@elyracode/ai 0.9.0 → 0.9.1

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.
@@ -391,8 +391,8 @@ export const MODELS = {
391
391
  thinkingLevelMap: { "xhigh": "max" },
392
392
  input: ["text", "image"],
393
393
  cost: {
394
- input: 5,
395
- output: 25,
394
+ input: 5.5,
395
+ output: 27.5,
396
396
  cacheRead: 0.5,
397
397
  cacheWrite: 6.25,
398
398
  },
@@ -409,10 +409,10 @@ export const MODELS = {
409
409
  thinkingLevelMap: { "xhigh": "xhigh" },
410
410
  input: ["text", "image"],
411
411
  cost: {
412
- input: 5,
413
- output: 25,
414
- cacheRead: 0.5,
415
- cacheWrite: 6.25,
412
+ input: 5.5,
413
+ output: 27.5,
414
+ cacheRead: 0.55,
415
+ cacheWrite: 6.875,
416
416
  },
417
417
  contextWindow: 1000000,
418
418
  maxTokens: 128000,
@@ -427,10 +427,10 @@ export const MODELS = {
427
427
  thinkingLevelMap: { "minimal": "low", "low": "medium", "medium": "high", "high": "max", "xhigh": "ultracode" },
428
428
  input: ["text", "image"],
429
429
  cost: {
430
- input: 5,
431
- output: 25,
432
- cacheRead: 0.5,
433
- cacheWrite: 6.25,
430
+ input: 5.5,
431
+ output: 27.5,
432
+ cacheRead: 0.55,
433
+ cacheWrite: 6.875,
434
434
  },
435
435
  contextWindow: 1000000,
436
436
  maxTokens: 128000,
@@ -444,10 +444,10 @@ export const MODELS = {
444
444
  reasoning: true,
445
445
  input: ["text", "image"],
446
446
  cost: {
447
- input: 3,
448
- output: 15,
449
- cacheRead: 0.3,
450
- cacheWrite: 3.75,
447
+ input: 3.3,
448
+ output: 16.5,
449
+ cacheRead: 0.33,
450
+ cacheWrite: 4.125,
451
451
  },
452
452
  contextWindow: 200000,
453
453
  maxTokens: 64000,
@@ -461,10 +461,10 @@ export const MODELS = {
461
461
  reasoning: true,
462
462
  input: ["text", "image"],
463
463
  cost: {
464
- input: 3,
465
- output: 15,
466
- cacheRead: 0.3,
467
- cacheWrite: 3.75,
464
+ input: 3.3,
465
+ output: 16.5,
466
+ cacheRead: 0.33,
467
+ cacheWrite: 4.125,
468
468
  },
469
469
  contextWindow: 1000000,
470
470
  maxTokens: 64000,
@@ -3321,24 +3321,6 @@ export const MODELS = {
3321
3321
  contextWindow: 128000,
3322
3322
  maxTokens: 128000,
3323
3323
  },
3324
- "@cf/moonshotai/kimi-k2.5": {
3325
- id: "@cf/moonshotai/kimi-k2.5",
3326
- name: "Kimi K2.5",
3327
- api: "openai-completions",
3328
- provider: "cloudflare-workers-ai",
3329
- baseUrl: "https://api.cloudflare.com/client/v4/accounts/{CLOUDFLARE_ACCOUNT_ID}/ai/v1",
3330
- compat: { "sendSessionAffinityHeaders": true },
3331
- reasoning: true,
3332
- input: ["text", "image"],
3333
- cost: {
3334
- input: 0.6,
3335
- output: 3,
3336
- cacheRead: 0.1,
3337
- cacheWrite: 0,
3338
- },
3339
- contextWindow: 256000,
3340
- maxTokens: 256000,
3341
- },
3342
3324
  "@cf/moonshotai/kimi-k2.6": {
3343
3325
  id: "@cf/moonshotai/kimi-k2.6",
3344
3326
  name: "Kimi K2.6",
@@ -7162,6 +7144,25 @@ export const MODELS = {
7162
7144
  contextWindow: 1000000,
7163
7145
  maxTokens: 64000,
7164
7146
  },
7147
+ "deepseek-v4-flash": {
7148
+ id: "deepseek-v4-flash",
7149
+ name: "DeepSeek V4 Flash",
7150
+ api: "openai-completions",
7151
+ provider: "opencode",
7152
+ baseUrl: "https://opencode.ai/zen/v1",
7153
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7154
+ reasoning: true,
7155
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7156
+ input: ["text"],
7157
+ cost: {
7158
+ input: 0.14,
7159
+ output: 0.28,
7160
+ cacheRead: 0.03,
7161
+ cacheWrite: 0,
7162
+ },
7163
+ contextWindow: 1000000,
7164
+ maxTokens: 384000,
