@dreb/ai 2.48.0 → 2.49.0

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.
@@ -1354,10 +1354,10 @@ export const MODELS = {
1354
1354
  reasoning: true,
1355
1355
  input: ["text", "image"],
1356
1356
  cost: {
1357
- input: 1,
1358
- output: 6,
1359
- cacheRead: 0.1,
1360
- cacheWrite: 1.25,
1357
+ input: 0.22,
1358
+ output: 1.32,
1359
+ cacheRead: 0.022,
1360
+ cacheWrite: 0.275,
1361
1361
  },
1362
1362
  contextWindow: 272000,
1363
1363
  maxTokens: 128000,
@@ -1371,10 +1371,10 @@ export const MODELS = {
1371
1371
  reasoning: true,
1372
1372
  input: ["text", "image"],
1373
1373
  cost: {
1374
- input: 5,
1375
- output: 30,
1376
- cacheRead: 0.5,
1377
- cacheWrite: 6.25,
1374
+ input: 5.5,
1375
+ output: 33,
1376
+ cacheRead: 0.55,
1377
+ cacheWrite: 6.88,
1378
1378
  },
1379
1379
  contextWindow: 272000,
1380
1380
  maxTokens: 128000,
@@ -1388,10 +1388,10 @@ export const MODELS = {
1388
1388
  reasoning: true,
1389
1389
  input: ["text", "image"],
1390
1390
  cost: {
1391
- input: 2.5,
1392
- output: 15,
1393
- cacheRead: 0.25,
1394
- cacheWrite: 3.125,
1391
+ input: 2.2,
1392
+ output: 13.2,
1393
+ cacheRead: 0.22,
1394
+ cacheWrite: 2.75,
1395
1395
  },
1396
1396
  contextWindow: 272000,
1397
1397
  maxTokens: 128000,
@@ -2752,10 +2752,10 @@ export const MODELS = {
2752
2752
  reasoning: true,
2753
2753
  input: ["text", "image"],
2754
2754
  cost: {
2755
- input: 1,
2756
- output: 6,
2757
- cacheRead: 0.1,
2758
- cacheWrite: 1.25,
2755
+ input: 0.2,
2756
+ output: 1.2,
2757
+ cacheRead: 0.02,
2758
+ cacheWrite: 0.25,
2759
2759
  },
2760
2760
  contextWindow: 1050000,
2761
2761
  maxTokens: 128000,
@@ -2786,10 +2786,10 @@ export const MODELS = {
2786
2786
  reasoning: true,
2787
2787
  input: ["text", "image"],
2788
2788
  cost: {
2789
- input: 2.5,
2790
- output: 15,
2791
- cacheRead: 0.25,
2792
- cacheWrite: 3.125,
2789
+ input: 2,
2790
+ output: 12,
2791
+ cacheRead: 0.2,
2792
+ cacheWrite: 2.5,
2793
2793
  },
2794
2794
  contextWindow: 1050000,
2795
2795
  maxTokens: 128000,
@@ -6700,10 +6700,10 @@ export const MODELS = {
6700
6700
  reasoning: true,
6701
6701
  input: ["text", "image"],
6702
6702
  cost: {
6703
- input: 1,
6704
- output: 6,
6705
- cacheRead: 0.1,
6706
- cacheWrite: 1.25,
6703
+ input: 0.2,
6704
+ output: 1.2,
6705
+ cacheRead: 0.02,
6706
+ cacheWrite: 0.25,
6707
6707
  },
6708
6708
  contextWindow: 1050000,
6709
6709
  maxTokens: 128000,
@@ -6734,10 +6734,10 @@ export const MODELS = {
6734
6734
  reasoning: true,
6735
6735
  input: ["text", "image"],
6736
6736
  cost: {
6737
- input: 2.5,
6738
- output: 15,
6739
- cacheRead: 0.25,
6740
- cacheWrite: 3.125,
6737
+ input: 2,
6738
+ output: 12,
6739
+ cacheRead: 0.2,
6740
+ cacheWrite: 2.5,
6741
6741
  },
6742
6742
  contextWindow: 1050000,
6743
6743
  maxTokens: 128000,
@@ -7292,7 +7292,7 @@ export const MODELS = {
7292
7292
  },
7293
7293
  "deepseek-v4-flash-free": {
7294
7294
  id: "deepseek-v4-flash-free",
7295
- name: "DeepSeek V4 Flash Free",
7295
+ name: "DeepSeek V4 Flash Free (New)",
7296
7296
  api: "openai-completions",
7297
7297
  provider: "opencode",
7298
7298
  baseUrl: "https://opencode.ai/zen/v1",
@@ -7741,10 +7741,10 @@ export const MODELS = {
7741
7741
  reasoning: true,
7742
7742
  input: ["text", "image"],
