@dreb/ai 2.45.2 → 2.45.4

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.
@@ -7814,6 +7814,23 @@ export const MODELS = {
7814
7814
  contextWindow: 262144,
7815
7815
  maxTokens: 262144,
7816
7816
  },
7817
+ "kimi-k3": {
7818
+ id: "kimi-k3",
7819
+ name: "Kimi K3",
7820
+ api: "openai-completions",
7821
+ provider: "opencode",
7822
+ baseUrl: "https://opencode.ai/zen/v1",
7823
+ reasoning: true,
7824
+ input: ["text", "image"],
7825
+ cost: {
7826
+ input: 3,
7827
+ output: 15,
7828
+ cacheRead: 0.3,
7829
+ cacheWrite: 0,
7830
+ },
7831
+ contextWindow: 1048576,
7832
+ maxTokens: 131072,
7833
+ },
7817
7834
  "laguna-s-2.1-free": {
7818
7835
  id: "laguna-s-2.1-free",
7819
7836
  name: "Laguna S 2.1 Free",
@@ -8711,13 +8728,13 @@ export const MODELS = {
8711
8728
  reasoning: true,
8712
8729
  input: ["text"],
8713
8730
  cost: {
8714
- input: 0.25,
8715
- output: 0.7999999999999999,
8731
+ input: 0.22,
8732
+ output: 0.85,
8716
8733
  cacheRead: 0.06,
8717
8734
  cacheWrite: 0,
8718
8735
  },
8719
8736
  contextWindow: 262144,
8720
- maxTokens: 80000,
8737
+ maxTokens: 262144,
8721
8738
  },
8722
8739
  "arcee-ai/virtuoso-large": {
8723
8740
  id: "arcee-ai/virtuoso-large",
@@ -9017,13 +9034,13 @@ export const MODELS = {
9017
9034
  reasoning: true,
9018
9035
  input: ["text"],
9019
9036
  cost: {
9020
- input: 0.098,
9021
- output: 0.196,
9022
- cacheRead: 0.0196,
9037
+ input: 0.14,
9038
+ output: 0.28,
9039
+ cacheRead: 0.028,
9023
9040
  cacheWrite: 0,
9024
9041
  },
9025
9042
  contextWindow: 1048576,
9026
- maxTokens: 4096,
9043
+ maxTokens: 393216,
9027
9044
  },
9028
9045
  "deepseek/deepseek-v4-pro": {
9029
9046
  id: "deepseek/deepseek-v4-pro",
@@ -9323,8 +9340,8 @@ export const MODELS = {
9323
9340
  reasoning: true,
9324
9341
  input: ["text", "image"],
9325
9342
  cost: {
9326
- input: 0.12,
9327
- output: 0.35,
9343
+ input: 0.14,
9344
+ output: 0.42,
9328
9345
  cacheRead: 0.049999999999999996,
9329
9346
  cacheWrite: 0,
9330
9347
  },
@@ -10122,9 +10139,9 @@ export const MODELS = {
10122
10139
  reasoning: true,
10123
10140
  input: ["text", "image"],
10124
10141
  cost: {
10125
- input: 0.684,
10126
- output: 3.42,
10127
- cacheRead: 0.144,
10142
+ input: 0.646,
10143
+ output: 2.7199999999999998,
10144
+ cacheRead: 0.1088,
10128
10145
  cacheWrite: 0,
10129
10146
  },
10130
10147
  contextWindow: 262144,
@@ -10139,7 +10156,7 @@ export const MODELS = {
10139
10156
  reasoning: true,
10140
10157
  input: ["text", "image"],
10141
10158
  cost: {
10142
- input: 0.78,
10159
+ input: 0.73,
10143
10160
  output: 3.5,
10144
10161
  cacheRead: 0.15,
10145
10162
  cacheWrite: 0,
@@ -10719,11 +10736,11 @@ export const MODELS = {
10719
10736
  cost: {
10720
10737
  input: 1.25,
10721
10738
  output: 10,
10722
- cacheRead: 0.125,
10739
+ cacheRead: 0.13,
10723
10740
  cacheWrite: 0,
10724
10741
  },
10725
10742
  contextWindow: 128000,
10726
- maxTokens: 16384,
10743
+ maxTokens: 32000,
10727
10744
  },
10728
10745
