@caupulican/pi-ai 0.80.8 → 0.80.9

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.
@@ -1086,6 +1086,59 @@ export const MODELS = {
1086
1086
  contextWindow: 262144,
1087
1087
  maxTokens: 131072,
1088
1088
  },
1089
+ "openai.gpt-5.4": {
1090
+ id: "openai.gpt-5.4",
1091
+ name: "GPT-5.4",
1092
+ api: "bedrock-converse-stream",
1093
+ provider: "amazon-bedrock",
1094
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1095
+ reasoning: true,
1096
+ thinkingLevelMap: { "xhigh": "xhigh" },
1097
+ input: ["text", "image"],
1098
+ cost: {
1099
+ input: 2.75,
1100
+ output: 16.5,
1101
+ cacheRead: 0.275,
1102
+ cacheWrite: 0,
1103
+ },
1104
+ contextWindow: 272000,
1105
+ maxTokens: 128000,
1106
+ },
1107
+ "openai.gpt-5.5": {
1108
+ id: "openai.gpt-5.5",
1109
+ name: "GPT-5.5",
1110
+ api: "bedrock-converse-stream",
1111
+ provider: "amazon-bedrock",
1112
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1113
+ reasoning: true,
1114
+ thinkingLevelMap: { "xhigh": "xhigh" },
1115
+ input: ["text", "image"],
1116
+ cost: {
1117
+ input: 5.5,
1118
+ output: 33,
1119
+ cacheRead: 0.55,
1120
+ cacheWrite: 0,
1121
+ },
1122
+ contextWindow: 272000,
1123
+ maxTokens: 128000,
1124
+ },
1125
+ "openai.gpt-oss-120b": {
1126
+ id: "openai.gpt-oss-120b",
1127
+ name: "gpt-oss-120b",
1128
+ api: "bedrock-converse-stream",
1129
+ provider: "amazon-bedrock",
1130
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1131
+ reasoning: false,
1132
+ input: ["text"],
1133
+ cost: {
1134
+ input: 0.15,
1135
+ output: 0.6,
1136
+ cacheRead: 0,
1137
+ cacheWrite: 0,
1138
+ },
1139
+ contextWindow: 128000,
1140
+ maxTokens: 16384,
1141
+ },
1089
1142
  "openai.gpt-oss-120b-1:0": {
1090
1143
  id: "openai.gpt-oss-120b-1:0",
1091
1144
  name: "gpt-oss-120b",
@@ -1103,6 +1156,23 @@ export const MODELS = {
1103
1156
  contextWindow: 128000,
1104
1157
  maxTokens: 16384,
1105
1158
  },
1159
+ "openai.gpt-oss-20b": {
1160
+ id: "openai.gpt-oss-20b",
1161
+ name: "gpt-oss-20b",
1162
+ api: "bedrock-converse-stream",
1163
+ provider: "amazon-bedrock",
1164
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1165
+ reasoning: false,
1166
+ input: ["text"],
1167
+ cost: {
1168
+ input: 0.07,
1169
+ output: 0.3,
1170
+ cacheRead: 0,
1171
+ cacheWrite: 0,
1172
+ },
1173
+ contextWindow: 128000,
1174
+ maxTokens: 16384,
1175
+ },
1106
1176
  "openai.gpt-oss-20b-1:0": {
1107
1177
  id: "openai.gpt-oss-20b-1:0",
1108
1178
  name: "gpt-oss-20b",
@@ -3830,6 +3900,24 @@ export const MODELS = {
3830
3900
  contextWindow: 202800,
3831
3901
  maxTokens: 131072,
3832
3902
  },
3903
+ "accounts/fireworks/routers/kimi-k2p6-fast": {
3904
+ id: "accounts/fireworks/routers/kimi-k2p6-fast",
3905
+ name: "Kimi K2.6 Fast",
3906
+ api: "anthropic-messages",
3907
+ provider: "fireworks",
3908
+ baseUrl: "https://api.fireworks.ai/inference",
3909
+ compat: { "sendSessionAffinityHeaders": true, "supportsEagerToolInputStreaming": false, "supportsCacheControlOnTools": false, "supportsLongCacheRetention": false },
3910
+ reasoning: true,
3911
+ input: ["text", "image"],
3912
+ cost: {
3913
+ input: 2,
3914
+ output: 8,
3915
+ cacheRead: 0.3,
3916
+ cacheWrite: 0,
3917
+ },
3918
+ contextWindow: 262000,
3919
+ maxTokens: 262000,
3920
+ },
3833
3921
  "accounts/fireworks/routers/kimi-k2p6-turbo": {
