@mariozechner/pi-ai 0.9.3 → 0.9.4

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.
@@ -106,7 +106,7 @@ export const MODELS = {
106
106
  },
107
107
  "claude-opus-4-5": {
108
108
  id: "claude-opus-4-5",
109
- name: "Claude Opus 4.5",
109
+ name: "Claude Opus 4.5 (latest)",
110
110
  api: "anthropic-messages",
111
111
  provider: "anthropic",
112
112
  baseUrl: "https://api.anthropic.com",
@@ -1988,23 +1988,6 @@ export const MODELS = {
1988
1988
  contextWindow: 65536,
1989
1989
  maxTokens: 65536,
1990
1990
  },
1991
- "x-ai/grok-4.1-fast": {
1992
- id: "x-ai/grok-4.1-fast",
1993
- name: "xAI: Grok 4.1 Fast",
1994
- api: "openai-completions",
1995
- provider: "openrouter",
1996
- baseUrl: "https://openrouter.ai/api/v1",
1997
- reasoning: true,
1998
- input: ["text", "image"],
1999
- cost: {
2000
- input: 0,
2001
- output: 0,
2002
- cacheRead: 0,
2003
- cacheWrite: 0,
2004
- },
2005
- contextWindow: 2000000,
2006
- maxTokens: 30000,
2007
- },
2008
1991
  "x-ai/grok-4.1-fast:free": {
2009
1992
  id: "x-ai/grok-4.1-fast:free",
2010
1993
  name: "xAI: Grok 4.1 Fast (free)",
@@ -2122,7 +2105,7 @@ export const MODELS = {
2122
2105
  cacheWrite: 0,
2123
2106
  },
2124
2107
  contextWindow: 256000,
2125
- maxTokens: 32000,
2108
+ maxTokens: 32768,
2126
2109
  },
2127
2110
  "moonshotai/kimi-k2-thinking": {
2128
2111
  id: "moonshotai/kimi-k2-thinking",
@@ -3442,13 +3425,13 @@ export const MODELS = {
3442
3425
  reasoning: true,
3443
3426
  input: ["text"],
3444
3427
  cost: {
3445
- input: 1,
3446
- output: 4,
3428
+ input: 0.38,
3429
+ output: 1.53,
3447
3430
  cacheRead: 0,
3448
3431
  cacheWrite: 0,
3449
3432
  },
3450
3433
  contextWindow: 262144,
3451
- maxTokens: 32768,
3434
+ maxTokens: 262144,
3452
3435
  },
3453
3436
  "google/gemini-2.5-flash-lite": {
3454
3437
  id: "google/gemini-2.5-flash-lite",
@@ -3884,8 +3867,8 @@ export const MODELS = {
3884
3867
  reasoning: true,
3885
3868
  input: ["text"],
3886
3869
  cost: {
3887
- input: 0.15,
3888
- output: 0.59,
3870
+ input: 0.049999999999999996,
3871
+ output: 0.19999999999999998,
3889
3872
  cacheRead: 0,
3890
3873
  cacheWrite: 0,
3891
3874
  },
@@ -4827,9 +4810,9 @@ export const MODELS = {
4827
4810
  contextWindow: 32768,
4828
4811
  maxTokens: 4096,
4829
4812
  },
4830
- "anthropic/claude-3.5-haiku-20241022": {
4831
- id: "anthropic/claude-3.5-haiku-20241022",
4832
- name: "Anthropic: Claude 3.5 Haiku (2024-10-22)",
4813
+ "anthropic/claude-3.5-haiku": {
4814
+ id: "anthropic/claude-3.5-haiku",
4815
+ name: "Anthropic: Claude 3.5 Haiku",
4833
4816
  api: "openai-completions",
4834
4817
  provider: "openrouter",
4835
4818
  baseUrl: "https://openrouter.ai/api/v1",
@@ -4844,9 +4827,9 @@ export const MODELS = {
4844
4827
  contextWindow: 200000,