  "openai/gpt-5.1-codex": {
10729
10746
  id: "openai/gpt-5.1-codex",
@@ -10736,7 +10753,7 @@ export const MODELS = {
10736
10753
  cost: {
10737
10754
  input: 1.25,
10738
10755
  output: 10,
10739
- cacheRead: 0.125,
10756
+ cacheRead: 0.13,
10740
10757
  cacheWrite: 0,
10741
10758
  },
10742
10759
  contextWindow: 400000,
@@ -10770,11 +10787,11 @@ export const MODELS = {
10770
10787
  cost: {
10771
10788
  input: 0.25,
10772
10789
  output: 2,
10773
- cacheRead: 0.024999999999999998,
10790
+ cacheRead: 0.03,
10774
10791
  cacheWrite: 0,
10775
10792
  },
10776
10793
  contextWindow: 400000,
10777
- maxTokens: 100000,
10794
+ maxTokens: 128000,
10778
10795
  },
10779
10796
  "openai/gpt-5.2": {
10780
10797
  id: "openai/gpt-5.2",
@@ -10989,10 +11006,10 @@ export const MODELS = {
10989
11006
  reasoning: true,
10990
11007
  input: ["text", "image"],
10991
11008
  cost: {
10992
- input: 1,
10993
- output: 6,
10994
- cacheRead: 0.09999999999999999,
10995
- cacheWrite: 1.25,
11009
+ input: 0.5,
11010
+ output: 3,
11011
+ cacheRead: 0.049999999999999996,
11012
+ cacheWrite: 0.625,
10996
11013
  },
10997
11014
  contextWindow: 1050000,
10998
11015
  maxTokens: 128000,
@@ -11006,10 +11023,10 @@ export const MODELS = {
11006
11023
  reasoning: true,
11007
11024
  input: ["text", "image"],
11008
11025
  cost: {
11009
- input: 1,
11010
- output: 6,
11011
- cacheRead: 0.09999999999999999,
11012
- cacheWrite: 1.25,
11026
+ input: 0.5,
11027
+ output: 3,
11028
+ cacheRead: 0.049999999999999996,
11029
+ cacheWrite: 0.625,
11013
11030
  },
11014
11031
  contextWindow: 1050000,
11015
11032
  maxTokens: 128000,
@@ -11057,10 +11074,10 @@ export const MODELS = {
11057
11074
  reasoning: true,
11058
11075
  input: ["text", "image"],
11059
11076
  cost: {
11060
- input: 2.5,
11061
- output: 15,
11062
- cacheRead: 0.25,
11063
- cacheWrite: 3.125,
11077
+ input: 1.25,
11078
+ output: 7.5,
11079
+ cacheRead: 0.125,
11080
+ cacheWrite: 1.5625,
11064
11081
  },
11065
11082
  contextWindow: 1050000,
11066
11083
  maxTokens: 128000,
@@ -11074,10 +11091,10 @@ export const MODELS = {
11074
11091
  reasoning: true,
11075
11092
  input: ["text", "image"],
11076
11093
  cost: {
11077
- input: 2.5,
11078
- output: 15,
11079
- cacheRead: 0.25,
11080
- cacheWrite: 3.125,
11094
+ input: 1.25,
11095
+ output: 7.5,
11096
+ cacheRead: 0.125,
11097
+ cacheWrite: 1.5625,
11081
11098
  },
11082
11099
  contextWindow: 1050000,
11083
11100
  maxTokens: 128000,
@@ -11160,8 +11177,8 @@ export const MODELS = {
11160
11177
  input: ["text"],
11161
11178
  cost: {
11162
11179
  input: 0.03,
11163
- output: 0.13,
11164
- cacheRead: 0.03,
11180
+ output: 0.14,
11181
+ cacheRead: 0,
11165
11182
  cacheWrite: 0,
11166
11183
  },
11167
11184
  contextWindow: 131072,
@@ -11669,13 +11686,13 @@ export const MODELS = {
11669
11686
  reasoning: true,
11670
11687
  input: ["text"],
11671
11688
  cost: {
11672
- input: 0.13,
11673
- output: 0.52,
11689
+ input: 0.12,
11690
+ output: 0.5,
11674
11691