3834
3922
  id: "accounts/fireworks/routers/kimi-k2p6-turbo",
3835
3923
  name: "Kimi K2.6 Turbo",
@@ -5948,11 +6036,11 @@ export const MODELS = {
5948
6036
  api: "mistral-conversations",
5949
6037
  provider: "mistral",
5950
6038
  baseUrl: "https://api.mistral.ai",
5951
- reasoning: true,
6039
+ reasoning: false,
5952
6040
  input: ["text", "image"],
5953
6041
  cost: {
5954
- input: 1.5,
5955
- output: 7.5,
6042
+ input: 0.4,
6043
+ output: 2,
5956
6044
  cacheRead: 0,
5957
6045
  cacheWrite: 0,
5958
6046
  },
@@ -7950,9 +8038,9 @@ export const MODELS = {
7950
8038
  contextWindow: 200000,
7951
8039
  maxTokens: 32000,
7952
8040
  },
7953
- "nemotron-3-super-free": {
7954
- id: "nemotron-3-super-free",
7955
- name: "Nemotron 3 Super Free",
8041
+ "nemotron-3-ultra-free": {
8042
+ id: "nemotron-3-ultra-free",
8043
+ name: "Nemotron 3 Ultra Free",
7956
8044
  api: "openai-completions",
7957
8045
  provider: "opencode",
7958
8046
  baseUrl: "https://opencode.ai/zen/v1",
@@ -7964,7 +8052,7 @@ export const MODELS = {
7964
8052
  cacheRead: 0,
7965
8053
  cacheWrite: 0,
7966
8054
  },
7967
- contextWindow: 204800,
8055
+ contextWindow: 1000000,
7968
8056
  maxTokens: 128000,
7969
8057
  },
7970
8058
  "qwen3.5-plus": {
@@ -8211,7 +8299,7 @@ export const MODELS = {
8211
8299
  cacheRead: 0.05,
8212
8300
  cacheWrite: 0.625,
8213
8301
  },
8214
- contextWindow: 262144,
8302
+ contextWindow: 1000000,
8215
8303
  maxTokens: 65536,
8216
8304
  },
8217
8305
  "qwen3.7-max": {
@@ -8245,7 +8333,7 @@ export const MODELS = {
8245
8333
  cacheRead: 0.04,
8246
8334
  cacheWrite: 0.5,
8247
8335
  },
8248
- contextWindow: 262144,
8336
+ contextWindow: 1000000,
8249
8337
  maxTokens: 65536,
8250
8338
  },
8251
8339
  },
@@ -9273,12 +9361,12 @@ export const MODELS = {
9273
9361
  input: ["text", "image"],
9274
9362
  cost: {
9275
9363
  input: 0.12,
9276
- output: 0.37,
9277
- cacheRead: 0,
9364
+ output: 0.36,
9365
+ cacheRead: 0.09,
9278
9366
  cacheWrite: 0,
9279
9367
  },
9280
9368
  contextWindow: 262144,
9281
- maxTokens: 16384,
9369
+ maxTokens: 8192,
9282
9370
  },
9283
9371
  "google/gemma-4-31b-it:free": {
9284
9372
  id: "google/gemma-4-31b-it:free",
@@ -9427,7 +9515,7 @@ export const MODELS = {
9427
9515
  input: ["text"],
9428
9516
  cost: {
9429
9517
  input: 0.02,
9430
- output: 0.049999999999999996,
9518
+ output: 0.03,
9431
9519
  cacheRead: 0,
9432
9520
  cacheWrite: 0,
9433
9521
  },
@@ -9585,7 +9673,7 @@ export const MODELS = {
9585
9673
  cacheWrite: 0,
9586
9674
  },
9587
9675
  contextWindow: 204800,
9588
- maxTokens: 131072,
9676
+ maxTokens: 196608,
9589
9677
  },
9590
9678
  "minimax/minimax-m3": {
9591
9679
  id: "minimax/minimax-m3",
@@ -10116,6 +10204,23 @@ export const MODELS = {
10116
10204
  contextWindow: 1000000,
10117
10205
  maxTokens: 262144,
10118
10206
  },
10207
+ "nvidia/nemotron-3-ultra-550b-a55b": {
10208
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
10209
+ name: "NVIDIA: Nemotron 3 Ultra",
10210
+ api: "openai-completions",
10211
+ provider: "openrouter",
10212
+ baseUrl: "https://openrouter.ai/api/v1",
10213
+ reasoning: true,
10214
+ input: ["text"],
10215
+ cost: {
10216
+ input: 0.5,
10217
+ output: 2.5,
10218
+ cacheRead: 0.15,
10219
+ cacheWrite: 0,
10220
+ },
10221
