@mariozechner/pi-ai 0.9.1 → 0.9.2

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.
@@ -1455,23 +1455,6 @@ export const MODELS = {
1455
1455
  contextWindow: 131072,
1456
1456
  maxTokens: 40960,
1457
1457
  },
1458
- "qwen-3-coder-480b": {
1459
- id: "qwen-3-coder-480b",
1460
- name: "Qwen 3 Coder 480B",
1461
- api: "openai-completions",
1462
- provider: "cerebras",
1463
- baseUrl: "https://api.cerebras.ai/v1",
1464
- reasoning: false,
1465
- input: ["text"],
1466
- cost: {
1467
- input: 2,
1468
- output: 2,
1469
- cacheRead: 0,
1470
- cacheWrite: 0,
1471
- },
1472
- contextWindow: 131000,
1473
- maxTokens: 32000,
1474
- },
1475
1458
  "gpt-oss-120b": {
1476
1459
  id: "gpt-oss-120b",
1477
1460
  name: "GPT OSS 120B",
@@ -1954,6 +1937,23 @@ export const MODELS = {
1954
1937
  },
1955
1938
  },
1956
1939
  openrouter: {
1940
+ "allenai/olmo-3-7b-instruct": {
1941
+ id: "allenai/olmo-3-7b-instruct",
1942
+ name: "AllenAI: Olmo 3 7B Instruct",
1943
+ api: "openai-completions",
1944
+ provider: "openrouter",
1945
+ baseUrl: "https://openrouter.ai/api/v1",
1946
+ reasoning: false,
1947
+ input: ["text"],
1948
+ cost: {
1949
+ input: 0.09999999999999999,
1950
+ output: 0.19999999999999998,
1951
+ cacheRead: 0,
1952
+ cacheWrite: 0,
1953
+ },
1954
+ contextWindow: 65536,
1955
+ maxTokens: 65536,
1956
+ },
1957
1957
  "x-ai/grok-4.1-fast": {
1958
1958
  id: "x-ai/grok-4.1-fast",
1959
1959
  name: "xAI: Grok 4.1 Fast",
@@ -2184,8 +2184,8 @@ export const MODELS = {
2184
2184
  reasoning: true,
2185
2185
  input: ["text"],
2186
2186
  cost: {
2187
- input: 0.255,
2188
- output: 1.02,
2187
+ input: 0.24,
2188
+ output: 0.96,
2189
2189
  cacheRead: 0,
2190
2190
  cacheWrite: 0,
2191
2191
  },
@@ -2269,8 +2269,8 @@ export const MODELS = {
2269
2269
  reasoning: false,
2270
2270
  input: ["text", "image"],
2271
2271
  cost: {
2272
- input: 0.08,
2273
- output: 0.5,
2272
+ input: 0.064,
2273
+ output: 0.39999999999999997,
2274
2274
  cacheRead: 0,
2275
2275
  cacheWrite: 0,
2276
2276
  },
@@ -2354,8 +2354,8 @@ export const MODELS = {
2354
2354
  reasoning: true,
2355
2355
  input: ["text", "image"],
2356
2356
  cost: {
2357
- input: 0.19999999999999998,
2358
- output: 1,
2357
+ input: 0.16,
2358
+ output: 0.7999999999999999,
2359
2359
  cacheRead: 0,
2360
2360
  cacheWrite: 0,
2361
2361
  },
@@ -2456,13 +2456,13 @@ export const MODELS = {
2456
2456
  reasoning: true,
2457
2457
  input: ["text"],
2458
2458
  cost: {
2459
- input: 0.27,
2460
- output: 0.39999999999999997,
2459
+ input: 0.216,
2460
+ output: 0.328,
2461
2461
  cacheRead: 0,
2462
2462
  cacheWrite: 0,
2463
2463
  },
2464
2464
  contextWindow: 163840,
2465
- maxTokens: 4096,
2465