7165
+ },
7165
7166
  "deepseek-v4-flash-free": {
7166
7167
  id: "deepseek-v4-flash-free",
7167
7168
  name: "DeepSeek V4 Flash Free",
@@ -7622,8 +7623,8 @@ export const MODELS = {
7622
7623
  cacheRead: 0,
7623
7624
  cacheWrite: 0,
7624
7625
  },
7625
- contextWindow: 1000000,
7626
- maxTokens: 128000,
7626
+ contextWindow: 200000,
7627
+ maxTokens: 32000,
7627
7628
  },
7628
7629
  "minimax-m2.5": {
7629
7630
  id: "minimax-m2.5",
@@ -7659,6 +7660,23 @@ export const MODELS = {
7659
7660
  contextWindow: 204800,
7660
7661
  maxTokens: 131072,
7661
7662
  },
7663
+ "minimax-m3-free": {
7664
+ id: "minimax-m3-free",
7665
+ name: "MiniMax M3 Free",
7666
+ api: "anthropic-messages",
7667
+ provider: "opencode",
7668
+ baseUrl: "https://opencode.ai/zen",
7669
+ reasoning: true,
7670
+ input: ["text", "image"],
7671
+ cost: {
7672
+ input: 0,
7673
+ output: 0,
7674
+ cacheRead: 0,
7675
+ cacheWrite: 0,
7676
+ },
7677
+ contextWindow: 200000,
7678
+ maxTokens: 32000,
7679
+ },
7662
7680
  "nemotron-3-super-free": {
7663
7681
  id: "nemotron-3-super-free",
7664
7682
  name: "Nemotron 3 Super Free",
@@ -7886,6 +7904,23 @@ export const MODELS = {
7886
7904
  contextWindow: 204800,
7887
7905
  maxTokens: 131072,
7888
7906
  },
7907
+ "minimax-m3": {
7908
+ id: "minimax-m3",
7909
+ name: "MiniMax M3",
7910
+ api: "anthropic-messages",
7911
+ provider: "opencode-go",
7912
+ baseUrl: "https://opencode.ai/zen/go",
7913
+ reasoning: true,
7914
+ input: ["text", "image"],
7915
+ cost: {
7916
+ input: 0.6,
7917
+ output: 2.4,
7918
+ cacheRead: 0.12,
7919
+ cacheWrite: 0,
7920
+ },
7921
+ contextWindow: 512000,
7922
+ maxTokens: 131072,
7923
+ },
7889
7924
  "qwen3.6-plus": {
7890
7925
  id: "qwen3.6-plus",
7891
7926
  name: "Qwen3.6 Plus",
@@ -8611,25 +8646,6 @@ export const MODELS = {
8611
8646
  contextWindow: 1048576,
8612
8647
  maxTokens: 131072,
8613
8648
  },
8614
- "deepseek/deepseek-v4-flash:free": {
8615
- id: "deepseek/deepseek-v4-flash:free",
8616
- name: "DeepSeek: DeepSeek V4 Flash (free)",
8617
- api: "openai-completions",
8618
- provider: "openrouter",
8619
- baseUrl: "https://openrouter.ai/api/v1",
8620
- compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
8621
- reasoning: true,
8622
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
8623
- input: ["text"],
8624
- cost: {
8625
- input: 0,
8626
- output: 0,
8627
- cacheRead: 0,
8628
- cacheWrite: 0,
8629
- },
8630
- contextWindow: 1048576,
8631
- maxTokens: 384000,
8632
- },
8633
8649
  "deepseek/deepseek-v4-pro": {
8634
8650
  id: "deepseek/deepseek-v4-pro",
8635
8651
  name: "DeepSeek: DeepSeek V4 Pro",
@@ -9083,9 +9099,9 @@ export const MODELS = {
9083
9099
  reasoning: true,
9084
9100
  input: ["text"],
9085
9101
  cost: {
9086
- input: 0.075,
9087
- output: 0.625,
9088
- cacheRead: 0.015,
9102
+ input: 0.3,
9103
+ output: 2.5,
9104
+ cacheRead: 0.06,
9089
9105
  cacheWrite: 0,
9090
9106
  },
9091
9107
  contextWindow: 262144,
@@ -9295,6 +9311,23 @@ export const MODELS = {
9295
9311
  contextWindow: 204800,
9296
9312
  maxTokens: 4096,
9297
9313
  },
9314
+ "minimax/minimax-m3": {
9315
+ id: "minimax/minimax-m3",
9316
+ name: "MiniMax: MiniMax M3",
9317
+ api: "openai-completions",
9318
+ provider: "openrouter",
9319
+ baseUrl: "https://openrouter.ai/api/v1",
9320
+ reasoning: true,
9321
+ input: ["text", "image"],
9322
+ cost: {
9323
+ input: 0.3,
9324
+ output: 1.2,
9325
+ cacheRead: 0.06,