+ contextWindow: 1000000,
10222
+ maxTokens: 16384,
10223
+ },
10119
10224
  "nvidia/nemotron-3-ultra-550b-a55b:free": {
10120
10225
  id: "nvidia/nemotron-3-ultra-550b-a55b:free",
10121
10226
  name: "NVIDIA: Nemotron 3 Ultra (free)",
@@ -11384,7 +11489,7 @@ export const MODELS = {
11384
11489
  cacheWrite: 0,
11385
11490
  },
11386
11491
  contextWindow: 131072,
11387
- maxTokens: 20000,
11492
+ maxTokens: 16384,
11388
11493
  },
11389
11494
  "qwen/qwen3-30b-a3b-instruct-2507": {
11390
11495
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -12049,23 +12154,6 @@ export const MODELS = {
12049
12154
  contextWindow: 256000,
12050
12155
  maxTokens: 128000,
12051
12156
  },
12052
- "sao10k/l3-euryale-70b": {
12053
- id: "sao10k/l3-euryale-70b",
12054
- name: "Sao10k: Llama 3 Euryale 70B v2.1",
12055
- api: "openai-completions",
12056
- provider: "openrouter",
12057
- baseUrl: "https://openrouter.ai/api/v1",
12058
- reasoning: false,
12059
- input: ["text"],
12060
- cost: {
12061
- input: 1.48,
12062
- output: 1.48,
12063
- cacheRead: 0,
12064
- cacheWrite: 0,
12065
- },
12066
- contextWindow: 8192,
12067
- maxTokens: 8192,
12068
- },
12069
12157
  "sao10k/l3.1-euryale-70b": {
12070
12158
  id: "sao10k/l3.1-euryale-70b",
12071
12159
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -12799,7 +12887,7 @@ export const MODELS = {
12799
12887
  },
12800
12888
  "deepseek-ai/DeepSeek-V3": {
12801
12889
  id: "deepseek-ai/DeepSeek-V3",
12802
- name: "DeepSeek V3",
12890
+ name: "DeepSeek-V3",
12803
12891
  api: "openai-completions",
12804
12892
  provider: "together",
12805
12893
  baseUrl: "https://api.together.ai/v1",
@@ -12947,6 +13035,25 @@ export const MODELS = {
12947
13035
  contextWindow: 262144,
12948
13036
  maxTokens: 131000,
12949
13037
  },
13038
+ "nvidia/nemotron-3-ultra-550b-a55b": {
13039
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
13040
+ name: "Nemotron 3 Ultra 550B A55B",
13041
+ api: "openai-completions",
13042
+ provider: "together",
13043
+ baseUrl: "https://api.together.ai/v1",
13044
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13045
+ reasoning: true,
13046
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13047
+ input: ["text"],
13048
+ cost: {
13049
+ input: 0.6,
13050
+ output: 3.6,
13051
+ cacheRead: 0.2,
13052
+ cacheWrite: 0,
13053
+ },
13054
+ contextWindow: 512300,
13055
+ maxTokens: 512300,
13056
+ },
12950
13057
  "openai/gpt-oss-120b": {
12951
13058
  id: "openai/gpt-oss-120b",
12952
13059
  name: "GPT OSS 120B",
@@ -14643,6 +14750,23 @@ export const MODELS = {
14643
14750
  contextWindow: 256000,
14644
14751
  maxTokens: 32000,
14645
14752
  },
14753
+ "nvidia/nemotron-3-ultra-550b-a55b": {
14754
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
14755
+ name: "Nemotron 3 Ultra",
14756
+ api: "anthropic-messages",
14757
+ provider: "vercel-ai-gateway",
14758
+ baseUrl: "https://ai-gateway.vercel.sh",
14759
+ reasoning: true,
14760
+ input: ["text"],
14761
+ cost: {
14762
+ input: 0.6,
14763
+ output: 2.4,
14764
+ cacheRead: 0.12,
14765
+ cacheWrite: 0,
14766
+ },
14767
+ contextWindow: 1000000,
14768
+ maxTokens: 65000,
14769
+ },
14646
14770
  "nvidia/nemotron-nano-12b-v2-vl": {
14647
14771
  id: "nvidia/nemotron-nano-12b-v2-vl",
14648
14772
  name: "Nvidia Nemotron Nano 12B V2 VL",