+ maxTokens: 65536,
2466
2466
  },
2467
2467
  "google/gemini-2.5-flash-preview-09-2025": {
2468
2468
  id: "google/gemini-2.5-flash-preview-09-2025",
@@ -2583,34 +2583,34 @@ export const MODELS = {
2583
2583
  contextWindow: 400000,
2584
2584
  maxTokens: 128000,
2585
2585
  },
2586
- "deepseek/deepseek-v3.1-terminus": {
2587
- id: "deepseek/deepseek-v3.1-terminus",
2588
- name: "DeepSeek: DeepSeek V3.1 Terminus",
2586
+ "deepseek/deepseek-v3.1-terminus:exacto": {
2587
+ id: "deepseek/deepseek-v3.1-terminus:exacto",
2588
+ name: "DeepSeek: DeepSeek V3.1 Terminus (exacto)",
2589
2589
  api: "openai-completions",
2590
2590
  provider: "openrouter",
2591
2591
  baseUrl: "https://openrouter.ai/api/v1",
2592
2592
  reasoning: true,
2593
2593
  input: ["text"],
2594
2594
  cost: {
2595
- input: 0.22999999999999998,
2596
- output: 0.8999999999999999,
2595
+ input: 0.216,
2596
+ output: 0.7999999999999999,
2597
2597
  cacheRead: 0,
2598
2598
  cacheWrite: 0,
2599
2599
  },
2600
- contextWindow: 163840,
2601
- maxTokens: 163840,
2600
+ contextWindow: 131072,
2601
+ maxTokens: 65536,
2602
2602
  },
2603
- "deepseek/deepseek-v3.1-terminus:exacto": {
2604
- id: "deepseek/deepseek-v3.1-terminus:exacto",
2605
- name: "DeepSeek: DeepSeek V3.1 Terminus (exacto)",
2603
+ "deepseek/deepseek-v3.1-terminus": {
2604
+ id: "deepseek/deepseek-v3.1-terminus",
2605
+ name: "DeepSeek: DeepSeek V3.1 Terminus",
2606
2606
  api: "openai-completions",
2607
2607
  provider: "openrouter",
2608
2608
  baseUrl: "https://openrouter.ai/api/v1",
2609
2609
  reasoning: true,
2610
2610
  input: ["text"],
2611
2611
  cost: {
2612
- input: 0.27,
2613
- output: 1,
2612
+ input: 0.216,
2613
+ output: 0.7999999999999999,
2614
2614
  cacheRead: 0,
2615
2615
  cacheWrite: 0,
2616
2616
  },
@@ -2694,13 +2694,13 @@ export const MODELS = {
2694
2694
  reasoning: true,
2695
2695
  input: ["text"],
2696
2696
  cost: {
2697
- input: 0.15,
2697
+ input: 0.12,
2698
2698
  output: 1.2,
2699
2699
  cacheRead: 0,
2700
2700
  cacheWrite: 0,
2701
2701
  },
2702
- contextWindow: 262144,
2703
- maxTokens: 262144,
2702
+ contextWindow: 131072,
2703
+ maxTokens: 32768,
2704
2704
  },
2705
2705
  "qwen/qwen3-next-80b-a3b-instruct": {
2706
2706
  id: "qwen/qwen3-next-80b-a3b-instruct",
@@ -3017,8 +3017,8 @@ export const MODELS = {
3017
3017
  reasoning: false,
3018
3018
  input: ["text"],
3019
3019
  cost: {
3020
- input: 0.07,
3021
- output: 0.28,
3020
+ input: 0.056,
3021
+ output: 0.224,
3022
3022
  cacheRead: 0,
3023
3023
  cacheWrite: 0,
3024
3024
  },
@@ -3034,8 +3034,8 @@ export const MODELS = {
3034
3034
  reasoning: true,
3035
3035
  input: ["text", "image"],
3036
3036
  cost: {
3037
