@dreb/ai 2.38.1 → 2.40.0

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.
@@ -8553,8 +8553,8 @@ export const MODELS = {
8553
8553
  reasoning: true,
8554
8554
  input: ["text"],
8555
8555
  cost: {
8556
- input: 0.21,
8557
- output: 0.7899999999999999,
8556
+ input: 0.25,
8557
+ output: 0.95,
8558
8558
  cacheRead: 0.13,
8559
8559
  cacheWrite: 0,
8560
8560
  },
@@ -8655,13 +8655,13 @@ export const MODELS = {
8655
8655
  reasoning: true,
8656
8656
  input: ["text"],
8657
8657
  cost: {
8658
- input: 0.08399999999999999,
8659
- output: 0.16799999999999998,
8660
- cacheRead: 0.016800000000000002,
8658
+ input: 0.09,
8659
+ output: 0.18,
8660
+ cacheRead: 0.018,
8661
8661
  cacheWrite: 0,
8662
8662
  },
8663
8663
  contextWindow: 1048576,
8664
- maxTokens: 384000,
8664
+ maxTokens: 65536,
8665
8665
  },
8666
8666
  "deepseek/deepseek-v4-pro": {
8667
8667
  id: "deepseek/deepseek-v4-pro",
@@ -8984,7 +8984,7 @@ export const MODELS = {
8984
8984
  cacheWrite: 0,
8985
8985
  },
8986
8986
  contextWindow: 262144,
8987
- maxTokens: 8192,
8987
+ maxTokens: 32768,
8988
8988
  },
8989
8989
  "ibm-granite/granite-4.1-8b": {
8990
8990
  id: "ibm-granite/granite-4.1-8b",
@@ -9088,23 +9088,6 @@ export const MODELS = {
9088
9088
  contextWindow: 256000,
9089
9089
  maxTokens: 80000,
9090
9090
  },
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)",
9094
- api: "openai-completions",
9095
- provider: "openrouter",
9096
- baseUrl: "https://openrouter.ai/api/v1",
9097
- reasoning: true,
9098
- input: ["text"],
9099
- cost: {
9100
- input: 0,
9101
- output: 0,
9102
- cacheRead: 0,
9103
- cacheWrite: 0,
9104
- },
9105
- contextWindow: 32768,
9106
- maxTokens: 4096,
9107
- },
9108
9091
  "meta-llama/llama-3.1-70b-instruct": {
9109
9092
  id: "meta-llama/llama-3.1-70b-instruct",
9110
9093
  name: "Meta: Llama 3.1 70B Instruct",
@@ -9182,8 +9165,8 @@ export const MODELS = {
9182
9165
  reasoning: false,
9183
9166
  input: ["text", "image"],
9184
9167
  cost: {
9185
- input: 0.15,
9186
- output: 0.6,
9168
+ input: 0.19999999999999998,
9169
+ output: 0.7999999999999999,
9187
9170
  cacheRead: 0,
9188
9171
  cacheWrite: 0,
9189
9172
  },
@@ -9692,9 +9675,9 @@ export const MODELS = {
9692
9675
  reasoning: true,
9693
9676
  input: ["text", "image"],
9694
9677
  cost: {
9695
- input: 0.72,
9678
+ input: 0.719,
9696
9679
  output: 3.49,
9697
- cacheRead: 0.159,
9680
+ cacheRead: 0.149,
9698
9681
  cacheWrite: 0,
9699
9682
  },
9700
9683
  contextWindow: 262144,
@@ -10221,11 +10204,11 @@ export const MODELS = {
10221
10204
  cost: {
10222
10205
  input: 0.049999999999999996,
10223
10206
  output: 0.39999999999999997,
10224
- cacheRead: 0.01,
10207
+ cacheRead: 0.005,
10225
10208
  cacheWrite: 0,
10226
10209
  },
10227
10210
  contextWindow: 400000,
10228
- maxTokens: 4096,
10211
+ maxTokens: 128000,
10229
10212
  },
10230
10213
  "openai/gpt-5-pro": {
10231
10214
  id: "openai/gpt-5-pro",
