@dreb/ai 2.39.0 → 2.40.1

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.
@@ -1260,6 +1260,57 @@ export const MODELS = {
1260
1260
  contextWindow: 272000,
1261
1261
  maxTokens: 128000,
1262
1262
  },
1263
+ "openai.gpt-5.6-luna": {
1264
+ id: "openai.gpt-5.6-luna",
1265
+ name: "GPT-5.6 Luna",
1266
+ api: "bedrock-converse-stream",
1267
+ provider: "amazon-bedrock",
1268
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1269
+ reasoning: true,
1270
+ input: ["text", "image"],
1271
+ cost: {
1272
+ input: 1,
1273
+ output: 6,
1274
+ cacheRead: 0.1,
1275
+ cacheWrite: 1.25,
1276
+ },
1277
+ contextWindow: 272000,
1278
+ maxTokens: 128000,
1279
+ },
1280
+ "openai.gpt-5.6-sol": {
1281
+ id: "openai.gpt-5.6-sol",
1282
+ name: "GPT-5.6 Sol",
1283
+ api: "bedrock-converse-stream",
1284
+ provider: "amazon-bedrock",
1285
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1286
+ reasoning: true,
1287
+ input: ["text", "image"],
1288
+ cost: {
1289
+ input: 5,
1290
+ output: 30,
1291
+ cacheRead: 0.5,
1292
+ cacheWrite: 6.25,
1293
+ },
1294
+ contextWindow: 272000,
1295
+ maxTokens: 128000,
1296
+ },
1297
+ "openai.gpt-5.6-terra": {
1298
+ id: "openai.gpt-5.6-terra",
1299
+ name: "GPT-5.6 Terra",
1300
+ api: "bedrock-converse-stream",
1301
+ provider: "amazon-bedrock",
1302
+ baseUrl: "https://bedrock-runtime.us-east-1.amazonaws.com",
1303
+ reasoning: true,
1304
+ input: ["text", "image"],
1305
+ cost: {
1306
+ input: 2.5,
1307
+ output: 15,
1308
+ cacheRead: 0.25,
1309
+ cacheWrite: 3.125,
1310
+ },
1311
+ contextWindow: 272000,
1312
+ maxTokens: 128000,
1313
+ },
1263
1314
  "openai.gpt-oss-120b": {
1264
1315
  id: "openai.gpt-oss-120b",
1265
1316
  name: "gpt-oss-120b",
@@ -8553,8 +8604,8 @@ export const MODELS = {
8553
8604
  reasoning: true,
8554
8605
  input: ["text"],
8555
8606
  cost: {
8556
- input: 0.21,
8557
- output: 0.7899999999999999,
8607
+ input: 0.25,
8608
+ output: 0.95,
8558
8609
  cacheRead: 0.13,
8559
8610
  cacheWrite: 0,
8560
8611
  },
@@ -8655,13 +8706,13 @@ export const MODELS = {
8655
8706
  reasoning: true,
8656
8707
  input: ["text"],
8657
8708
  cost: {
8658
- input: 0.077,
8659
- output: 0.154,
8660
- cacheRead: 0.015399999999999999,
8709
+ input: 0.09,
8710
+ output: 0.18,
8711
+ cacheRead: 0.018,
8661
8712
  cacheWrite: 0,
8662
8713
  },
8663
8714
  contextWindow: 1048576,
8664
- maxTokens: 384000,
8715
+ maxTokens: 65536,
8665
8716
  },
8666
8717
  "deepseek/deepseek-v4-pro": {
8667
8718
  id: "deepseek/deepseek-v4-pro",
@@ -8961,13 +9012,13 @@ export const MODELS = {
8961
9012
  reasoning: true,
8962
9013
  input: ["text", "image"],
8963
9014
  cost: {
8964
- input: 0.12,
9015
+ input: 0.06,
8965
9016
  output: 0.35,
8966
- cacheRead: 0.09,
9017
+ cacheRead: 0,
8967
9018
  cacheWrite: 0,
8968
9019
  },
8969
9020
  contextWindow: 262144,
8970
- maxTokens: 262144,
9021
+ maxTokens: 8192,
8971
9022
  },
8972
9023
  "google/gemma-4-31b-it:free": {
