reasoning: true,
11945
11895
  input: ["text", "image"],
11946
11896
  cost: {
11947
- input: 0.14,
11897
+ input: 0.15,
11948
11898
  output: 1,
11949
- cacheRead: 0,
11899
+ cacheRead: 0.049999999999999996,
11950
11900
  cacheWrite: 0,
11951
11901
  },
11952
11902
  contextWindow: 262144,
11953
- maxTokens: 262140,
11903
+ maxTokens: 262144,
11954
11904
  },
11955
11905
  "qwen/qwen3.6-flash": {
11956
11906
  id: "qwen/qwen3.6-flash",
@@ -12292,23 +12242,6 @@ export const MODELS = {
12292
12242
  contextWindow: 1048576,
12293
12243
  maxTokens: 131072,
12294
12244
  },
12295
- "z-ai/glm-4-32b": {
12296
- id: "z-ai/glm-4-32b",
12297
- name: "Z.ai: GLM 4 32B ",
12298
- api: "openai-completions",
12299
- provider: "openrouter",
12300
- baseUrl: "https://openrouter.ai/api/v1",
12301
- reasoning: false,
12302
- input: ["text"],
12303
- cost: {
12304
- input: 0.09999999999999999,
12305
- output: 0.09999999999999999,
12306
- cacheRead: 0,
12307
- cacheWrite: 0,
12308
- },
12309
- contextWindow: 128000,
12310
- maxTokens: 4096,
12311
- },
12312
12245
  "z-ai/glm-4.5": {
12313
12246
  id: "z-ai/glm-4.5",
12314
12247
  name: "Z.ai: GLM 4.5",
@@ -12343,23 +12276,6 @@ export const MODELS = {
12343
12276
  contextWindow: 131072,
12344
12277
  maxTokens: 131070,
12345
12278
  },
12346
- "z-ai/glm-4.5-air:free": {
12347
- id: "z-ai/glm-4.5-air:free",
12348
- name: "Z.ai: GLM 4.5 Air (free)",
12349
- api: "openai-completions",
12350
- provider: "openrouter",
12351
- baseUrl: "https://openrouter.ai/api/v1",
12352
- reasoning: true,
12353
- input: ["text"],
12354
- cost: {
12355
- input: 0,
12356
- output: 0,
12357
- cacheRead: 0,
12358
- cacheWrite: 0,
12359
- },
12360
- contextWindow: 131072,
12361
- maxTokens: 96000,
12362
- },
12363
12279
  "z-ai/glm-4.5v": {
12364
12280
  id: "z-ai/glm-4.5v",
12365
12281
  name: "Z.ai: GLM 4.5V",
@@ -12405,11 +12321,11 @@ export const MODELS = {
12405
12321
  cost: {
12406
12322
  input: 0.3,
12407
12323
  output: 0.8999999999999999,
12408
- cacheRead: 0.049999999999999996,
12324
+ cacheRead: 0.055,
12409
12325
  cacheWrite: 0,
12410
12326
  },
12411
12327
  contextWindow: 131072,
12412
- maxTokens: 24000,
12328
+ maxTokens: 32768,
12413
12329
  },
12414
12330
  "z-ai/glm-4.7": {
12415
12331
  id: "z-ai/glm-4.7",
@@ -12476,7 +12392,7 @@ export const MODELS = {
12476
12392
  cacheRead: 0.24,
12477
12393
  cacheWrite: 0,
12478
12394
  },
12479
- contextWindow: 202752,
12395
+ contextWindow: 262144,
12480
12396
  maxTokens: 131072,
12481
12397
  },
12482
12398
  "z-ai/glm-5.1": {
@@ -12496,22 +12412,22 @@ export const MODELS = {
12496
12412
  contextWindow: 202752,
12497
12413
  maxTokens: 4096,
12498
12414
  },
12499
- "z-ai/glm-5v-turbo": {
12500
- id: "z-ai/glm-5v-turbo",
12501
- name: "Z.ai: GLM 5V Turbo",
12415
+ "~anthropic/claude-fable-latest": {
12416
+ id: "~anthropic/claude-fable-latest",
12417
+ name: "Anthropic: Claude Fable Latest",
12502
12418
  api: "openai-completions",
12503
12419
  provider: "openrouter",
12504
12420
  baseUrl: "https://openrouter.ai/api/v1",
12505
12421
  reasoning: true,
12506
12422
  input: ["text", "image"],
12507
12423
  cost: {
12508
- input: 1.2,
12509
- output: 4,
12510
- cacheRead: 0.24,
12511
- cacheWrite: 0,
12424
+ input: 10,
12425
+ output: 50,
12426
+ cacheRead: 1,
12427
+ cacheWrite: 12.5,
12512
12428
  },
12513
- contextWindow: 202752,