@@ -10255,7 +10238,7 @@ export const MODELS = {
10255
10238
  cost: {
10256
10239
  input: 1.25,
10257
10240
  output: 10,
10258
- cacheRead: 0.13,
10241
+ cacheRead: 0.125,
10259
10242
  cacheWrite: 0,
10260
10243
  },
10261
10244
  contextWindow: 400000,
@@ -10695,25 +10678,8 @@ export const MODELS = {
10695
10678
  reasoning: true,
10696
10679
  input: ["text"],
10697
10680
  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,
10681
+ input: 3e-8,
10682
+ output: 1.5e-7,
10717
10683
  cacheRead: 0,
10718
10684
  cacheWrite: 0,
10719
10685
  },
@@ -11155,7 +11121,7 @@ export const MODELS = {
11155
11121
  input: ["text"],
11156
11122
  cost: {
11157
11123
  input: 0.09,
11158
- output: 0.09999999999999999,
11124
+ output: 0.55,
11159
11125
  cacheRead: 0,
11160
11126
  cacheWrite: 0,
11161
11127
  },
@@ -12175,12 +12141,12 @@ export const MODELS = {
12175
12141
  input: ["text"],
12176
12142
  cost: {
12177
12143
  input: 0.43,
12178
- output: 1.74,
12144
+ output: 1.75,
12179
12145
  cacheRead: 0.08,
12180
12146
  cacheWrite: 0,
12181
12147
  },
12182
- contextWindow: 202752,
12183
- maxTokens: 131072,
12148
+ contextWindow: 200000,
12149
+ maxTokens: 16384,
12184
12150
  },
12185
12151
  "z-ai/glm-4.6v": {
12186
12152
  id: "z-ai/glm-4.6v",
@@ -12293,13 +12259,13 @@ export const MODELS = {
12293
12259
  reasoning: true,
12294
12260
  input: ["text"],
12295
12261
  cost: {
12296
- input: 0.42,
12297
- output: 1.32,
12298
- cacheRead: 0.078,
12262
+ input: 0.9299999999999999,
12263
+ output: 3,
12264
+ cacheRead: 0.18,
12299
12265
  cacheWrite: 0,
12300
12266
  },
12301
12267
  contextWindow: 1048576,
12302
- maxTokens: 131072,
12268
+ maxTokens: 32768,
12303
12269
  },
12304
12270
  "z-ai/glm-5v-turbo": {
12305
12271
  id: "z-ai/glm-5v-turbo",
@@ -12983,23 +12949,6 @@ export const MODELS = {
12983
12949
  contextWindow: 200000,
12984
12950
  maxTokens: 4096,
12985
12951
  },
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
12952
  "anthropic/claude-fable-5": {
13004
12953
  id: "anthropic/claude-fable-5",
13005
12954
  name: "Claude Fable 5",
@@ -13049,7 +12998,7 @@ export const MODELS = {
13049
12998
  cacheWrite: 18.75,
13050
12999
  },
13051
13000
  contextWindow: 200000,
13052
- maxTokens: 32000,
13001
+ maxTokens: 8192,
13053
13002
  },
13054
13003
  "anthropic/claude-opus-4.1": {
13055
13004
  id: "anthropic/claude-opus-4.1",
@@ -13151,7 +13100,7 @@ export const MODELS = {
13151
13100
  cacheWrite: 3.75,
13152
13101
  },
13153
13102
  contextWindow: 1000000,
13154
- maxTokens: 64000,
13103
+ maxTokens: 8192,
13155
13104
  },
13156
13105
  "anthropic/claude-sonnet-4.5": {
13157
13106
  id: "anthropic/claude-sonnet-4.5",
@@ -13204,23 +13153,6 @@ export const MODELS = {
13204
13153
  contextWindow: 1000000,
13205
13154
  maxTokens: 128000,
13206
13155
  },
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
13156
  "arcee-ai/trinity-large-thinking": {
