@dreb/ai 2.24.0 → 2.25.1

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.
@@ -1073,6 +1073,57 @@ export const MODELS = {
1073
1073
  contextWindow: 262144,
1074
1074
  maxTokens: 131072,
1075
1075
  },
1076
+ "openai.gpt-5.4": {
1077
+ id: "openai.gpt-5.4",
1078
+ name: "GPT-5.4",
1079
+ api: "bedrock-converse-stream",
1080
+ provider: "amazon-bedrock",
1081
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1082
+ reasoning: true,
1083
+ input: ["text", "image"],
1084
+ cost: {
1085
+ input: 2.75,
1086
+ output: 16.5,
1087
+ cacheRead: 0.275,
1088
+ cacheWrite: 0,
1089
+ },
1090
+ contextWindow: 272000,
1091
+ maxTokens: 128000,
1092
+ },
1093
+ "openai.gpt-5.5": {
1094
+ id: "openai.gpt-5.5",
1095
+ name: "GPT-5.5",
1096
+ api: "bedrock-converse-stream",
1097
+ provider: "amazon-bedrock",
1098
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1099
+ reasoning: true,
1100
+ input: ["text", "image"],
1101
+ cost: {
1102
+ input: 5.5,
1103
+ output: 33,
1104
+ cacheRead: 0.55,
1105
+ cacheWrite: 0,
1106
+ },
1107
+ contextWindow: 272000,
1108
+ maxTokens: 128000,
1109
+ },
1110
+ "openai.gpt-oss-120b": {
1111
+ id: "openai.gpt-oss-120b",
1112
+ name: "gpt-oss-120b",
1113
+ api: "bedrock-converse-stream",
1114
+ provider: "amazon-bedrock",
1115
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1116
+ reasoning: false,
1117
+ input: ["text"],
1118
+ cost: {
1119
+ input: 0.15,
1120
+ output: 0.6,
1121
+ cacheRead: 0,
1122
+ cacheWrite: 0,
1123
+ },
1124
+ contextWindow: 128000,
1125
+ maxTokens: 16384,
1126
+ },
1076
1127
  "openai.gpt-oss-120b-1:0": {
1077
1128
  id: "openai.gpt-oss-120b-1:0",
1078
1129
  name: "gpt-oss-120b",
@@ -1090,6 +1141,23 @@ export const MODELS = {
1090
1141
  contextWindow: 128000,
1091
1142
  maxTokens: 16384,
1092
1143
  },
1144
+ "openai.gpt-oss-20b": {
1145
+ id: "openai.gpt-oss-20b",
1146
+ name: "gpt-oss-20b",
1147
+ api: "bedrock-converse-stream",
1148
+ provider: "amazon-bedrock",
1149
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1150
+ reasoning: false,
1151
+ input: ["text"],
1152
+ cost: {
1153
+ input: 0.07,
1154
+ output: 0.3,
1155
+ cacheRead: 0,
1156
+ cacheWrite: 0,
1157
+ },
1158
+ contextWindow: 128000,
1159
+ maxTokens: 16384,
1160
+ },
1093
1161
  "openai.gpt-oss-20b-1:0": {
1094
1162
  id: "openai.gpt-oss-20b-1:0",
1095
1163
  name: "gpt-oss-20b",
@@ -5002,11 +5070,11 @@ export const MODELS = {
5002
5070
  api: "mistral-conversations",
5003
5071
  provider: "mistral",
5004
5072
  baseUrl: "https://api.mistral.ai",
5005
- reasoning: true,
5073
+ reasoning: false,
5006
5074
  input: ["text", "image"],
5007
5075
  cost: {
5008
- input: 1.5,
5009
- output: 7.5,
5076
+ input: 0.4,
5077
+ output: 2,
5010
5078
  cacheRead: 0,
5011
5079
  cacheWrite: 0,
5012
5080
  },
@@ -6719,9 +6787,9 @@ export const MODELS = {
6719
6787
  contextWindow: 200000,
6720
6788
  maxTokens: 32000,
6721
6789
  },
6722
- "nemotron-3-super-free": {
6723
- id: "nemotron-3-super-free",
6724
- name: "Nemotron 3 Super Free",
6790
+ "nemotron-3-ultra-free": {
6791
+ id: "nemotron-3-ultra-free",
6792
+ name: "Nemotron 3 Ultra Free",
6725
6793
  api: "openai-completions",
6726
6794
  provider: "opencode",
6727
6795
  baseUrl: "https://opencode.ai/zen/v1",
@@ -6733,7 +6801,7 @@ export const MODELS = {
6733
6801
  cacheRead: 0,
6734
6802
  cacheWrite: 0,
6735
6803
  },
6736
- contextWindow: 204800,
6804
+ contextWindow: 1000000,