4845
4828
  maxTokens: 8192,
4846
4829
  },
4847
- "anthropic/claude-3.5-haiku": {
4848
- id: "anthropic/claude-3.5-haiku",
4849
- name: "Anthropic: Claude 3.5 Haiku",
4830
+ "anthropic/claude-3.5-haiku-20241022": {
4831
+ id: "anthropic/claude-3.5-haiku-20241022",
4832
+ name: "Anthropic: Claude 3.5 Haiku (2024-10-22)",
4850
4833
  api: "openai-completions",
4851
4834
  provider: "openrouter",
4852
4835
  baseUrl: "https://openrouter.ai/api/v1",
@@ -4912,23 +4895,6 @@ export const MODELS = {
4912
4895
  contextWindow: 131072,
4913
4896
  maxTokens: 4096,
4914
4897
  },
4915
- "qwen/qwen-2.5-7b-instruct": {
4916
- id: "qwen/qwen-2.5-7b-instruct",
4917
- name: "Qwen: Qwen2.5 7B Instruct",
4918
- api: "openai-completions",
4919
- provider: "openrouter",
4920
- baseUrl: "https://openrouter.ai/api/v1",
4921
- reasoning: false,
4922
- input: ["text"],
4923
- cost: {
4924
- input: 0.04,
4925
- output: 0.09999999999999999,
4926
- cacheRead: 0,
4927
- cacheWrite: 0,
4928
- },
4929
- contextWindow: 32768,
4930
- maxTokens: 4096,
4931
- },
4932
4898
  "nvidia/llama-3.1-nemotron-70b-instruct": {
4933
4899
  id: "nvidia/llama-3.1-nemotron-70b-instruct",
4934
4900
  name: "NVIDIA: Llama 3.1 Nemotron 70B Instruct",
@@ -5014,34 +4980,34 @@ export const MODELS = {
5014
4980
  contextWindow: 32768,
5015
4981
  maxTokens: 4096,
5016
4982
  },
5017
- "cohere/command-r-plus-08-2024": {
5018
- id: "cohere/command-r-plus-08-2024",
5019
- name: "Cohere: Command R+ (08-2024)",
4983
+ "cohere/command-r-08-2024": {
4984
+ id: "cohere/command-r-08-2024",
4985
+ name: "Cohere: Command R (08-2024)",
5020
4986
  api: "openai-completions",
5021
4987
  provider: "openrouter",
5022
4988
  baseUrl: "https://openrouter.ai/api/v1",
5023
4989
  reasoning: false,
5024
4990
  input: ["text"],
5025
4991
  cost: {
5026
- input: 2.5,
5027
- output: 10,
4992
+ input: 0.15,
4993
+ output: 0.6,
5028
4994
  cacheRead: 0,
5029
4995
  cacheWrite: 0,
5030
4996
  },
5031
4997
  contextWindow: 128000,
5032
4998
  maxTokens: 4000,
5033
4999
  },
5034
- "cohere/command-r-08-2024": {
5035
- id: "cohere/command-r-08-2024",
5036
- name: "Cohere: Command R (08-2024)",
5000
+ "cohere/command-r-plus-08-2024": {
5001
+ id: "cohere/command-r-plus-08-2024",
5002
+ name: "Cohere: Command R+ (08-2024)",
5037
5003
  api: "openai-completions",
5038
5004
  provider: "openrouter",
5039
5005
  baseUrl: "https://openrouter.ai/api/v1",
5040
5006
  reasoning: false,
5041
5007
  input: ["text"],
5042
5008
  cost: {
5043
- input: 0.15,
5044
- output: 0.6,
5009
+ input: 2.5,
5010
+ output: 10,
5045
5011
  cacheRead: 0,
5046
5012
  cacheWrite: 0,
5047
5013
  },
@@ -5116,6 +5082,23 @@ export const MODELS = {
5116
5082
  contextWindow: 128000,
5117
5083
  maxTokens: 16384,
5118