8973
9024
  id: "google/gemma-4-31b-it:free",
@@ -8984,7 +9035,7 @@ export const MODELS = {
8984
9035
  cacheWrite: 0,
8985
9036
  },
8986
9037
  contextWindow: 262144,
8987
- maxTokens: 8192,
9038
+ maxTokens: 32768,
8988
9039
  },
8989
9040
  "ibm-granite/granite-4.1-8b": {
8990
9041
  id: "ibm-granite/granite-4.1-8b",
@@ -9071,6 +9122,23 @@ export const MODELS = {
9071
9122
  contextWindow: 262144,
9072
9123
  maxTokens: 65536,
9073
9124
  },
9125
+ "kwaipilot/kat-coder-air-v2.5": {
9126
+ id: "kwaipilot/kat-coder-air-v2.5",
9127
+ name: "Kwaipilot: KAT-Coder-Air V2.5",
9128
+ api: "openai-completions",
9129
+ provider: "openrouter",
9130
+ baseUrl: "https://openrouter.ai/api/v1",
9131
+ reasoning: false,
9132
+ input: ["text"],
9133
+ cost: {
9134
+ input: 0.15,
9135
+ output: 0.6,
9136
+ cacheRead: 0.03,
9137
+ cacheWrite: 0,
9138
+ },
9139
+ contextWindow: 256000,
9140
+ maxTokens: 80000,
9141
+ },
9074
9142
  "kwaipilot/kat-coder-pro-v2": {
9075
9143
  id: "kwaipilot/kat-coder-pro-v2",
9076
9144
  name: "Kwaipilot: KAT-Coder-Pro V2",
@@ -9088,22 +9156,22 @@ export const MODELS = {
9088
9156
  contextWindow: 256000,
9089
9157
  maxTokens: 80000,
9090
9158
  },
9091
- "liquid/lfm-2.5-1.2b-thinking:free": {
9092
- id: "liquid/lfm-2.5-1.2b-thinking:free",
9093
- name: "LiquidAI: LFM2.5-1.2B-Thinking (free)",
9159
+ "kwaipilot/kat-coder-pro-v2.5": {
9160
+ id: "kwaipilot/kat-coder-pro-v2.5",
9161
+ name: "Kwaipilot: KAT-Coder-Pro V2.5",
9094
9162
  api: "openai-completions",
9095
9163
  provider: "openrouter",
9096
9164
  baseUrl: "https://openrouter.ai/api/v1",
9097
- reasoning: true,
9165
+ reasoning: false,
9098
9166
  input: ["text"],
9099
9167
  cost: {
9100
- input: 0,
9101
- output: 0,
9102
- cacheRead: 0,
9168
+ input: 0.74,
9169
+ output: 2.96,
9170
+ cacheRead: 0.15,
9103
9171
  cacheWrite: 0,
9104
9172
  },
9105
- contextWindow: 32768,
9106
- maxTokens: 4096,
9173
+ contextWindow: 256000,
9174
+ maxTokens: 80000,
9107
9175
  },
9108
9176
  "meta-llama/llama-3.1-70b-instruct": {
9109
9177
  id: "meta-llama/llama-3.1-70b-instruct",
@@ -9692,9 +9760,9 @@ export const MODELS = {
9692
9760
  reasoning: true,
9693
9761
  input: ["text", "image"],
9694
9762
  cost: {
9695
- input: 0.72,
9696
- output: 3.5,
9697
- cacheRead: 0.15,
9763
+ input: 0.719,
9764
+ output: 3.49,
9765
+ cacheRead: 0.149,
9698
9766
  cacheWrite: 0,
9699
9767
  },
9700
9768
  contextWindow: 262144,
@@ -10221,11 +10289,11 @@ export const MODELS = {
10221
10289
  cost: {
10222
10290
  input: 0.049999999999999996,
10223
10291
  output: 0.39999999999999997,
10224
- cacheRead: 0.01,
10292
+ cacheRead: 0.005,
10225
10293
  cacheWrite: 0,
10226
10294
  },
10227
10295
  contextWindow: 400000,
10228
- maxTokens: 4096,
10296
+ maxTokens: 128000,
10229
10297
  },
10230
10298
  "openai/gpt-5-pro": {
10231
10299
  id: "openai/gpt-5-pro",
@@ -10255,7 +10323,7 @@ export const MODELS = {
