@dreb/ai 2.43.0 → 2.43.3

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.
@@ -3368,7 +3368,7 @@ export const MODELS = {
3368
3368
  input: 1,
3369
3369
  output: 6,
3370
3370
  cacheRead: 0.1,
3371
- cacheWrite: 0,
3371
+ cacheWrite: 1.25,
3372
3372
  },
3373
3373
  contextWindow: 1050000,
3374
3374
  maxTokens: 128000,
@@ -3386,7 +3386,7 @@ export const MODELS = {
3386
3386
  input: 5,
3387
3387
  output: 30,
3388
3388
  cacheRead: 0.5,
3389
- cacheWrite: 0,
3389
+ cacheWrite: 6.25,
3390
3390
  },
3391
3391
  contextWindow: 1050000,
3392
3392
  maxTokens: 128000,
@@ -3404,7 +3404,7 @@ export const MODELS = {
3404
3404
  input: 2.5,
3405
3405
  output: 15,
3406
3406
  cacheRead: 0.25,
3407
- cacheWrite: 0,
3407
+ cacheWrite: 3.125,
3408
3408
  },
3409
3409
  contextWindow: 1050000,
3410
3410
  maxTokens: 128000,
@@ -5280,13 +5280,49 @@ export const MODELS = {
5280
5280
  },
5281
5281
  },
5282
5282
  "kimi-coding-oauth": {
5283
+ "k3": {
5284
+ id: "k3",
5285
+ name: "Kimi K3",
5286
+ api: "openai-completions",
5287
+ provider: "kimi-coding-oauth",
5288
+ baseUrl: "https://api.kimi.com/coding/v1",
5289
+ compat: { "thinkingFormat": "kimi", "supportsDeveloperRole": false, "reasoningEffortMap": { "minimal": "low", "low": "low", "medium": "high", "high": "high", "xhigh": "max" } },
5290
+ reasoning: true,
5291
+ input: ["text", "image"],
5292
+ cost: {
5293
+ input: 0,
5294
+ output: 0,
5295
+ cacheRead: 0,
5296
+ cacheWrite: 0,
5297
+ },
5298
+ contextWindow: 1048576,
5299
+ maxTokens: 32768,
5300
+ },
5283
5301
  "kimi-for-coding": {
5284
5302
  id: "kimi-for-coding",
5285
5303
  name: "Kimi For Coding",
5286
5304
  api: "openai-completions",
5287
5305
  provider: "kimi-coding-oauth",
5288
5306
  baseUrl: "https://api.kimi.com/coding/v1",
5289
- compat: { "thinkingFormat": "kimi", "supportsDeveloperRole": false },
5307
+ compat: { "thinkingFormat": "kimi", "supportsDeveloperRole": false, "reasoningEffortMap": { "minimal": "auto", "low": "auto", "medium": "auto", "high": "auto", "xhigh": "auto" } },
5308
+ reasoning: true,
5309
+ input: ["text", "image"],
5310
+ cost: {
5311
+ input: 0,
5312
+ output: 0,
5313
+ cacheRead: 0,
5314
+ cacheWrite: 0,
5315
+ },
5316
+ contextWindow: 262144,
5317
+ maxTokens: 32768,
5318
+ },
5319
+ "kimi-for-coding-highspeed": {
5320
+ id: "kimi-for-coding-highspeed",
5321
+ name: "Kimi For Coding Highspeed",
5322
+ api: "openai-completions",
5323
+ provider: "kimi-coding-oauth",
5324
+ baseUrl: "https://api.kimi.com/coding/v1",
5325
+ compat: { "thinkingFormat": "kimi", "supportsDeveloperRole": false, "reasoningEffortMap": { "minimal": "auto", "low": "auto", "medium": "auto", "high": "auto", "xhigh": "auto" } },
5290
5326
  reasoning: true,
5291
5327
  input: ["text", "image"],
5292
5328
  cost: {
@@ -7540,7 +7576,7 @@ export const MODELS = {
7540
7576
  "grok-4.5": {
7541
7577
  id: "grok-4.5",
7542
7578
  name: "Grok 4.5",
7543
- api: "openai-completions",
7579
+ api: "openai-responses",
7544
7580
  provider: "opencode",
7545
7581
  baseUrl: "https://opencode.ai/zen/v1",
7546
7582