13225
13157
  id: "arcee-ai/trinity-large-thinking",
13226
13158
  name: "Trinity Large Thinking",
@@ -13349,12 +13281,12 @@ export const MODELS = {
13349
13281
  reasoning: true,
13350
13282
  input: ["text"],
13351
13283
  cost: {
13352
- input: 0.6,
13353
- output: 1.7,
13354
- cacheRead: 0,
13284
+ input: 0.21,
13285
+ output: 0.7899999999999999,
13286
+ cacheRead: 0.13,
13355
13287
  cacheWrite: 0,
13356
13288
  },
13357
- contextWindow: 128000,
13289
+ contextWindow: 163840,
13358
13290
  maxTokens: 128000,
13359
13291
  },
13360
13292
  "deepseek/deepseek-v3.1-terminus": {
@@ -13680,6 +13612,23 @@ export const MODELS = {
13680
13612
  contextWindow: 1000000,
13681
13613
  maxTokens: 32000,
13682
13614
  },
13615
+ "kwaipilot/kat-coder-air-v2.5": {
13616
+ id: "kwaipilot/kat-coder-air-v2.5",
13617
+ name: "Kat Coder Air V2.5",
13618
+ api: "anthropic-messages",
13619
+ provider: "vercel-ai-gateway",
13620
+ baseUrl: "https://ai-gateway.vercel.sh",
13621
+ reasoning: true,
13622
+ input: ["text"],
13623
+ cost: {
13624
+ input: 0.15,
13625
+ output: 0.6,
13626
+ cacheRead: 0.03,
13627
+ cacheWrite: 0,
13628
+ },
13629
+ contextWindow: 256000,
13630
+ maxTokens: 80000,
13631
+ },
13683
13632
  "kwaipilot/kat-coder-pro-v1": {
13684
13633
  id: "kwaipilot/kat-coder-pro-v1",
13685
13634
  name: "KAT-Coder-Pro V1",
@@ -13714,39 +13663,22 @@ export const MODELS = {
13714
13663
  contextWindow: 256000,
13715
13664
  maxTokens: 256000,
13716
13665
  },
13717
- "meituan/longcat-flash-chat": {
13718
- id: "meituan/longcat-flash-chat",
13719
- name: "LongCat Flash Chat",
13720
- api: "anthropic-messages",
13721
- provider: "vercel-ai-gateway",
13722
- baseUrl: "https://ai-gateway.vercel.sh",
13723
- reasoning: false,
13724
- input: ["text"],
13725
- cost: {
13726
- input: 0,
13727
- output: 0,
13728
- cacheRead: 0,
13729
- cacheWrite: 0,
13730
- },
13731
- contextWindow: 128000,
13732
- maxTokens: 100000,
13733
- },
13734
- "meituan/longcat-flash-thinking-2601": {
13735
- id: "meituan/longcat-flash-thinking-2601",
13736
- name: "LongCat Flash Thinking 2601",
13666
+ "kwaipilot/kat-coder-pro-v2.5": {
13667
+ id: "kwaipilot/kat-coder-pro-v2.5",
13668
+ name: "Kat Coder Pro V2.5",
13737
13669
  api: "anthropic-messages",
13738
13670
  provider: "vercel-ai-gateway",
13739
13671
  baseUrl: "https://ai-gateway.vercel.sh",
13740
13672
  reasoning: true,
13741
13673
  input: ["text"],
13742
13674
  cost: {
13743
- input: 0,
13744
- output: 0,
13745
- cacheRead: 0,
13675
+ input: 0.74,
13676
+ output: 2.96,
13677
+ cacheRead: 0.15,
13746
13678
  cacheWrite: 0,
13747
13679
  },
13748
- contextWindow: 32768,
13749
- maxTokens: 32768,
13680
+ contextWindow: 256000,
13681
+ maxTokens: 80000,
13750
13682
  },
13751
13683
  "meta/llama-3.1-70b": {
13752
13684
  id: "meta/llama-3.1-70b",
@@ -14054,23 +13986,6 @@ export const MODELS = {
14054
13986
  contextWindow: 256000,
14055
13987