10255
10323
  cost: {
10256
10324
  input: 1.25,
10257
10325
  output: 10,
10258
- cacheRead: 0.13,
10326
+ cacheRead: 0.125,
10259
10327
  cacheWrite: 0,
10260
10328
  },
10261
10329
  contextWindow: 400000,
@@ -10695,25 +10763,8 @@ export const MODELS = {
10695
10763
  reasoning: true,
10696
10764
  input: ["text"],
10697
10765
  cost: {
10698
- input: 0.036,
10699
- output: 0.18,
10700
- cacheRead: 0,
10701
- cacheWrite: 0,
10702
- },
10703
- contextWindow: 131072,
10704
- maxTokens: 4096,
10705
- },
10706
- "openai/gpt-oss-120b:free": {
10707
- id: "openai/gpt-oss-120b:free",
10708
- name: "OpenAI: gpt-oss-120b (free)",
10709
- api: "openai-completions",
10710
- provider: "openrouter",
10711
- baseUrl: "https://openrouter.ai/api/v1",
10712
- reasoning: true,
10713
- input: ["text"],
10714
- cost: {
10715
- input: 0,
10716
- output: 0,
10766
+ input: 0.03,
10767
+ output: 0.15,
10717
10768
  cacheRead: 0,
10718
10769
  cacheWrite: 0,
10719
10770
  },
@@ -11715,13 +11766,13 @@ export const MODELS = {
11715
11766
  reasoning: true,
11716
11767
  input: ["text", "image"],
11717
11768
  cost: {
11718
- input: 0.28500000000000003,
11769
+ input: 0.28900000000000003,
11719
11770
  output: 2.4,
11720
- cacheRead: 0.15,
11771
+ cacheRead: 0,
11721
11772
  cacheWrite: 0,
11722
11773
  },
11723
11774
  contextWindow: 262144,
11724
- maxTokens: 262140,
11775
+ maxTokens: 131072,
11725
11776
  },
11726
11777
  "qwen/qwen3.6-35b-a3b": {
11727
11778
  id: "qwen/qwen3.6-35b-a3b",
@@ -12248,7 +12299,7 @@ export const MODELS = {
12248
12299
  cacheWrite: 0,
12249
12300
  },
12250
12301
  contextWindow: 202752,
12251
- maxTokens: 4096,
12302
+ maxTokens: 128000,
12252
12303
  },
12253
12304
  "z-ai/glm-5-turbo": {
12254
12305
  id: "z-ai/glm-5-turbo",
@@ -12293,13 +12344,13 @@ export const MODELS = {
12293
12344
  reasoning: true,
12294
12345
  input: ["text"],
12295
12346
  cost: {
12296
- input: 0.42,
12297
- output: 1.32,
12298
- cacheRead: 0.078,
12347
+ input: 0.9239999999999999,
12348
+ output: 2.904,
12349
+ cacheRead: 0.1716,
12299
12350
  cacheWrite: 0,
12300
12351
  },
12301
12352
  contextWindow: 1048576,
12302
- maxTokens: 131072,
12353
+ maxTokens: 128000,
12303
12354
  },
12304
12355
  "z-ai/glm-5v-turbo": {
12305
12356
  id: "z-ai/glm-5v-turbo",
@@ -12983,23 +13034,6 @@ export const MODELS = {
12983
13034
  contextWindow: 200000,
12984
13035
  maxTokens: 4096,
12985
13036
  },
12986
- "anthropic/claude-3.5-haiku": {
12987
- id: "anthropic/claude-3.5-haiku",
12988
- name: "Claude 3.5 Haiku",
12989
- api: "anthropic-messages",
12990
- provider: "vercel-ai-gateway",
12991
- baseUrl: "https://ai-gateway.vercel.sh",
12992
- reasoning: false,
12993
- input: ["text", "image"],
12994
- cost: {
12995
- input: 0.7999999999999999,
12996
- output: 4,
12997
- cacheRead: 0.08,
12998
- cacheWrite: 1,
12999
- },
13000
- contextWindow: 200000,
13001
- maxTokens: 8192,
13002
- },
13003