  reasoning: true,
@@ -7786,9 +7822,9 @@ export const MODELS = {
7786
7822
  reasoning: true,
7787
7823
  input: ["text"],
7788
7824
  cost: {
7789
- input: 0.435,
7790
- output: 0.87,
7791
- cacheRead: 0.003625,
7825
+ input: 0.14,
7826
+ output: 0.28,
7827
+ cacheRead: 0.0028,
7792
7828
  cacheWrite: 0,
7793
7829
  },
7794
7830
  contextWindow: 1000000,
@@ -7848,7 +7884,7 @@ export const MODELS = {
7848
7884
  "grok-4.5": {
7849
7885
  id: "grok-4.5",
7850
7886
  name: "Grok 4.5",
7851
- api: "openai-completions",
7887
+ api: "openai-responses",
7852
7888
  provider: "opencode-go",
7853
7889
  baseUrl: "https://opencode.ai/zen/go/v1",
7854
7890
  reasoning: true,
@@ -7898,7 +7934,7 @@ export const MODELS = {
7898
7934
  },
7899
7935
  "kimi-k3": {
7900
7936
  id: "kimi-k3",
7901
- name: "Kimi K3",
7937
+ name: "Kimi K3 (2x usage)",
7902
7938
  api: "openai-completions",
7903
7939
  provider: "opencode-go",
7904
7940
  baseUrl: "https://opencode.ai/zen/go/v1",
@@ -9029,13 +9065,13 @@ export const MODELS = {
9029
9065
  reasoning: true,
9030
9066
  input: ["text", "image"],
9031
9067
  cost: {
9032
- input: 0.09999999999999999,
9033
- output: 0.3,
9068
+ input: 0.07,
9069
+ output: 0.33999999999999997,
9034
9070
  cacheRead: 0,
9035
9071
  cacheWrite: 0,
9036
9072
  },
9037
9073
  contextWindow: 262144,
9038
- maxTokens: 256000,
9074
+ maxTokens: 16384,
9039
9075
  },
9040
9076
  "google/gemma-4-26b-a4b-it:free": {
9041
9077
  id: "google/gemma-4-26b-a4b-it:free",
@@ -9063,13 +9099,13 @@ export const MODELS = {
9063
9099
  reasoning: true,
9064
9100
  input: ["text", "image"],
9065
9101
  cost: {
9066
- input: 0.12,
9067
- output: 0.37,
9068
- cacheRead: 0,
9102
+ input: 0.22,
9103
+ output: 0.55,
9104
+ cacheRead: 0.12,
9069
9105
  cacheWrite: 0,
9070
9106
  },
9071
9107
  contextWindow: 262144,
9072
- maxTokens: 16384,
9108
+ maxTokens: 262144,
9073
9109
  },
9074
9110
  "google/gemma-4-31b-it:free": {
9075
9111
  id: "google/gemma-4-31b-it:free",
@@ -9275,23 +9311,6 @@ export const MODELS = {
9275
9311
  contextWindow: 131072,
9276
9312
  maxTokens: 128000,
9277
9313
  },
9278
- "meta-llama/llama-3.3-70b-instruct:free": {
9279
- id: "meta-llama/llama-3.3-70b-instruct:free",
9280
- name: "Meta: Llama 3.3 70B Instruct (free)",
9281
- api: "openai-completions",
9282
- provider: "openrouter",
9283
- baseUrl: "https://openrouter.ai/api/v1",
9284
- reasoning: false,
9285
- input: ["text"],
9286
- cost: {
9287
- input: 0,
9288
- output: 0,
9289
- cacheRead: 0,
9290
- cacheWrite: 0,
9291
- },
9292
- contextWindow: 131072,
9293
- maxTokens: 4096,
9294
- },
9295
9314
  "meta-llama/llama-4-maverick": {
9296
9315
  id: "meta-llama/llama-4-maverick",
9297
9316
  name: "Meta: Llama 4 Maverick",
@@ -9369,9 +9388,9 @@ export const MODELS = {
9369
9388
  reasoning: true,
9370
9389
  input: ["text"],
9371
9390
  cost: {
9372
- input: 0.255,
9373
- output: 1.02,
9374
- cacheRead: 0,
9391
+ input: 0.3,
9392
+ output: 1.2,
9393
+ cacheRead: 0.03,
9375
9394
  cacheWrite: 0,
9376
9395
  },
9377
9396
  contextWindow: 204800,
@@ -9779,11 +9798,11 @@ export const MODELS = {