12514
- maxTokens: 131072,
12429
+ contextWindow: 1000000,
12430
+ maxTokens: 128000,
12515
12431
  },
12516
12432
  "~anthropic/claude-haiku-latest": {
12517
12433
  id: "~anthropic/claude-haiku-latest",
@@ -12607,13 +12523,13 @@ export const MODELS = {
12607
12523
  reasoning: true,
12608
12524
  input: ["text", "image"],
12609
12525
  cost: {
12610
- input: 0.684,
12611
- output: 3.42,
12612
- cacheRead: 0.144,
12526
+ input: 0.6799999999999999,
12527
+ output: 3.41,
12528
+ cacheRead: 0.33999999999999997,
12613
12529
  cacheWrite: 0,
12614
12530
  },
12615
12531
  contextWindow: 262144,
12616
- maxTokens: 262144,
12532
+ maxTokens: 262142,
12617
12533
  },
12618
12534
  "~openai/gpt-latest": {
12619
12535
  id: "~openai/gpt-latest",
@@ -13054,8 +12970,8 @@ export const MODELS = {
13054
12970
  reasoning: true,
13055
12971
  input: ["text"],
13056
12972
  cost: {
13057
- input: 0.08,
13058
- output: 0.29,
12973
+ input: 0.12,
12974
+ output: 0.5,
13059
12975
  cacheRead: 0,
13060
12976
  cacheWrite: 0,
13061
12977
  },
@@ -13419,6 +13335,23 @@ export const MODELS = {
13419
13335
  contextWindow: 200000,
13420
13336
  maxTokens: 8192,
13421
13337
  },
13338
+ "anthropic/claude-fable-5": {
13339
+ id: "anthropic/claude-fable-5",
13340
+ name: "Claude Fable 5",
13341
+ api: "anthropic-messages",
13342
+ provider: "vercel-ai-gateway",
13343
+ baseUrl: "https://ai-gateway.vercel.sh",
13344
+ reasoning: true,
13345
+ input: ["text", "image"],
13346
+ cost: {
13347
+ input: 10,
13348
+ output: 50,
13349
+ cacheRead: 1,
13350
+ cacheWrite: 12.5,
13351
+ },
13352
+ contextWindow: 1000000,
13353
+ maxTokens: 128000,
13354
+ },
13422
13355
  "anthropic/claude-haiku-4.5": {
13423
13356
  id: "anthropic/claude-haiku-4.5",
13424
13357
  name: "Claude Haiku 4.5",
@@ -13799,40 +13732,6 @@ export const MODELS = {
13799
13732
  contextWindow: 1000000,
13800
13733
  maxTokens: 384000,
13801
13734
  },
13802
- "google/gemini-2.0-flash": {
13803
- id: "google/gemini-2.0-flash",
13804
- name: "Gemini 2.0 Flash",
13805
- api: "anthropic-messages",
13806
- provider: "vercel-ai-gateway",
13807
- baseUrl: "https://ai-gateway.vercel.sh",
13808
- reasoning: false,
13809
- input: ["text", "image"],
13810
- cost: {
13811
- input: 0.15,
13812
- output: 0.6,
13813
- cacheRead: 0.024999999999999998,
13814
- cacheWrite: 0,
13815
- },
13816
- contextWindow: 1048576,
13817
- maxTokens: 8192,
13818
- },
13819
- "google/gemini-2.0-flash-lite": {
13820
- id: "google/gemini-2.0-flash-lite",
13821
- name: "Gemini 2.0 Flash Lite",
13822
- api: "anthropic-messages",
13823
- provider: "vercel-ai-gateway",
13824
- baseUrl: "https://ai-gateway.vercel.sh",
13825
- reasoning: false,
13826
- input: ["text", "image"],
13827
- cost: {
13828
- input: 0.075,
13829
- output: 0.3,
13830
- cacheRead: 0.02,
13831
- cacheWrite: 0,
13832
- },
13833
- contextWindow: 1048576,
13834
- maxTokens: 8192,
13835
- },
13836
13735
  "google/gemini-2.5-flash": {
13837
13736
  id: "google/gemini-2.5-flash",
13838
13737
  name: "Gemini 2.5 Flash",
@@ -15952,7 +15851,7 @@ export const MODELS = {
15952
15851
  cacheRead: 0.2,
15953
15852
  cacheWrite: 0,
15954
15853
  },
15955
- contextWindow: 2000000,
15854
+ contextWindow: 1000000,
15956
15855
  maxTokens: 30000,
15957
15856
  },
15958
15857
  "grok-4.20-0309-reasoning": {
@@ -15969,7 +15868,7 @@ export const MODELS = {