13037
  "anthropic/claude-fable-5": {
13004
13038
  id: "anthropic/claude-fable-5",
13005
13039
  name: "Claude Fable 5",
@@ -13049,7 +13083,7 @@ export const MODELS = {
13049
13083
  cacheWrite: 18.75,
13050
13084
  },
13051
13085
  contextWindow: 200000,
13052
- maxTokens: 32000,
13086
+ maxTokens: 8192,
13053
13087
  },
13054
13088
  "anthropic/claude-opus-4.1": {
13055
13089
  id: "anthropic/claude-opus-4.1",
@@ -13151,7 +13185,7 @@ export const MODELS = {
13151
13185
  cacheWrite: 3.75,
13152
13186
  },
13153
13187
  contextWindow: 1000000,
13154
- maxTokens: 64000,
13188
+ maxTokens: 8192,
13155
13189
  },
13156
13190
  "anthropic/claude-sonnet-4.5": {
13157
13191
  id: "anthropic/claude-sonnet-4.5",
@@ -13204,23 +13238,6 @@ export const MODELS = {
13204
13238
  contextWindow: 1000000,
13205
13239
  maxTokens: 128000,
13206
13240
  },
13207
- "arcee-ai/trinity-large-preview": {
13208
- id: "arcee-ai/trinity-large-preview",
13209
- name: "Trinity Large Preview",
13210
- api: "anthropic-messages",
13211
- provider: "vercel-ai-gateway",
13212
- baseUrl: "https://ai-gateway.vercel.sh",
13213
- reasoning: false,
13214
- input: ["text"],
13215
- cost: {
13216
- input: 0.25,
13217
- output: 1,
13218
- cacheRead: 0,
13219
- cacheWrite: 0,
13220
- },
13221
- contextWindow: 131000,
13222
- maxTokens: 131000,
13223
- },
13224
13241
  "arcee-ai/trinity-large-thinking": {
13225
13242
  id: "arcee-ai/trinity-large-thinking",
13226
13243
  name: "Trinity Large Thinking",
@@ -13349,12 +13366,12 @@ export const MODELS = {
13349
13366
  reasoning: true,
13350
13367
  input: ["text"],
13351
13368
  cost: {
13352
- input: 0.6,
13353
- output: 1.7,
13354
- cacheRead: 0,
13369
+ input: 0.21,
13370
+ output: 0.7899999999999999,
13371
+ cacheRead: 0.13,
13355
13372
  cacheWrite: 0,
13356
13373
  },
13357
- contextWindow: 128000,
13374
+ contextWindow: 163840,
13358
13375
  maxTokens: 128000,
13359
13376
  },
13360
13377
  "deepseek/deepseek-v3.1-terminus": {
@@ -13748,40 +13765,6 @@ export const MODELS = {
13748
13765
  contextWindow: 256000,
13749
13766
  maxTokens: 80000,
13750
13767
  },
13751
- "meituan/longcat-flash-chat": {
13752
- id: "meituan/longcat-flash-chat",
13753
- name: "LongCat Flash Chat",
13754
- api: "anthropic-messages",
13755
- provider: "vercel-ai-gateway",
13756
- baseUrl: "https://ai-gateway.vercel.sh",
13757
- reasoning: false,
13758
- input: ["text"],
13759
- cost: {
13760
- input: 0,
13761
- output: 0,
13762
- cacheRead: 0,
13763
- cacheWrite: 0,
13764
- },
13765
- contextWindow: 128000,
13766
- maxTokens: 100000,
13767
- },
13768
- "meituan/longcat-flash-thinking-2601": {
13769
- id: "meituan/longcat-flash-thinking-2601",
13770
- name: "LongCat Flash Thinking 2601",
13771
- api: "anthropic-messages",
13772
- provider: "vercel-ai-gateway",
13773
- baseUrl: "https://ai-gateway.vercel.sh",
13774
- reasoning: true,
13775
- input: ["text"],
13776
- cost: {
13777