6737
6805
  maxTokens: 128000,
6738
6806
  },
6739
6807
  "qwen3.5-plus": {
@@ -6973,7 +7041,7 @@ export const MODELS = {
6973
7041
  cacheRead: 0.05,
6974
7042
  cacheWrite: 0.625,
6975
7043
  },
6976
- contextWindow: 262144,
7044
+ contextWindow: 1000000,
6977
7045
  maxTokens: 65536,
6978
7046
  },
6979
7047
  "qwen3.7-max": {
@@ -7007,7 +7075,7 @@ export const MODELS = {
7007
7075
  cacheRead: 0.04,
7008
7076
  cacheWrite: 0.5,
7009
7077
  },
7010
- contextWindow: 262144,
7078
+ contextWindow: 1000000,
7011
7079
  maxTokens: 65536,
7012
7080
  },
7013
7081
  },
@@ -8178,7 +8246,7 @@ export const MODELS = {
8178
8246
  input: ["text"],
8179
8247
  cost: {
8180
8248
  input: 0.02,
8181
- output: 0.049999999999999996,
8249
+ output: 0.03,
8182
8250
  cacheRead: 0,
8183
8251
  cacheWrite: 0,
8184
8252
  },
@@ -8865,6 +8933,40 @@ export const MODELS = {
8865
8933
  contextWindow: 1000000,
8866
8934
  maxTokens: 262144,
8867
8935
  },
8936
+ "nvidia/nemotron-3-ultra-550b-a55b": {
8937
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
8938
+ name: "NVIDIA: Nemotron 3 Ultra",
8939
+ api: "openai-completions",
8940
+ provider: "openrouter",
8941
+ baseUrl: "https://openrouter.ai/api/v1",
8942
+ reasoning: true,
8943
+ input: ["text"],
8944
+ cost: {
8945
+ input: 0.5,
8946
+ output: 2.5,
8947
+ cacheRead: 0.15,
8948
+ cacheWrite: 0,
8949
+ },
8950
+ contextWindow: 1000000,
8951
+ maxTokens: 16384,
8952
+ },
8953
+ "nvidia/nemotron-3-ultra-550b-a55b:free": {
8954
+ id: "nvidia/nemotron-3-ultra-550b-a55b:free",
8955
+ name: "NVIDIA: Nemotron 3 Ultra (free)",
8956
+ api: "openai-completions",
8957
+ provider: "openrouter",
8958
+ baseUrl: "https://openrouter.ai/api/v1",
8959
+ reasoning: true,
8960
+ input: ["text"],
8961
+ cost: {
8962
+ input: 0,
8963
+ output: 0,
8964
+ cacheRead: 0,
8965
+ cacheWrite: 0,
8966
+ },
8967
+ contextWindow: 1000000,
8968
+ maxTokens: 65536,
8969
+ },
8868
8970
  "nvidia/nemotron-nano-12b-v2-vl:free": {
8869
8971
  id: "nvidia/nemotron-nano-12b-v2-vl:free",
8870
8972
  name: "NVIDIA: Nemotron Nano 12B 2 VL (free)",
@@ -8984,23 +9086,6 @@ export const MODELS = {
8984
9086
  contextWindow: 8191,
8985
9087
  maxTokens: 4096,
8986
9088
  },
8987
- "openai/gpt-4-0314": {
8988
- id: "openai/gpt-4-0314",
8989
- name: "OpenAI: GPT-4 (older v0314)",
8990
- api: "openai-completions",
8991
- provider: "openrouter",
8992
- baseUrl: "https://openrouter.ai/api/v1",
8993
- reasoning: false,
8994
- input: ["text"],
8995
- cost: {
8996
- input: 30,
8997
- output: 60,
8998
- cacheRead: 0,
8999
- cacheWrite: 0,
9000
- },
9001
- contextWindow: 8191,
9002
- maxTokens: 4096,
9003
- },
9004
9089
  "openai/gpt-4-1106-preview": {
9005
9090
  id: "openai/gpt-4-1106-preview",
9006
9091
  name: "OpenAI: GPT-4 Turbo (older v1106)",
@@ -10121,7 +10206,7 @@ export const MODELS = {
10121
10206
  cacheWrite: 0,
10122
10207
  },
10123
10208
  contextWindow: 131072,
10124
- maxTokens: 20000,
10209
+ maxTokens: 16384,
10125
10210
  },
10126
10211
  "qwen/qwen3-30b-a3b-instruct-2507": {
10127
10212
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -10132,8 +10217,8 @@ export const MODELS = {
10132
10217
  reasoning: false,
10133
10218
  input: ["text"],
10134
10219
  cost: {
10135
- input: 0.0428,
10136
- output: 0.1716,
10220
+ input: 0.04815,
10221
+ output: 0.19305,
10137
10222
  cacheRead: 0,
10138
10223
  cacheWrite: 0,
10139
10224
  },
