@mariozechner/pi-ai 0.9.3 → 0.10.0

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",
@@ -115,8 +115,8 @@ export const MODELS = {
115
115
  cost: {
116
116
  input: 5,
117
117
  output: 25,
118
- cacheRead: 1.5,
119
- cacheWrite: 18.75,
118
+ cacheRead: 0.5,
119
+ cacheWrite: 6.25,
120
120
  },
121
121
  contextWindow: 200000,
122
122
  maxTokens: 64000,
@@ -1954,6 +1954,40 @@ export const MODELS = {
1954
1954
  },
1955
1955
  },
1956
1956
  openrouter: {
1957
+ "tngtech/tng-r1t-chimera:free": {
1958
+ id: "tngtech/tng-r1t-chimera:free",
1959
+ name: "TNG: R1T Chimera (free)",
1960
+ api: "openai-completions",
1961
+ provider: "openrouter",
1962
+ baseUrl: "https://openrouter.ai/api/v1",
1963
+ reasoning: true,
1964
+ input: ["text"],
1965
+ cost: {
1966
+ input: 0,
1967
+ output: 0,
1968
+ cacheRead: 0,
1969
+ cacheWrite: 0,
1970
+ },
1971
+ contextWindow: 163840,
1972
+ maxTokens: 163840,
1973
+ },
1974
+ "tngtech/tng-r1t-chimera": {
1975
+ id: "tngtech/tng-r1t-chimera",
1976
+ name: "TNG: R1T Chimera",
1977
+ api: "openai-completions",
1978
+ provider: "openrouter",
1979
+ baseUrl: "https://openrouter.ai/api/v1",
1980
+ reasoning: true,
1981
+ input: ["text"],
1982
+ cost: {
1983
+ input: 0.3,
1984
+ output: 1.2,
1985
+ cacheRead: 0,
1986
+ cacheWrite: 0,
1987
+ },
1988
+ contextWindow: 163840,
1989
+ maxTokens: 163840,
1990
+ },
1957
1991
  "anthropic/claude-opus-4.5": {
1958
1992
  id: "anthropic/claude-opus-4.5",
1959
1993
  name: "Anthropic: Claude Opus 4.5",
@@ -1988,23 +2022,6 @@ export const MODELS = {
1988
2022
  contextWindow: 65536,
1989
2023
  maxTokens: 65536,
1990
2024
  },
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
2025
  "x-ai/grok-4.1-fast:free": {
2009
2026
  id: "x-ai/grok-4.1-fast:free",
2010
2027
  name: "xAI: Grok 4.1 Fast (free)",
@@ -2122,7 +2139,7 @@ export const MODELS = {
2122
2139
  cacheWrite: 0,
2123
2140
  },
2124
2141
  contextWindow: 256000,
2125
- maxTokens: 32000,
2142
+ maxTokens: 32768,
2126
2143
  },
2127
2144
  "moonshotai/kimi-k2-thinking": {
2128
2145
  id: "moonshotai/kimi-k2-thinking",
@@ -3442,13 +3459,13 @@ export const MODELS = {
3442
3459
  reasoning: true,
3443
3460
  input: ["text"],
3444
3461
  cost: {
3445
- input: 1,
3446
- output: 4,
3462
+ input: 0.38,
3463
+ output: 1.53,
3447
3464
  cacheRead: 0,
3448
3465
  cacheWrite: 0,
3449
3466
  },
3450
3467
  contextWindow: 262144,
3451
- maxTokens: 32768,
3468
+ maxTokens: 262144,
3452
3469
  },
3453
3470
  "google/gemini-2.5-flash-lite": {
3454
3471
  id: "google/gemini-2.5-flash-lite",
@@ -3586,23 +3603,6 @@ export const MODELS = {
3586
3603
  contextWindow: 128000,
3587
3604
  maxTokens: 16384,
3588
3605
  },
3589
- "mistralai/mistral-small-3.2-24b-instruct:free": {
3590
- id: "mistralai/mistral-small-3.2-24b-instruct:free",
3591
- name: "Mistral: Mistral Small 3.2 24B (free)",