- input: 0,
13778
- output: 0,
13779
- cacheRead: 0,
13780
- cacheWrite: 0,
13781
- },
13782
- contextWindow: 32768,
13783
- maxTokens: 32768,
13784
- },
13785
13768
  "meta/llama-3.1-70b": {
13786
13769
  id: "meta/llama-3.1-70b",
13787
13770
  name: "Llama 3.1 70B Instruct",
@@ -14088,23 +14071,6 @@ export const MODELS = {
14088
14071
  contextWindow: 256000,
14089
14072
  maxTokens: 256000,
14090
14073
  },
14091
- "mistral/devstral-small": {
14092
- id: "mistral/devstral-small",
14093
- name: "Devstral Small 1.1",
14094
- api: "anthropic-messages",
14095
- provider: "vercel-ai-gateway",
14096
- baseUrl: "https://ai-gateway.vercel.sh",
14097
- reasoning: false,
14098
- input: ["text"],
14099
- cost: {
14100
- input: 0.09999999999999999,
14101
- output: 0.3,
14102
- cacheRead: 0,
14103
- cacheWrite: 0,
14104
- },
14105
- contextWindow: 128000,
14106
- maxTokens: 64000,
14107
- },
14108
14074
  "mistral/devstral-small-2": {
14109
14075
  id: "mistral/devstral-small-2",
14110
14076
  name: "Devstral Small 2",
@@ -14309,23 +14275,6 @@ export const MODELS = {
14309
14275
  contextWindow: 128000,
14310
14276
  maxTokens: 4000,
14311
14277
  },
14312
- "mistral/pixtral-large": {
14313
- id: "mistral/pixtral-large",
14314
- name: "Pixtral Large",
14315
- api: "anthropic-messages",
14316
- provider: "vercel-ai-gateway",
14317
- baseUrl: "https://ai-gateway.vercel.sh",
14318
- reasoning: false,
14319
- input: ["text", "image"],
14320
- cost: {
14321
- input: 2,
14322
- output: 6,
14323
- cacheRead: 0,
14324
- cacheWrite: 0,
14325
- },
14326
- contextWindow: 128000,
14327
- maxTokens: 4000,
14328
- },
14329
14278
  "moonshotai/kimi-k2": {
14330
14279
  id: "moonshotai/kimi-k2",
14331
14280
  name: "Kimi K2 Instruct",
@@ -15465,40 +15414,6 @@ export const MODELS = {
15465
15414
  contextWindow: 256000,
15466
15415
  maxTokens: 256000,
15467
15416
  },
15468
- "xiaomi/mimo-v2-flash": {
15469
- id: "xiaomi/mimo-v2-flash",
15470
- name: "MiMo V2 Flash",
15471
- api: "anthropic-messages",
15472
- provider: "vercel-ai-gateway",
15473
- baseUrl: "https://ai-gateway.vercel.sh",
15474
- reasoning: true,
15475
- input: ["text"],
15476
- cost: {
15477
- input: 0.09999999999999999,
15478
- output: 0.3,
15479
- cacheRead: 0.01,
15480
- cacheWrite: 0,
15481
- },
15482
- contextWindow: 262144,
15483
- maxTokens: 32000,
15484
- },
15485
- "xiaomi/mimo-v2-pro": {
15486
- id: "xiaomi/mimo-v2-pro",
15487
- name: "MiMo V2 Pro",
15488
- api: "anthropic-messages",
15489
- provider: "vercel-ai-gateway",
15490
- baseUrl: "https://ai-gateway.vercel.sh",
15491
- reasoning: true,
15492
- input: ["text"],
15493
- cost: {
15494
- input: 1,
15495
- output: 3,
15496
- cacheRead: 0.19999999999999998,
15497
- cacheWrite: 0,
15498
- },
15499
- contextWindow: 1000000,
15500
- maxTokens: 128000,
15501
- },
15502
15417
  "xiaomi/mimo-v2.5": {
15503
15418
  id: "xiaomi/mimo-v2.5",
15504
15419
  name: "MiMo M2.5",