  maxTokens: 256000,
14056
13988
  },
14057
- "mistral/devstral-small": {
14058
- id: "mistral/devstral-small",
14059
- name: "Devstral Small 1.1",
14060
- api: "anthropic-messages",
14061
- provider: "vercel-ai-gateway",
14062
- baseUrl: "https://ai-gateway.vercel.sh",
14063
- reasoning: false,
14064
- input: ["text"],
14065
- cost: {
14066
- input: 0.09999999999999999,
14067
- output: 0.3,
14068
- cacheRead: 0,
14069
- cacheWrite: 0,
14070
- },
14071
- contextWindow: 128000,
14072
- maxTokens: 64000,
14073
- },
14074
13989
  "mistral/devstral-small-2": {
14075
13990
  id: "mistral/devstral-small-2",
14076
13991
  name: "Devstral Small 2",
@@ -14275,23 +14190,6 @@ export const MODELS = {
14275
14190
  contextWindow: 128000,
14276
14191
  maxTokens: 4000,
14277
14192
  },
14278
- "mistral/pixtral-large": {
14279
- id: "mistral/pixtral-large",
14280
- name: "Pixtral Large",
14281
- api: "anthropic-messages",
14282
- provider: "vercel-ai-gateway",
14283
- baseUrl: "https://ai-gateway.vercel.sh",
14284
- reasoning: false,
14285
- input: ["text", "image"],
14286
- cost: {
14287
- input: 2,
14288
- output: 6,
14289
- cacheRead: 0,
14290
- cacheWrite: 0,
14291
- },
14292
- contextWindow: 128000,
14293
- maxTokens: 4000,
14294
- },
14295
14193
  "moonshotai/kimi-k2": {
14296
14194
  id: "moonshotai/kimi-k2",
14297
14195
  name: "Kimi K2 Instruct",
@@ -15431,40 +15329,6 @@ export const MODELS = {
15431
15329
  contextWindow: 256000,
15432
15330
  maxTokens: 256000,
15433
15331
  },
15434
- "xiaomi/mimo-v2-flash": {
15435
- id: "xiaomi/mimo-v2-flash",
15436
- name: "MiMo V2 Flash",
15437
- api: "anthropic-messages",
15438
- provider: "vercel-ai-gateway",
15439
- baseUrl: "https://ai-gateway.vercel.sh",
15440
- reasoning: true,
15441
- input: ["text"],
15442
- cost: {
15443
- input: 0.09999999999999999,
15444
- output: 0.3,
15445
- cacheRead: 0.01,
15446
- cacheWrite: 0,
15447
- },
15448
- contextWindow: 262144,
15449
- maxTokens: 32000,
15450
- },
15451
- "xiaomi/mimo-v2-pro": {
15452
- id: "xiaomi/mimo-v2-pro",
15453
- name: "MiMo V2 Pro",
15454
- api: "anthropic-messages",
15455
- provider: "vercel-ai-gateway",
15456
- baseUrl: "https://ai-gateway.vercel.sh",
15457
- reasoning: true,
15458
- input: ["text"],
15459
- cost: {
15460
- input: 1,
15461
- output: 3,
15462
- cacheRead: 0.19999999999999998,
15463
- cacheWrite: 0,
15464
- },
15465
- contextWindow: 1000000,
15466
- maxTokens: 128000,
15467
- },
15468
15332
  "xiaomi/mimo-v2.5": {
15469
15333
  id: "xiaomi/mimo-v2.5",
15470
15334
  name: "MiMo M2.5",
@@ -15729,9 +15593,9 @@ export const MODELS = {
15729
15593
  reasoning: true,
15730
15594
  input: ["text"],
15731
15595
  cost: {
15732
- input: 3,
15733
- output: 10.25,
15734
- cacheRead: 0.5,
15596
+ input: 2.0999999999999996,
15597
+ output: 6.6000000000000005,
15598
+ cacheRead: 0.21,
15735
15599
  cacheWrite: 0,
15736
15600
  },
15737
15601
  contextWindow: 1000000,