5084
  },
5085
+ "meta-llama/llama-3.1-8b-instruct": {
5086
+ id: "meta-llama/llama-3.1-8b-instruct",
5087
+ name: "Meta: Llama 3.1 8B Instruct",
5088
+ api: "openai-completions",
5089
+ provider: "openrouter",
5090
+ baseUrl: "https://openrouter.ai/api/v1",
5091
+ reasoning: false,
5092
+ input: ["text"],
5093
+ cost: {
5094
+ input: 0.02,
5095
+ output: 0.03,
5096
+ cacheRead: 0,
5097
+ cacheWrite: 0,
5098
+ },
5099
+ contextWindow: 131072,
5100
+ maxTokens: 16384,
5101
+ },
5119
5102
  "meta-llama/llama-3.1-405b-instruct": {
5120
5103
  id: "meta-llama/llama-3.1-405b-instruct",
5121
5104
  name: "Meta: Llama 3.1 405B Instruct",
@@ -5150,23 +5133,6 @@ export const MODELS = {
5150
5133
  contextWindow: 131072,
5151
5134
  maxTokens: 4096,
5152
5135
  },
5153
- "meta-llama/llama-3.1-8b-instruct": {
5154
- id: "meta-llama/llama-3.1-8b-instruct",
5155
- name: "Meta: Llama 3.1 8B Instruct",
5156
- api: "openai-completions",
5157
- provider: "openrouter",
5158
- baseUrl: "https://openrouter.ai/api/v1",
5159
- reasoning: false,
5160
- input: ["text"],
5161
- cost: {
5162
- input: 0.02,
5163
- output: 0.03,
5164
- cacheRead: 0,
5165
- cacheWrite: 0,
5166
- },
5167
- contextWindow: 131072,
5168
- maxTokens: 16384,
5169
- },
5170
5136
  "mistralai/mistral-nemo": {
5171
5137
  id: "mistralai/mistral-nemo",
5172
5138
  name: "Mistral: Mistral Nemo",
@@ -5184,9 +5150,9 @@ export const MODELS = {
5184
5150
  contextWindow: 131072,
5185
5151
  maxTokens: 16384,
5186
5152
  },
5187
- "openai/gpt-4o-mini-2024-07-18": {
5188
- id: "openai/gpt-4o-mini-2024-07-18",
5189
- name: "OpenAI: GPT-4o-mini (2024-07-18)",
5153
+ "openai/gpt-4o-mini": {
5154
+ id: "openai/gpt-4o-mini",
5155
+ name: "OpenAI: GPT-4o-mini",
5190
5156
  api: "openai-completions",
5191
5157
  provider: "openrouter",
5192
5158
  baseUrl: "https://openrouter.ai/api/v1",
@@ -5201,9 +5167,9 @@ export const MODELS = {
5201
5167
  contextWindow: 128000,
5202
5168
  maxTokens: 16384,
5203
5169
  },
5204
- "openai/gpt-4o-mini": {
5205
- id: "openai/gpt-4o-mini",
5206
- name: "OpenAI: GPT-4o-mini",
5170
+ "openai/gpt-4o-mini-2024-07-18": {
5171
+ id: "openai/gpt-4o-mini-2024-07-18",
5172
+ name: "OpenAI: GPT-4o-mini (2024-07-18)",
5207
5173
  api: "openai-completions",
5208
5174
  provider: "openrouter",
5209
5175
  baseUrl: "https://openrouter.ai/api/v1",
@@ -5303,23 +5269,6 @@ export const MODELS = {
5303
5269
  contextWindow: 128000,
5304
5270
  maxTokens: 4096,
5305
5271
  },
5306
- "openai/gpt-4o-2024-05-13": {
5307
- id: "openai/gpt-4o-2024-05-13",
5308
- name: "OpenAI: GPT-4o (2024-05-13)",
5309
- api: "openai-completions",
5310
- provider: "openrouter",
5311
- baseUrl: "https://openrouter.ai/api/v1",
5312
- reasoning: false,
5313
- input: ["text", "image"],