- input: 0.14,
3038
- output: 0.56,
3037
+ input: 0.112,
3038
+ output: 0.448,
3039
3039
  cacheRead: 0,
3040
3040
  cacheWrite: 0,
3041
3041
  },
@@ -3051,8 +3051,8 @@ export const MODELS = {
3051
3051
  reasoning: true,
3052
3052
  input: ["text", "image"],
3053
3053
  cost: {
3054
- input: 0.6,
3055
- output: 1.7999999999999998,
3054
+ input: 0.48,
3055
+ output: 1.44,
3056
3056
  cacheRead: 0.11,
3057
3057
  cacheWrite: 0,
3058
3058
  },
@@ -3144,9 +3144,9 @@ export const MODELS = {
3144
3144
  contextWindow: 400000,
3145
3145
  maxTokens: 128000,
3146
3146
  },
3147
- "openai/gpt-oss-120b": {
3148
- id: "openai/gpt-oss-120b",
3149
- name: "OpenAI: gpt-oss-120b",
3147
+ "openai/gpt-oss-120b:exacto": {
3148
+ id: "openai/gpt-oss-120b:exacto",
3149
+ name: "OpenAI: gpt-oss-120b (exacto)",
3150
3150
  api: "openai-completions",
3151
3151
  provider: "openrouter",
3152
3152
  baseUrl: "https://openrouter.ai/api/v1",
@@ -3154,29 +3154,29 @@ export const MODELS = {
3154
3154
  input: ["text"],
3155
3155
  cost: {
3156
3156
  input: 0.04,
3157
- output: 0.39999999999999997,
3157
+ output: 0.19999999999999998,
3158
3158
  cacheRead: 0,
3159
3159
  cacheWrite: 0,
3160
3160
  },
3161
3161
  contextWindow: 131072,
3162
- maxTokens: 131072,
3162
+ maxTokens: 32768,
3163
3163
  },
3164
- "openai/gpt-oss-120b:exacto": {
3165
- id: "openai/gpt-oss-120b:exacto",
3166
- name: "OpenAI: gpt-oss-120b (exacto)",
3164
+ "openai/gpt-oss-120b": {
3165
+ id: "openai/gpt-oss-120b",
3166
+ name: "OpenAI: gpt-oss-120b",
3167
3167
  api: "openai-completions",
3168
3168
  provider: "openrouter",
3169
3169
  baseUrl: "https://openrouter.ai/api/v1",
3170
3170
  reasoning: true,
3171
3171
  input: ["text"],
3172
3172
  cost: {
3173
- input: 0.049999999999999996,
3174
- output: 0.24,
3173
+ input: 0.04,
3174
+ output: 0.19999999999999998,
3175
3175
  cacheRead: 0,
3176
3176
  cacheWrite: 0,
3177
3177
  },
3178
3178
  contextWindow: 131072,
3179
- maxTokens: 4096,
3179
+ maxTokens: 32768,
3180
3180
  },
3181
3181
  "openai/gpt-oss-20b:free": {
3182
3182
  id: "openai/gpt-oss-20b:free",
@@ -3323,8 +3323,8 @@ export const MODELS = {
3323
3323
  reasoning: true,
3324
3324
  input: ["text"],
3325
3325
  cost: {
3326
- input: 0.13,
3327
- output: 0.85,
3326
+ input: 0.10400000000000001,
3327
+ output: 0.6799999999999999,
3328
3328
  cacheRead: 0,
3329
3329
  cacheWrite: 0,
3330
3330
  },
@@ -3408,13 +3408,13 @@ export const MODELS = {
3408
3408
  reasoning: true,
3409
3409
  input: ["text"],
3410
3410
  cost: {
3411
- input: 0.38,
3412
- output: 1.53,
3411
+ input: 1,
3412
+ output: 4,
3413
3413
  cacheRead: 0,
3414
3414
  cacheWrite: 0,
3415
3415
  },
3416
3416