9779
9798
  cost: {
9780
9799
  input: 0.6,
9781
9800
  output: 2.5,
9782
- cacheRead: 0,
9801
+ cacheRead: 0.15,
9783
9802
  cacheWrite: 0,
9784
9803
  },
9785
9804
  contextWindow: 262144,
9786
- maxTokens: 262144,
9805
+ maxTokens: 100352,
9787
9806
  },
9788
9807
  "moonshotai/kimi-k2.5": {
9789
9808
  id: "moonshotai/kimi-k2.5",
@@ -9811,13 +9830,13 @@ export const MODELS = {
9811
9830
  reasoning: true,
9812
9831
  input: ["text", "image"],
9813
9832
  cost: {
9814
- input: 0.95,
9815
- output: 4,
9816
- cacheRead: 0.16,
9833
+ input: 0.684,
9834
+ output: 3.42,
9835
+ cacheRead: 0.144,
9817
9836
  cacheWrite: 0,
9818
9837
  },
9819
9838
  contextWindow: 262144,
9820
- maxTokens: 4096,
9839
+ maxTokens: 262144,
9821
9840
  },
9822
9841
  "moonshotai/kimi-k2.7-code": {
9823
9842
  id: "moonshotai/kimi-k2.7-code",
@@ -9828,9 +9847,9 @@ export const MODELS = {
9828
9847
  reasoning: true,
9829
9848
  input: ["text", "image"],
9830
9849
  cost: {
9831
- input: 0.75,
9832
- output: 3.5,
9833
- cacheRead: 0.16,
9850
+ input: 0.85,
9851
+ output: 3.8,
9852
+ cacheRead: 0.16999999999999998,
9834
9853
  cacheWrite: 0,
9835
9854
  },
9836
9855
  contextWindow: 262144,
@@ -9947,13 +9966,13 @@ export const MODELS = {
9947
9966
  reasoning: true,
9948
9967
  input: ["text"],
9949
9968
  cost: {
9950
- input: 0.21,
9951
- output: 0.45499999999999996,
9952
- cacheRead: 0.06,
9969
+ input: 0.08499999999999999,
9970
+ output: 0.39999999999999997,
9971
+ cacheRead: 0,
9953
9972
  cacheWrite: 0,
9954
9973
  },
9955
9974
  contextWindow: 1000000,
9956
- maxTokens: 4096,
9975
+ maxTokens: 16384,
9957
9976
  },
9958
9977
  "nvidia/nemotron-3-super-120b-a12b:free": {
9959
9978
  id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -11324,13 +11343,13 @@ export const MODELS = {
11324
11343
  reasoning: true,
11325
11344
  input: ["text"],
11326
11345
  cost: {
11327
- input: 0.12,
11328
- output: 0.5,
11346
+ input: 0.13,
11347
+ output: 0.52,
11329
11348
  cacheRead: 0,
11330
11349
  cacheWrite: 0,
11331
11350
  },
11332
11351
  contextWindow: 131072,
11333
- maxTokens: 16384,
11352
+ maxTokens: 8192,
11334
11353
  },
11335
11354
  "qwen/qwen3-30b-a3b-instruct-2507": {
11336
11355
  id: "qwen/qwen3-30b-a3b-instruct-2507",
@@ -11485,23 +11504,6 @@ export const MODELS = {
11485
11504
  contextWindow: 1000000,
11486
11505
  maxTokens: 65536,
11487
11506
  },
11488
- "qwen/qwen3-coder:free": {
11489
- id: "qwen/qwen3-coder:free",
11490
- name: "Qwen: Qwen3 Coder 480B A35B (free)",
11491
- api: "openai-completions",
11492
- provider: "openrouter",
11493
- baseUrl: "https://openrouter.ai/api/v1",
11494
- reasoning: false,
11495
- input: ["text"],
11496
- cost: {
11497
- input: 0,
11498
- output: 0,
11499
- cacheRead: 0,
11500
- cacheWrite: 0,
11501
- },
11502
- contextWindow: 1048576,
11503
- maxTokens: 262000,
11504
- },
11505
11507
  "qwen/qwen3-max": {
11506
11508
  id: "qwen/qwen3-max",
11507
11509
  name: "Qwen: Qwen3 Max",
@@ -11553,23 +11555,6 @@ export const MODELS = {
11553
11555
  contextWindow: 262144,
11554