5314
- cost: {
5315
- input: 5,
5316
- output: 15,
5317
- cacheRead: 0,
5318
- cacheWrite: 0,
5319
- },
5320
- contextWindow: 128000,
5321
- maxTokens: 4096,
5322
- },
5323
5272
  "openai/gpt-4o": {
5324
5273
  id: "openai/gpt-4o",
5325
5274
  name: "OpenAI: GPT-4o",
@@ -5354,22 +5303,22 @@ export const MODELS = {
5354
5303
  contextWindow: 128000,
5355
5304
  maxTokens: 64000,
5356
5305
  },
5357
- "meta-llama/llama-3-70b-instruct": {
5358
- id: "meta-llama/llama-3-70b-instruct",
5359
- name: "Meta: Llama 3 70B Instruct",
5306
+ "openai/gpt-4o-2024-05-13": {
5307
+ id: "openai/gpt-4o-2024-05-13",
5308
+ name: "OpenAI: GPT-4o (2024-05-13)",
5360
5309
  api: "openai-completions",
5361
5310
  provider: "openrouter",
5362
5311
  baseUrl: "https://openrouter.ai/api/v1",
5363
5312
  reasoning: false,
5364
- input: ["text"],
5313
+ input: ["text", "image"],
5365
5314
  cost: {
5366
- input: 0.3,
5367
- output: 0.39999999999999997,
5315
+ input: 5,
5316
+ output: 15,
5368
5317
  cacheRead: 0,
5369
5318
  cacheWrite: 0,
5370
5319
  },
5371
- contextWindow: 8192,
5372
- maxTokens: 16384,
5320
+ contextWindow: 128000,
5321
+ maxTokens: 4096,
5373
5322
  },
5374
5323
  "meta-llama/llama-3-8b-instruct": {
5375
5324
  id: "meta-llama/llama-3-8b-instruct",
@@ -5388,6 +5337,23 @@ export const MODELS = {
5388
5337
  contextWindow: 8192,
5389
5338
  maxTokens: 16384,
5390
5339
  },
5340
+ "meta-llama/llama-3-70b-instruct": {
5341
+ id: "meta-llama/llama-3-70b-instruct",
5342
+ name: "Meta: Llama 3 70B Instruct",
5343
+ api: "openai-completions",
5344
+ provider: "openrouter",
5345
+ baseUrl: "https://openrouter.ai/api/v1",
5346
+ reasoning: false,
5347
+ input: ["text"],
5348
+ cost: {
5349
+ input: 0.3,
5350
+ output: 0.39999999999999997,
5351
+ cacheRead: 0,
5352
+ cacheWrite: 0,
5353
+ },
5354
+ contextWindow: 8192,
5355
+ maxTokens: 16384,
5356
+ },
5391
5357
  "mistralai/mixtral-8x22b-instruct": {
5392
5358
  id: "mistralai/mixtral-8x22b-instruct",
5393
5359
  name: "Mistral: Mixtral 8x22B Instruct",
@@ -5473,38 +5439,38 @@ export const MODELS = {
5473
5439
  contextWindow: 128000,
5474
5440
  maxTokens: 4096,
5475
5441
  },
5476
- "openai/gpt-3.5-turbo-0613": {
5477
- id: "openai/gpt-3.5-turbo-0613",
5478
- name: "OpenAI: GPT-3.5 Turbo (older v0613)",
5442
+ "openai/gpt-4-turbo-preview": {
5443
+ id: "openai/gpt-4-turbo-preview",
5444
+ name: "OpenAI: GPT-4 Turbo Preview",
5479
5445
  api: "openai-completions",
5480
5446
  provider: "openrouter",
5481
5447
  baseUrl: "https://openrouter.ai/api/v1",
5482
5448
  reasoning: false,
5483
5449
  input: ["text"],
5484
5450
  cost: {
5485
- input: 1,
5486
- output: 2,
5451
+ input: 10,
5452
+ output: 30,
5487
5453
  cacheRead: 0,
5488
5454
  cacheWrite: 0,
5489
5455