  cacheRead: 0,
11675
11692
  cacheWrite: 0,
11676
11693
  },
11677
11694
  contextWindow: 131072,
11678
- maxTokens: 8192,
11695
+ maxTokens: 16384,
11679
11696
  },
11680
11697
  "qwen/qwen3-30b-a3b-instruct-2507": {
11681
11698
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -11686,13 +11703,13 @@ export const MODELS = {
11686
11703
  reasoning: false,
11687
11704
  input: ["text"],
11688
11705
  cost: {
11689
- input: 0.09999999999999999,
11690
- output: 0.3,
11706
+ input: 0.04815,
11707
+ output: 0.19305,
11691
11708
  cacheRead: 0,
11692
11709
  cacheWrite: 0,
11693
11710
  },
11694
11711
  contextWindow: 262144,
11695
- maxTokens: 4096,
11712
+ maxTokens: 32000,
11696
11713
  },
11697
11714
  "qwen/qwen3-30b-a3b-thinking-2507": {
11698
11715
  id: "qwen/qwen3-30b-a3b-thinking-2507",
@@ -12162,13 +12179,13 @@ export const MODELS = {
12162
12179
  reasoning: true,
12163
12180
  input: ["text", "image"],
12164
12181
  cost: {
12165
- input: 0.28900000000000003,
12166
- output: 2.4,
12167
- cacheRead: 0,
12182
+ input: 0.3,
12183
+ output: 2,
12184
+ cacheRead: 0.15,
12168
12185
  cacheWrite: 0,
12169
12186
  },
12170
12187
  contextWindow: 262144,
12171
- maxTokens: 131072,
12188
+ maxTokens: 65536,
12172
12189
  },
12173
12190
  "qwen/qwen3.6-35b-a3b": {
12174
12191
  id: "qwen/qwen3.6-35b-a3b",
@@ -12740,9 +12757,9 @@ export const MODELS = {
12740
12757
  reasoning: true,
12741
12758
  input: ["text"],
12742
12759
  cost: {
12743
- input: 0.7756000000000001,
12744
- output: 2.4376,
12745
- cacheRead: 0.14404,
12760
+ input: 0.7686000000000001,
12761
+ output: 2.4156,
12762
+ cacheRead: 0.14274,
12746
12763
  cacheWrite: 0,
12747
12764
  },
12748
12765
  contextWindow: 1048576,
@@ -12876,13 +12893,13 @@ export const MODELS = {
12876
12893
  reasoning: true,
12877
12894
  input: ["text", "image"],
12878
12895
  cost: {
12879
- input: 3,
12880
- output: 15,
12881
- cacheRead: 0.3,
12896
+ input: 2.7,
12897
+ output: 13.5,
12898
+ cacheRead: 0.27,
12882
12899
  cacheWrite: 0,
12883
12900
  },
12884
12901
  contextWindow: 1048576,
12885
- maxTokens: 4096,
12902
+ maxTokens: 16384,
12886
12903
  },
12887
12904
  "~openai/gpt-latest": {
12888
12905
  id: "~openai/gpt-latest",
@@ -13549,23 +13566,6 @@ export const MODELS = {
13549
13566
  contextWindow: 1000000,
13550
13567
  maxTokens: 128000,
13551
13568
  },
13552
- "anthropic/claude-opus-4.7-fast": {
13553
- id: "anthropic/claude-opus-4.7-fast",
13554
- name: "Claude Opus 4.7 (Fast)",
13555
- api: "anthropic-messages",
13556
- provider: "vercel-ai-gateway",
13557
- baseUrl: "https://ai-gateway.vercel.sh",
13558
- reasoning: true,
13559
- input: ["text", "image"],
13560
- cost: {
13561
- input: 30,
13562
- output: 150,
13563
- cacheRead: 3,
13564
- cacheWrite: 37.5,
13565
- },
13566
- contextWindow: 1000000,
13567
- maxTokens: 128000,
13568
- },
13569
13569
  "anthropic/claude-opus-4.8": {
13570
13570
  id: "anthropic/claude-opus-4.8",
13571
13571
  name: "Claude Opus 4.8",