@@ -10786,23 +10871,6 @@ export const MODELS = {
10786
10871
  contextWindow: 256000,
10787
10872
  maxTokens: 128000,
10788
10873
  },
10789
- "sao10k/l3-euryale-70b": {
10790
- id: "sao10k/l3-euryale-70b",
10791
- name: "Sao10k: Llama 3 Euryale 70B v2.1",
10792
- api: "openai-completions",
10793
- provider: "openrouter",
10794
- baseUrl: "https://openrouter.ai/api/v1",
10795
- reasoning: false,
10796
- input: ["text"],
10797
- cost: {
10798
- input: 1.48,
10799
- output: 1.48,
10800
- cacheRead: 0,
10801
- cacheWrite: 0,
10802
- },
10803
- contextWindow: 8192,
10804
- maxTokens: 8192,
10805
- },
10806
10874
  "sao10k/l3.1-euryale-70b": {
10807
10875
  id: "sao10k/l3.1-euryale-70b",
10808
10876
  name: "Sao10K: Llama 3.1 Euryale 70B v2.2",
@@ -11458,7 +11526,7 @@ export const MODELS = {
11458
11526
  provider: "vercel-ai-gateway",
11459
11527
  baseUrl: "https://ai-gateway.vercel.sh",
11460
11528
  reasoning: true,
11461
- input: ["text", "image"],
11529
+ input: ["text"],
11462
11530
  cost: {
11463
11531
  input: 1.3,
11464
11532
  output: 7.8,
@@ -11730,7 +11798,7 @@ export const MODELS = {
11730
11798
  provider: "vercel-ai-gateway",
11731
11799
  baseUrl: "https://ai-gateway.vercel.sh",
11732
11800
  reasoning: true,
11733
- input: ["text", "image"],
11801
+ input: ["text"],
11734
11802
  cost: {
11735
11803
  input: 1.25,
11736
11804
  output: 3.75,
@@ -12155,7 +12223,7 @@ export const MODELS = {
12155
12223
  provider: "vercel-ai-gateway",
12156
12224
  baseUrl: "https://ai-gateway.vercel.sh",
12157
12225
  reasoning: true,
12158
- input: ["text", "image"],
12226
+ input: ["text"],
12159
12227
  cost: {
12160
12228
  input: 0.435,
12161
12229
  output: 0.87,
@@ -12648,7 +12716,7 @@ export const MODELS = {
12648
12716
  provider: "vercel-ai-gateway",
12649
12717
  baseUrl: "https://ai-gateway.vercel.sh",
12650
12718
  reasoning: true,
12651
- input: ["text", "image"],
12719
+ input: ["text"],
12652
12720
  cost: {
12653
12721
  input: 0.6,
12654
12722
  output: 2.4,
@@ -12665,7 +12733,7 @@ export const MODELS = {
12665
12733
  provider: "vercel-ai-gateway",
12666
12734
  baseUrl: "https://ai-gateway.vercel.sh",
12667
12735
  reasoning: true,
12668
- input: ["text", "image"],
12736
+ input: ["text"],
12669
12737
  cost: {
12670
12738
  input: 0.3,
12671
12739
  output: 1.2,
@@ -12682,7 +12750,7 @@ export const MODELS = {
12682
12750
  provider: "vercel-ai-gateway",
12683
12751
  baseUrl: "https://ai-gateway.vercel.sh",
12684
12752
  reasoning: true,
12685
- input: ["text", "image"],
12753
+ input: ["text"],
12686
12754
  cost: {
12687
12755
  input: 0.6,
12688
12756
  output: 2.4,
@@ -13032,6 +13100,23 @@ export const MODELS = {
13032
13100
  contextWindow: 256000,
13033
13101
  maxTokens: 32000,
13034
13102
  },
13103
+ "nvidia/nemotron-3-ultra-550b-a55b": {
13104
+ id: "nvidia/nemotron-3-ultra-550b-a55b",
13105
+ name: "Nemotron 3 Ultra",
13106
+ api: "anthropic-messages",
13107
+ provider: "vercel-ai-gateway",
13108
+ baseUrl: "https://ai-gateway.vercel.sh",
13109
+ reasoning: true,
13110
+ input: ["text"],
13111
+ cost: {
13112
+ input: 0.6,
13113
+ output: 2.4,
13114
+ cacheRead: 0.12,
13115
+ cacheWrite: 0,
13116
+ },
13117
+ contextWindow: 1000000,
13118
+ maxTokens: 65000,
13119
+ },
13035
13120
  "nvidia/nemotron-nano-12b-v2-vl": {
13036
13121
  id: "nvidia/nemotron-nano-12b-v2-vl",
13037
13122
  name: "Nvidia Nemotron Nano 12B V2 VL",