7743
7743
  cost: {
7744
- input: 1,
7745
- output: 6,
7746
- cacheRead: 0.1,
7747
- cacheWrite: 1.25,
7744
+ input: 0.2,
7745
+ output: 1.2,
7746
+ cacheRead: 0.02,
7747
+ cacheWrite: 0.25,
7748
7748
  },
7749
7749
  contextWindow: 1050000,
7750
7750
  maxTokens: 128000,
@@ -8059,7 +8059,7 @@ export const MODELS = {
8059
8059
  "opencode-go": {
8060
8060
  "deepseek-v4-flash": {
8061
8061
  id: "deepseek-v4-flash",
8062
- name: "DeepSeek V4 Flash",
8062
+ name: "DeepSeek V4 Flash (New)",
8063
8063
  api: "openai-completions",
8064
8064
  provider: "opencode-go",
8065
8065
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -8125,6 +8125,23 @@ export const MODELS = {
8125
8125
  contextWindow: 1000000,
8126
8126
  maxTokens: 131072,
8127
8127
  },
8128
+ "gpt-5.6-luna": {
8129
+ id: "gpt-5.6-luna",
8130
+ name: "GPT-5.6 Luna (2x usage)",
8131
+ api: "openai-responses",
8132
+ provider: "opencode-go",
8133
+ baseUrl: "https://opencode.ai/zen/go/v1",
8134
+ reasoning: true,
8135
+ input: ["text", "image"],
8136
+ cost: {
8137
+ input: 0.1,
8138
+ output: 0.6,
8139
+ cacheRead: 0.01,
8140
+ cacheWrite: 0.125,
8141
+ },
8142
+ contextWindow: 1050000,
8143
+ maxTokens: 128000,
8144
+ },
8128
8145
  "grok-4.5": {
8129
8146
  id: "grok-4.5",
8130
8147
  name: "Grok 4.5",
@@ -8195,7 +8212,7 @@ export const MODELS = {
8195
8212
  },
8196
8213
  "kimi-k3": {
8197
8214
  id: "kimi-k3",
8198
- name: "Kimi K3 (2x usage)",
8215
+ name: "Kimi K3",
8199
8216
  api: "openai-completions",
8200
8217
  provider: "opencode-go",
8201
8218
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -9249,6 +9266,23 @@ export const MODELS = {
9249
9266
  contextWindow: 1048576,
9250
9267
  maxTokens: 393216,
9251
9268
  },
9269
+ "deepseek/deepseek-v4-flash-0731": {
9270
+ id: "deepseek/deepseek-v4-flash-0731",
9271
+ name: "DeepSeek: DeepSeek V4 Flash 0731",
9272
+ api: "openai-completions",
9273
+ provider: "openrouter",
9274
+ baseUrl: "https://openrouter.ai/api/v1",
9275
+ reasoning: true,
9276
+ input: ["text"],
9277
+ cost: {
9278
+ input: 0.14,
9279
+ output: 0.28,
9280
+ cacheRead: 0.0028,
9281
+ cacheWrite: 0,
9282
+ },
9283
+ contextWindow: 1048576,
9284
+ maxTokens: 384000,
9285
+ },
9252
9286
  "deepseek/deepseek-v4-pro": {
9253
9287
  id: "deepseek/deepseek-v4-pro",
9254
9288
  name: "DeepSeek: DeepSeek V4 Pro",
@@ -10167,23 +10201,6 @@ export const MODELS = {
10167
10201
  contextWindow: 256000,
10168
10202
  maxTokens: 4096,
10169
10203
  },
10170
- "mistralai/devstral-2512": {
10171
- id: "mistralai/devstral-2512",
10172
- name: "Mistral: Devstral 2 2512",
10173
- api: "openai-completions",
10174
- provider: "openrouter",
10175
- baseUrl: "https://openrouter.ai/api/v1",
10176
- reasoning: false,
10177
- input: ["text"],
10178
- cost: {
10179
- input: 0.39999999999999997,
10180
- output: 2,
10181
- cacheRead: 0.04,
10182
- cacheWrite: 0,
10183
- },
10184
- contextWindow: 262144,
10185
- maxTokens: 4096,
10186
- },
10187
10204
  "mistralai/ministral-14b-2512": {
10188
10205
  id: "mistralai/ministral-14b-2512",
10189
10206
  name: "Mistral: Ministral 3 14B 2512",
@@ -10397,13 +10414,13 @@ export const MODELS = {
10397
10414
  reasoning: false,
10398
10415
  input: ["text", "image"],
10399
10416
  cost: {
10400