3592
- api: "openai-completions",
3593
- provider: "openrouter",
3594
- baseUrl: "https://openrouter.ai/api/v1",
3595
- reasoning: false,
3596
- input: ["text", "image"],
3597
- cost: {
3598
- input: 0,
3599
- output: 0,
3600
- cacheRead: 0,
3601
- cacheWrite: 0,
3602
- },
3603
- contextWindow: 131072,
3604
- maxTokens: 4096,
3605
- },
3606
3606
  "mistralai/mistral-small-3.2-24b-instruct": {
3607
3607
  id: "mistralai/mistral-small-3.2-24b-instruct",
3608
3608
  name: "Mistral: Mistral Small 3.2 24B",
@@ -3884,8 +3884,8 @@ export const MODELS = {
3884
3884
  reasoning: true,
3885
3885
  input: ["text"],
3886
3886
  cost: {
3887
- input: 0.15,
3888
- output: 0.59,
3887
+ input: 0.049999999999999996,
3888
+ output: 0.19999999999999998,
3889
3889
  cacheRead: 0,
3890
3890
  cacheWrite: 0,
3891
3891
  },
@@ -4059,7 +4059,7 @@ export const MODELS = {
4059
4059
  cacheRead: 0,
4060
4060
  cacheWrite: 0,
4061
4061
  },
4062
- contextWindow: 40960,
4062
+ contextWindow: 131072,
4063
4063
  maxTokens: 4096,
4064
4064
  },
4065
4065
  "qwen/qwen3-235b-a22b": {
@@ -4249,23 +4249,6 @@ export const MODELS = {
4249
4249
  contextWindow: 327680,
4250
4250
  maxTokens: 16384,
4251
4251
  },
4252
- "deepseek/deepseek-chat-v3-0324:free": {
4253
- id: "deepseek/deepseek-chat-v3-0324:free",
4254
- name: "DeepSeek: DeepSeek V3 0324 (free)",
4255
- api: "openai-completions",
4256
- provider: "openrouter",
4257
- baseUrl: "https://openrouter.ai/api/v1",
4258
- reasoning: false,
4259
- input: ["text"],
4260
- cost: {
4261
- input: 0,
4262
- output: 0,
4263
- cacheRead: 0,
4264
- cacheWrite: 0,
4265
- },
4266
- contextWindow: 163840,
4267
- maxTokens: 4096,
4268
- },
4269
4252
  "deepseek/deepseek-chat-v3-0324": {
4270
4253
  id: "deepseek/deepseek-chat-v3-0324",
4271
4254
  name: "DeepSeek: DeepSeek V3 0324",
@@ -4275,13 +4258,13 @@ export const MODELS = {
4275
4258
  reasoning: true,
4276
4259
  input: ["text"],
4277
4260
  cost: {
4278
- input: 0.216,
4279
- output: 0.896,
4280
- cacheRead: 0.135,
4261
+ input: 0.19999999999999998,
4262
+ output: 0.88,
4263
+ cacheRead: 0,
4281
4264
  cacheWrite: 0,
4282
4265
  },
4283
4266
  contextWindow: 163840,
4284
- maxTokens: 163840,
4267
+ maxTokens: 4096,
4285
4268
  },
4286
4269
  "mistralai/mistral-small-3.1-24b-instruct:free": {
4287
4270
  id: "mistralai/mistral-small-3.1-24b-instruct:free",
@@ -4297,8 +4280,8 @@ export const MODELS = {
4297
4280
  cacheRead: 0,
4298
4281
  cacheWrite: 0,
4299
4282
  },
4300
- contextWindow: 96000,
4301
- maxTokens: 96000,
4283
+ contextWindow: 128000,
4284
+ maxTokens: 4096,
4302
4285
  },
4303
4286
  "mistralai/mistral-small-3.1-24b-instruct": {
4304
4287
  id: "mistralai/mistral-small-3.1-24b-instruct",
@@ -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",
@@ -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": {