  },
5490
- contextWindow: 4095,
5456
+ contextWindow: 128000,
5491
5457
  maxTokens: 4096,
5492
5458
  },
5493
- "openai/gpt-4-turbo-preview": {
5494
- id: "openai/gpt-4-turbo-preview",
5495
- name: "OpenAI: GPT-4 Turbo Preview",
5459
+ "openai/gpt-3.5-turbo-0613": {
5460
+ id: "openai/gpt-3.5-turbo-0613",
5461
+ name: "OpenAI: GPT-3.5 Turbo (older v0613)",
5496
5462
  api: "openai-completions",
5497
5463
  provider: "openrouter",
5498
5464
  baseUrl: "https://openrouter.ai/api/v1",
5499
5465
  reasoning: false,
5500
5466
  input: ["text"],
5501
5467
  cost: {
5502
- input: 10,
5503
- output: 30,
5468
+ input: 1,
5469
+ output: 2,
5504
5470
  cacheRead: 0,
5505
5471
  cacheWrite: 0,
5506
5472
  },
5507
- contextWindow: 128000,
5473
+ contextWindow: 4095,
5508
5474
  maxTokens: 4096,
5509
5475
  },
5510
5476
  "mistralai/mistral-small": {
@@ -5592,26 +5558,26 @@ export const MODELS = {
5592
5558
  contextWindow: 16385,
5593
5559
  maxTokens: 4096,
5594
5560
  },
5595
- "openai/gpt-3.5-turbo": {
5596
- id: "openai/gpt-3.5-turbo",
5597
- name: "OpenAI: GPT-3.5 Turbo",
5561
+ "openai/gpt-4-0314": {
5562
+ id: "openai/gpt-4-0314",
5563
+ name: "OpenAI: GPT-4 (older v0314)",
5598
5564
  api: "openai-completions",
5599
5565
  provider: "openrouter",
5600
5566
  baseUrl: "https://openrouter.ai/api/v1",
5601
5567
  reasoning: false,
5602
5568
  input: ["text"],
5603
5569
  cost: {
5604
- input: 0.5,
5605
- output: 1.5,
5570
+ input: 30,
5571
+ output: 60,
5606
5572
  cacheRead: 0,
5607
5573
  cacheWrite: 0,
5608
5574
  },
5609
- contextWindow: 16385,
5575
+ contextWindow: 8191,
5610
5576
  maxTokens: 4096,
5611
5577
  },
5612
- "openai/gpt-4-0314": {
5613
- id: "openai/gpt-4-0314",
5614
- name: "OpenAI: GPT-4 (older v0314)",
5578
+ "openai/gpt-4": {
5579
+ id: "openai/gpt-4",
5580
+ name: "OpenAI: GPT-4",
5615
5581
  api: "openai-completions",
5616
5582
  provider: "openrouter",
5617
5583
  baseUrl: "https://openrouter.ai/api/v1",
@@ -5626,21 +5592,21 @@ export const MODELS = {
5626
5592
  contextWindow: 8191,
5627
5593
  maxTokens: 4096,
5628
5594
  },
5629
- "openai/gpt-4": {
5630
- id: "openai/gpt-4",
5631
- name: "OpenAI: GPT-4",
5595
+ "openai/gpt-3.5-turbo": {
5596
+ id: "openai/gpt-3.5-turbo",
5597
+ name: "OpenAI: GPT-3.5 Turbo",
5632
5598
  api: "openai-completions",
5633
5599
  provider: "openrouter",
5634
5600
  baseUrl: "https://openrouter.ai/api/v1",
5635
5601
  reasoning: false,
5636
5602
  input: ["text"],
5637
5603
  cost: {
5638
- input: 30,
5639
- output: 60,
5604
+ input: 0.5,
5605
+ output: 1.5,
5640
5606
  cacheRead: 0,
5641
5607
  cacheWrite: 0,
5642
5608
  },
5643
- contextWindow: 8191,
5609
+ contextWindow: 16385,
5644
5610
  maxTokens: 4096,
5645
5611
  },
5646
5612
  "openrouter/auto": {