- input: 0.09999999999999999,
10401
- output: 0.3,
10402
- cacheRead: 0.01,
10417
+ input: 0.075,
10418
+ output: 0.19999999999999998,
10419
+ cacheRead: 0,
10403
10420
  cacheWrite: 0,
10404
10421
  },
10405
10422
  contextWindow: 256000,
10406
- maxTokens: 4096,
10423
+ maxTokens: 16384,
10407
10424
  },
10408
10425
  "mistralai/mixtral-8x22b-instruct": {
10409
10426
  id: "mistralai/mixtral-8x22b-instruct",
@@ -10516,9 +10533,9 @@ export const MODELS = {
10516
10533
  reasoning: true,
10517
10534
  input: ["text", "image"],
10518
10535
  cost: {
10519
- input: 0.95,
10520
- output: 4,
10521
- cacheRead: 0.16,
10536
+ input: 0.589,
10537
+ output: 2.48,
10538
+ cacheRead: 0.0992,
10522
10539
  cacheWrite: 0,
10523
10540
  },
10524
10541
  contextWindow: 262144,
@@ -11119,23 +11136,6 @@ export const MODELS = {
11119
11136
  contextWindow: 400000,
11120
11137
  maxTokens: 128000,
11121
11138
  },
11122
- "openai/gpt-5.1-chat": {
11123
- id: "openai/gpt-5.1-chat",
11124
- name: "OpenAI: GPT-5.1 Chat",
11125
- api: "openai-completions",
11126
- provider: "openrouter",
11127
- baseUrl: "https://openrouter.ai/api/v1",
11128
- reasoning: false,
11129
- input: ["text", "image"],
11130
- cost: {
11131
- input: 1.25,
11132
- output: 10,
11133
- cacheRead: 0.13,
11134
- cacheWrite: 0,
11135
- },
11136
- contextWindow: 128000,
11137
- maxTokens: 32000,
11138
- },
11139
11139
  "openai/gpt-5.1-codex": {
11140
11140
  id: "openai/gpt-5.1-codex",
11141
11141
  name: "OpenAI: GPT-5.1-Codex",
@@ -11690,8 +11690,8 @@ export const MODELS = {
11690
11690
  input: ["text"],
11691
11691
  cost: {
11692
11692
  input: 0.03,
11693
- output: 0.13,
11694
- cacheRead: 0.03,
11693
+ output: 0.14,
11694
+ cacheRead: 0,
11695
11695
  cacheWrite: 0,
11696
11696
  },
11697
11697
  contextWindow: 131072,
@@ -12403,13 +12403,13 @@ export const MODELS = {
12403
12403
  reasoning: false,
12404
12404
  input: ["text", "image"],
12405
12405
  cost: {
12406
- input: 0.15,
12407
- output: 0.6,
12406
+ input: 0.13,
12407
+ output: 0.52,
12408
12408
  cacheRead: 0,
12409
12409
  cacheWrite: 0,
12410
12410
  },
12411
12411
  contextWindow: 262144,
12412
- maxTokens: 16384,
12412
+ maxTokens: 32768,
12413
12413
  },
12414
12414
  "qwen/qwen3-vl-30b-a3b-thinking": {
12415
12415
  id: "qwen/qwen3-vl-30b-a3b-thinking",
@@ -12732,7 +12732,7 @@ export const MODELS = {
12732
12732
  cacheWrite: 1.84375,
12733
12733
  },
12734
12734
  contextWindow: 1000000,
12735
- maxTokens: 65536,
12735
+ maxTokens: 131072,
12736
12736
  },
12737
12737
  "qwen/qwen3.7-plus": {
12738
12738
  id: "qwen/qwen3.7-plus",
@@ -12749,7 +12749,7 @@ export const MODELS = {
12749
12749
  cacheWrite: 0.39999999999999997,
12750
12750
  },
12751
12751
  contextWindow: 1000000,
12752
- maxTokens: 65536,
12752
+ maxTokens: 131072,
12753
12753
  },
12754
12754
  "rekaai/reka-edge": {
12755
12755
  id: "rekaai/reka-edge",
@@ -13219,9 +13219,9 @@ export const MODELS = {
13219
13219
  reasoning: true,
13220
13220
  input: ["text"],
13221
13221
  cost: {
13222
- input: 0.966,
13223
- output: 3.036,
13224
- cacheRead: 0.1794,
13222
+ input: 1.12,
13223
+ output: 3.52,
13224
+ cacheRead: 0.20800000000000002,
13225
13225
  cacheWrite: 0,
13226
13226
  },
13227
13227
  contextWindow: 1048576,