11556
  maxTokens: 262144,
11555
11557
  },
11556
- "qwen/qwen3-next-80b-a3b-instruct:free": {
11557
- id: "qwen/qwen3-next-80b-a3b-instruct:free",
11558
- name: "Qwen: Qwen3 Next 80B A3B Instruct (free)",
11559
- api: "openai-completions",
11560
- provider: "openrouter",
11561
- baseUrl: "https://openrouter.ai/api/v1",
11562
- reasoning: false,
11563
- input: ["text"],
11564
- cost: {
11565
- input: 0,
11566
- output: 0,
11567
- cacheRead: 0,
11568
- cacheWrite: 0,
11569
- },
11570
- contextWindow: 262144,
11571
- maxTokens: 4096,
11572
- },
11573
11558
  "qwen/qwen3-next-80b-a3b-thinking": {
11574
11559
  id: "qwen/qwen3-next-80b-a3b-thinking",
11575
11560
  name: "Qwen: Qwen3 Next 80B A3B Thinking",
@@ -11732,13 +11717,13 @@ export const MODELS = {
11732
11717
  reasoning: true,
11733
11718
  input: ["text", "image"],
11734
11719
  cost: {
11735
- input: 0.195,
11736
- output: 1.56,
11720
+ input: 0.26,
11721
+ output: 2.6,
11737
11722
  cacheRead: 0,
11738
11723
  cacheWrite: 0,
11739
11724
  },
11740
11725
  contextWindow: 262144,
11741
- maxTokens: 65536,
11726
+ maxTokens: 81920,
11742
11727
  },
11743
11728
  "qwen/qwen3.5-35b-a3b": {
11744
11729
  id: "qwen/qwen3.5-35b-a3b",
@@ -12131,6 +12116,23 @@ export const MODELS = {
12131
12116
  contextWindow: 32768,
12132
12117
  maxTokens: 32768,
12133
12118
  },
12119
+ "thinkingmachines/inkling": {
12120
+ id: "thinkingmachines/inkling",
12121
+ name: "Thinking Machines: Inkling",
12122
+ api: "openai-completions",
12123
+ provider: "openrouter",
12124
+ baseUrl: "https://openrouter.ai/api/v1",
12125
+ reasoning: true,
12126
+ input: ["text", "image"],
12127
+ cost: {
12128
+ input: 1,
12129
+ output: 4.05,
12130
+ cacheRead: 0.16999999999999998,
12131
+ cacheWrite: 0,
12132
+ },
12133
+ contextWindow: 1048576,
12134
+ maxTokens: 4096,
12135
+ },
12134
12136
  "upstage/solar-pro-3": {
12135
12137
  id: "upstage/solar-pro-3",
12136
12138
  name: "Upstage: Solar Pro 3",
@@ -12193,7 +12195,7 @@ export const MODELS = {
12193
12195
  cost: {
12194
12196
  input: 2,
12195
12197
  output: 6,
12196
- cacheRead: 0.5,
12198
+ cacheRead: 0.3,
12197
12199
  cacheWrite: 0,
12198
12200
  },
12199
12201
  contextWindow: 500000,
@@ -12429,9 +12431,9 @@ export const MODELS = {
12429
12431
  reasoning: true,
12430
12432
  input: ["text"],
12431
12433
  cost: {
12432
- input: 0.9212,
12433
- output: 2.8952,
12434
- cacheRead: 0.17107999999999998,
12434
+ input: 0.252,
12435
+ output: 0.792,
12436
+ cacheRead: 0.0468,
12435
12437
  cacheWrite: 0,
12436
12438
  },
12437
12439
  contextWindow: 1048576,
@@ -12618,7 +12620,7 @@ export const MODELS = {
12618
12620
  cost: {
12619
12621
  input: 2,
12620
12622
  output: 6,
12621
- cacheRead: 0.5,
12623
+ cacheRead: 0.3,
12622
12624
  cacheWrite: 0,
12623
12625
  },
12624
12626
  contextWindow: 500000,
@@ -15886,7 +15888,7 @@ export const MODELS = {
15886
15888
  cost: {
15887
15889
  input: 2,
15888
15890
  output: 6,
15889
- cacheRead: 0.5,
15891
+ cacheRead: 0.3,
15890
15892
  cacheWrite: 0,
15891
15893
  },
15892
15894
  contextWindow: 500000,