  contextWindow: 262144,
3417
- maxTokens: 262144,
3417
+ maxTokens: 32768,
3418
3418
  },
3419
3419
  "google/gemini-2.5-flash-lite": {
3420
3420
  id: "google/gemini-2.5-flash-lite",
@@ -3442,13 +3442,13 @@ export const MODELS = {
3442
3442
  reasoning: true,
3443
3443
  input: ["text"],
3444
3444
  cost: {
3445
- input: 0.08,
3446
- output: 0.55,
3445
+ input: 0.072,
3446
+ output: 0.464,
3447
3447
  cacheRead: 0,
3448
3448
  cacheWrite: 0,
3449
3449
  },
3450
- contextWindow: 262144,
3451
- maxTokens: 262144,
3450
+ contextWindow: 131072,
3451
+ maxTokens: 16384,
3452
3452
  },
3453
3453
  "moonshotai/kimi-k2": {
3454
3454
  id: "moonshotai/kimi-k2",
@@ -3459,13 +3459,13 @@ export const MODELS = {
3459
3459
  reasoning: false,
3460
3460
  input: ["text"],
3461
3461
  cost: {
3462
- input: 0.5,
3463
- output: 2.4,
3462
+ input: 0.456,
3463
+ output: 1.8399999999999999,
3464
3464
  cacheRead: 0,
3465
3465
  cacheWrite: 0,
3466
3466
  },
3467
3467
  contextWindow: 131072,
3468
- maxTokens: 4096,
3468
+ maxTokens: 131072,
3469
3469
  },
3470
3470
  "mistralai/devstral-medium": {
3471
3471
  id: "mistralai/devstral-medium",
@@ -3969,8 +3969,8 @@ export const MODELS = {
3969
3969
  reasoning: true,
3970
3970
  input: ["text"],
3971
3971
  cost: {
3972
- input: 0.035,
3973
- output: 0.13799999999999998,
3972
+ input: 0.028,
3973
+ output: 0.1104,
3974
3974
  cacheRead: 0,
3975
3975
  cacheWrite: 0,
3976
3976
  },
@@ -4003,8 +4003,8 @@ export const MODELS = {
4003
4003
  reasoning: true,
4004
4004
  input: ["text"],
4005
4005
  cost: {
4006
- input: 0.049999999999999996,
4007
- output: 0.19999999999999998,
4006
+ input: 0.08,
4007
+ output: 0.24,
4008
4008
  cacheRead: 0,
4009
4009
  cacheWrite: 0,
4010
4010
  },
@@ -4190,13 +4190,13 @@ export const MODELS = {
4190
4190
  reasoning: false,
4191
4191
  input: ["text", "image"],
4192
4192
  cost: {
4193
- input: 0.15,
4194
- output: 0.6,
4193
+ input: 0.136,
4194
+ output: 0.6799999999999999,
4195
4195
  cacheRead: 0,
4196
4196
  cacheWrite: 0,
4197
4197
  },
4198
4198
  contextWindow: 1048576,
4199
- maxTokens: 16384,
4199
+ maxTokens: 8192,
4200
4200
  },
4201
4201
  "meta-llama/llama-4-scout": {
4202
4202
  id: "meta-llama/llama-4-scout",
@@ -4241,9 +4241,9 @@ export const MODELS = {
4241
4241
  reasoning: true,
4242
4242
  input: ["text"],
4243
4243
  cost: {
4244
- input: 0.24,
4245
- output: 0.84,
4246
- cacheRead: 0,
4244
+ input: 0.216,
4245
+ output: 0.896,
4246
+ cacheRead: 0.135,
4247
4247
  cacheWrite: 0,
4248
4248
  },
4249
4249
  contextWindow: 163840,
@@ -4649,13 +4649,13 @@ export const MODELS = {
4649
4649
  reasoning: false,
4650
4650
  input: ["text"],
4651
4651
  cost: {