15969
15868
  cacheRead: 0.2,
15970
15869
  cacheWrite: 0,
15971
15870
  },
15972
- contextWindow: 2000000,
15871
+ contextWindow: 1000000,
15973
15872
  maxTokens: 30000,
15974
15873
  },
15975
15874
  "grok-4.3": {
@@ -16115,6 +16014,24 @@ export const MODELS = {
16115
16014
  contextWindow: 1048576,
16116
16015
  maxTokens: 131072,
16117
16016
  },
16017
+ "mimo-v2.5-pro-ultraspeed": {
16018
+ id: "mimo-v2.5-pro-ultraspeed",
16019
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16020
+ api: "openai-completions",
16021
+ provider: "xiaomi",
16022
+ baseUrl: "https://api.xiaomimimo.com/v1",
16023
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16024
+ reasoning: true,
16025
+ input: ["text"],
16026
+ cost: {
16027
+ input: 1.305,
16028
+ output: 2.61,
16029
+ cacheRead: 0.0108,
16030
+ cacheWrite: 0,
16031
+ },
16032
+ contextWindow: 1048576,
16033
+ maxTokens: 131072,
16034
+ },
16118
16035
  },
16119
16036
  "xiaomi-token-plan-ams": {
16120
16037
  "mimo-v2-omni": {
@@ -16189,6 +16106,24 @@ export const MODELS = {
16189
16106
  contextWindow: 1048576,
16190
16107
  maxTokens: 131072,
16191
16108
  },
16109
+ "mimo-v2.5-pro-ultraspeed": {
16110
+ id: "mimo-v2.5-pro-ultraspeed",
16111
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16112
+ api: "openai-completions",
16113
+ provider: "xiaomi-token-plan-ams",
16114
+ baseUrl: "https://token-plan-ams.xiaomimimo.com/v1",
16115
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16116
+ reasoning: true,
16117
+ input: ["text"],
16118
+ cost: {
16119
+ input: 1.305,
16120
+ output: 2.61,
16121
+ cacheRead: 0.0108,
16122
+ cacheWrite: 0,
16123
+ },
16124
+ contextWindow: 1048576,
16125
+ maxTokens: 131072,
16126
+ },
16192
16127
  },
16193
16128
  "xiaomi-token-plan-cn": {
16194
16129
  "mimo-v2-omni": {
@@ -16263,6 +16198,24 @@ export const MODELS = {
16263
16198
  contextWindow: 1048576,
16264
16199
  maxTokens: 131072,
16265
16200
  },
16201
+ "mimo-v2.5-pro-ultraspeed": {
16202
+ id: "mimo-v2.5-pro-ultraspeed",
16203
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16204
+ api: "openai-completions",
16205
+ provider: "xiaomi-token-plan-cn",
16206
+ baseUrl: "https://token-plan-cn.xiaomimimo.com/v1",
16207
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16208
+ reasoning: true,
16209
+ input: ["text"],
16210
+ cost: {
16211
+ input: 1.305,
16212
+ output: 2.61,
16213
+ cacheRead: 0.0108,
16214
+ cacheWrite: 0,
16215
+ },
16216
+ contextWindow: 1048576,
16217
+ maxTokens: 131072,
16218
+ },
16266
16219
  },
16267
16220
  "xiaomi-token-plan-sgp": {
16268
16221
  "mimo-v2-omni": {
@@ -16337,6 +16290,24 @@ export const MODELS = {
16337
16290
  contextWindow: 1048576,
16338
16291
  maxTokens: 131072,
16339
16292
  },
16293
+ "mimo-v2.5-pro-ultraspeed": {
16294
+ id: "mimo-v2.5-pro-ultraspeed",
16295
+ name: "MiMo-V2.5-Pro-UltraSpeed",
16296
+ api: "openai-completions",
16297
+ provider: "xiaomi-token-plan-sgp",
16298
+ baseUrl: "https://token-plan-sgp.xiaomimimo.com/v1",
16299
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
16300
+ reasoning: true,
16301
+ input: ["text"],
16302
+ cost: {
16303
+ input: 1.305,
16304
+ output: 2.61,
16305
+ cacheRead: 0.0108,
16306
+ cacheWrite: 0,
16307
+ },
16308
+ contextWindow: 1048576,
16309
+ maxTokens: 131072,
16310
+ },
16340
16311
  },
16341
16312
  "zai": {
16342
16313
  "glm-4.5-air": {