@@ -14025,23 +14025,6 @@ export const MODELS = {
14025
14025
  contextWindow: 1000000,
14026
14026
  maxTokens: 65000,
14027
14027
  },
14028
- "google/gemini-3.1-flash-lite-preview": {
14029
- id: "google/gemini-3.1-flash-lite-preview",
14030
- name: "Gemini 3.1 Flash Lite Preview",
14031
- api: "anthropic-messages",
14032
- provider: "vercel-ai-gateway",
14033
- baseUrl: "https://ai-gateway.vercel.sh",
14034
- reasoning: true,
14035
- input: ["text", "image"],
14036
- cost: {
14037
- input: 0.25,
14038
- output: 1.5,
14039
- cacheRead: 0.03,
14040
- cacheWrite: 0,
14041
- },
14042
- contextWindow: 1000000,
14043
- maxTokens: 65000,
14044
- },
14045
14028
  "google/gemini-3.1-pro-preview": {
14046
14029
  id: "google/gemini-3.1-pro-preview",
14047
14030
  name: "Gemini 3.1 Pro Preview",
@@ -14875,6 +14858,23 @@ export const MODELS = {
14875
14858
  contextWindow: 1000000,
14876
14859
  maxTokens: 131072,
14877
14860
  },
14861
+ "moonshotai/kimi-k3-fast": {
14862
+ id: "moonshotai/kimi-k3-fast",
14863
+ name: "Kimi K3 Fast",
14864
+ api: "anthropic-messages",
14865
+ provider: "vercel-ai-gateway",
14866
+ baseUrl: "https://ai-gateway.vercel.sh",
14867
+ reasoning: true,
14868
+ input: ["text", "image"],
14869
+ cost: {
14870
+ input: 4.5,
14871
+ output: 22.5,
14872
+ cacheRead: 0.44999999999999996,
14873
+ cacheWrite: 0,
14874
+ },
14875
+ contextWindow: 1000000,
14876
+ maxTokens: 131072,
14877
+ },
14878
14878
  "nvidia/nemotron-3-nano-30b-a3b": {
14879
14879
  id: "nvidia/nemotron-3-nano-30b-a3b",
14880
14880
  name: "Nemotron 3 Nano 30B A3B",
@@ -15096,23 +15096,6 @@ export const MODELS = {
15096
15096
  contextWindow: 400000,
15097
15097
  maxTokens: 128000,
15098
15098
  },
15099
- "openai/gpt-5-chat": {
15100
- id: "openai/gpt-5-chat",
15101
- name: "GPT 5 Chat",
15102
- api: "anthropic-messages",
15103
- provider: "vercel-ai-gateway",
15104
- baseUrl: "https://ai-gateway.vercel.sh",
15105
- reasoning: false,
15106
- input: ["text", "image"],
15107
- cost: {
15108
- input: 1.25,
15109
- output: 10,
15110
- cacheRead: 0.125,
15111
- cacheWrite: 0,
15112
- },
15113
- contextWindow: 128000,
15114
- maxTokens: 16384,
15115
- },
15116
15099
  "openai/gpt-5-codex": {
15117
15100
  id: "openai/gpt-5-codex",
15118
15101
  name: "GPT-5-Codex",
@@ -15283,23 +15266,6 @@ export const MODELS = {
15283
15266
  contextWindow: 400000,
15284
15267
  maxTokens: 128000,
15285
15268
  },
15286
- "openai/gpt-5.2-chat": {
15287
- id: "openai/gpt-5.2-chat",
15288
- name: "GPT 5.2 Chat",
15289
- api: "anthropic-messages",
15290
- provider: "vercel-ai-gateway",
15291
- baseUrl: "https://ai-gateway.vercel.sh",
15292
- reasoning: false,
15293
- input: ["text", "image"],
15294
- cost: {
15295
- input: 1.75,
15296
- output: 14,
15297
- cacheRead: 0.175,
15298
- cacheWrite: 0,
15299
- },
15300
- contextWindow: 128000,
15301
- maxTokens: 16384,
15302
- },
15303
15269
  "openai/gpt-5.2-codex": {
15304
15270
  id: "openai/gpt-5.2-codex",
15305
15271
  name: "GPT 5.2 Codex",