4652
- input: 0.13,
4653
- output: 0.38,
4652
+ input: 0.10400000000000001,
4653
+ output: 0.312,
4654
4654
  cacheRead: 0,
4655
4655
  cacheWrite: 0,
4656
4656
  },
4657
4657
  contextWindow: 131072,
4658
- maxTokens: 16384,
4658
+ maxTokens: 120000,
4659
4659
  },
4660
4660
  "amazon/nova-lite-v1": {
4661
4661
  id: "amazon/nova-lite-v1",
@@ -5082,39 +5082,39 @@ export const MODELS = {
5082
5082
  contextWindow: 128000,
5083
5083
  maxTokens: 16384,
5084
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",
5085
+ "meta-llama/llama-3.1-70b-instruct": {
5086
+ id: "meta-llama/llama-3.1-70b-instruct",
5087
+ name: "Meta: Llama 3.1 70B Instruct",
5088
5088
  api: "openai-completions",
5089
5089
  provider: "openrouter",
5090
5090
  baseUrl: "https://openrouter.ai/api/v1",
5091
5091
  reasoning: false,
5092
5092
  input: ["text"],
5093
5093
  cost: {
5094
- input: 0.02,
5095
- output: 0.03,
5094
+ input: 0.39999999999999997,
5095
+ output: 0.39999999999999997,
5096
5096
  cacheRead: 0,
5097
5097
  cacheWrite: 0,
5098
5098
  },
5099
5099
  contextWindow: 131072,
5100
- maxTokens: 16384,
5100
+ maxTokens: 4096,
5101
5101
  },
5102
- "meta-llama/llama-3.1-70b-instruct": {
5103
- id: "meta-llama/llama-3.1-70b-instruct",
5104
- name: "Meta: Llama 3.1 70B Instruct",
5102
+ "meta-llama/llama-3.1-8b-instruct": {
5103
+ id: "meta-llama/llama-3.1-8b-instruct",
5104
+ name: "Meta: Llama 3.1 8B Instruct",
5105
5105
  api: "openai-completions",
5106
5106
  provider: "openrouter",
5107
5107
  baseUrl: "https://openrouter.ai/api/v1",
5108
5108
  reasoning: false,
5109
5109
  input: ["text"],
5110
5110
  cost: {
5111
- input: 0.39999999999999997,
5112
- output: 0.39999999999999997,
5111
+ input: 0.02,
5112
+ output: 0.03,
5113
5113
  cacheRead: 0,
5114
5114
  cacheWrite: 0,
5115
5115
  },
5116
5116
  contextWindow: 131072,
5117
- maxTokens: 4096,
5117
+ maxTokens: 16384,
5118
5118
  },
5119
5119
  "meta-llama/llama-3.1-405b-instruct": {
5120
5120
  id: "meta-llama/llama-3.1-405b-instruct",
@@ -5150,9 +5150,9 @@ export const MODELS = {
5150
5150
  contextWindow: 131072,
5151
5151
  maxTokens: 16384,
5152
5152
  },
5153
- "openai/gpt-4o-mini": {
5154
- id: "openai/gpt-4o-mini",
5155
- name: "OpenAI: GPT-4o-mini",
5153
+ "openai/gpt-4o-mini-2024-07-18": {
5154
+ id: "openai/gpt-4o-mini-2024-07-18",
5155
+ name: "OpenAI: GPT-4o-mini (2024-07-18)",
5156
5156
  api: "openai-completions",
5157
5157
  provider: "openrouter",
5158
5158
  baseUrl: "https://openrouter.ai/api/v1",
@@ -5167,9 +5167,9 @@ export const MODELS = {
5167
5167
  contextWindow: 128000,
5168
5168
  maxTokens: 16384,
5169
5169
  },
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)",
5170
+ "openai/gpt-4o-mini": {
5171
+ id: "openai/gpt-4o-mini",
5172
+ name: "OpenAI: GPT-4o-mini",
5173
5173
  api: "openai-completions",
5174
5174
  provider: "openrouter",
5175
5175
  baseUrl: "https://openrouter.ai/api/v1",
@@ -5269,6 +5269,23 @@ export const MODELS = {
5269
5269
  contextWindow: 128000,
5270
5270
  maxTokens: 4096,
5271
5271
  },
5272
+ "openai/gpt-4o-2024-05-13": {
5273
+ id: "openai/gpt-4o-2024-05-13",
5274
+ name: "OpenAI: GPT-4o (2024-05-13)",
5275
+ api: "openai-completions",
5276
+ provider: "openrouter",
5277
+ baseUrl: "https://openrouter.ai/api/v1",
5278
+ reasoning: false,
5279
+ input: ["text", "image"],
5280
+ cost: {
5281
+ input: 5,
5282
+ output: 15,
5283
+ cacheRead: 0,
5284
+ cacheWrite: 0,
5285
+ },
5286
+ contextWindow: 128000,
5287
+ maxTokens: 4096,
5288
+ },
5272
5289
  "openai/gpt-4o": {
5273
5290
  id: "openai/gpt-4o",
5274
5291
  name: "OpenAI: GPT-4o",
@@ -5303,23 +5320,6 @@ export const MODELS = {
5303
5320
  contextWindow: 128000,
5304
5321
  maxTokens: 64000,
5305
5322
  },
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
5323
  "meta-llama/llama-3-70b-instruct": {
5324
5324
  id: "meta-llama/llama-3-70b-instruct",
5325
5325
  name: "Meta: Llama 3 70B Instruct",
@@ -5575,38 +5575,38 @@ export const MODELS = {
5575
5575
  contextWindow: 8191,
5576
5576
  maxTokens: 4096,
5577
5577
  },
5578
- "openai/gpt-3.5-turbo": {
5579
- id: "openai/gpt-3.5-turbo",
5580
- name: "OpenAI: GPT-3.5 Turbo",
5578
+ "openai/gpt-4": {
5579
+ id: "openai/gpt-4",
5580
+ name: "OpenAI: GPT-4",
5581
5581
  api: "openai-completions",
5582
5582
  provider: "openrouter",
5583
5583
  baseUrl: "https://openrouter.ai/api/v1",
5584
5584
  reasoning: false,
5585
5585
  input: ["text"],
5586
5586
  cost: {
5587
- input: 0.5,
5588
- output: 1.5,
5587
+ input: 30,
5588
+ output: 60,
5589
5589
  cacheRead: 0,
5590
5590
  cacheWrite: 0,
5591
5591
  },
5592
- contextWindow: 16385,
5592
+ contextWindow: 8191,
5593
5593
  maxTokens: 4096,
5594
5594
  },
5595
- "openai/gpt-4": {
5596
- id: "openai/gpt-4",
5597
- name: "OpenAI: GPT-4",
5595
+ "openai/gpt-3.5-turbo": {
5596
+ id: "openai/gpt-3.5-turbo",
5597
+ name: "OpenAI: GPT-3.5 Turbo",
5598
5598
  api: "openai-completions",
5599
5599
  provider: "openrouter",
5600
5600
  baseUrl: "https://openrouter.ai/api/v1",
5601
5601
  reasoning: false,
5602
5602
  input: ["text"],
5603
5603
  cost: {
5604
- input: 30,
5605
- output: 60,
5604
+ input: 0.5,
5605
+ output: 1.5,
5606
5606
  cacheRead: 0,
5607
5607
  cacheWrite: 0,
5608
5608
  },
5609
- contextWindow: 8191,
5609
+ contextWindow: 16385,
5610
5610
  maxTokens: 4096,
5611
5611
  },
5612
5612
  "openrouter/auto": {