9326
+ cacheWrite: 0,
9327
+ },
9328
+ contextWindow: 1048576,
9329
+ maxTokens: 512000,
9330
+ },
9298
9331
  "mistralai/codestral-2508": {
9299
9332
  id: "mistralai/codestral-2508",
9300
9333
  name: "Mistral: Codestral 2508",
@@ -11721,6 +11754,23 @@ export const MODELS = {
11721
11754
  contextWindow: 32768,
11722
11755
  maxTokens: 32768,
11723
11756
  },
11757
+ "upstage/solar-pro-3": {
11758
+ id: "upstage/solar-pro-3",
11759
+ name: "Upstage: Solar Pro 3",
11760
+ api: "openai-completions",
11761
+ provider: "openrouter",
11762
+ baseUrl: "https://openrouter.ai/api/v1",
11763
+ reasoning: true,
11764
+ input: ["text"],
11765
+ cost: {
11766
+ input: 0.15,
11767
+ output: 0.6,
11768
+ cacheRead: 0.015,
11769
+ cacheWrite: 0,
11770
+ },
11771
+ contextWindow: 128000,
11772
+ maxTokens: 4096,
11773
+ },
11724
11774
  "x-ai/grok-4.20": {
11725
11775
  id: "x-ai/grok-4.20",
11726
11776
  name: "xAI: Grok 4.20",
@@ -11974,7 +12024,7 @@ export const MODELS = {
11974
12024
  cacheWrite: 0,
11975
12025
  },
11976
12026
  contextWindow: 202752,
11977
- maxTokens: 4096,
12027
+ maxTokens: 16384,
11978
12028
  },
11979
12029
  "z-ai/glm-5-turbo": {
11980
12030
  id: "z-ai/glm-5-turbo",
@@ -12008,7 +12058,7 @@ export const MODELS = {
12008
12058
  cacheWrite: 0,
12009
12059
  },
12010
12060
  contextWindow: 202752,
12011
- maxTokens: 4096,
12061
+ maxTokens: 131072,
12012
12062
  },
12013
12063
  "z-ai/glm-5v-turbo": {
12014
12064
  id: "z-ai/glm-5v-turbo",
@@ -12529,7 +12579,7 @@ export const MODELS = {
12529
12579
  api: "anthropic-messages",
12530
12580
  provider: "vercel-ai-gateway",
12531
12581
  baseUrl: "https://ai-gateway.vercel.sh",
12532
- reasoning: false,
12582
+ reasoning: true,
12533
12583
  input: ["text"],
12534
12584
  cost: {
12535
12585
  input: 0.22,
@@ -12614,7 +12664,7 @@ export const MODELS = {
12614
12664
  api: "anthropic-messages",
12615
12665
  provider: "vercel-ai-gateway",
12616
12666
  baseUrl: "https://ai-gateway.vercel.sh",
12617
- reasoning: false,
12667
+ reasoning: true,
12618
12668
  input: ["text"],
12619
12669
  cost: {
12620
12670
  input: 1.5,
@@ -12648,7 +12698,7 @@ export const MODELS = {
12648
12698
  api: "anthropic-messages",
12649
12699
  provider: "vercel-ai-gateway",
12650
12700
  baseUrl: "https://ai-gateway.vercel.sh",
12651
- reasoning: false,
12701
+ reasoning: true,
12652
12702
  input: ["text"],
12653
12703
  cost: {
12654
12704
  input: 0.5,
@@ -12727,6 +12777,40 @@ export const MODELS = {
12727
12777
  contextWindow: 256000,
12728
12778
  maxTokens: 65536,
12729
12779
  },
12780
+ "alibaba/qwen3-next-80b-a3b-instruct": {
12781
+ id: "alibaba/qwen3-next-80b-a3b-instruct",
12782
+ name: "Qwen3 Next 80B A3B Instruct",
12783
+ api: "anthropic-messages",
12784
+ provider: "vercel-ai-gateway",
12785
+ baseUrl: "https://ai-gateway.vercel.sh",
12786
+ reasoning: false,
12787
+ input: ["text"],
12788
+ cost: {
12789
+ input: 0.15,
12790
+ output: 1.2,
12791
+ cacheRead: 0,
12792
+ cacheWrite: 0,
12793
+ },
12794
+ contextWindow: 131072,
12795
+ maxTokens: 32768,
12796
+ },
12797
+ "alibaba/qwen3-next-80b-a3b-thinking": {
12798
+ id: "alibaba/qwen3-next-80b-a3b-thinking",
12799
+ name: "Qwen3 Next 80B A3B Thinking",
12800
+ api: "anthropic-messages",
12801
+ provider: "vercel-ai-gateway",
12802
+ baseUrl: "https://ai-gateway.vercel.sh",
12803
+ reasoning: true,
12804
+ input: ["text"],
12805
+ cost: {
12806
+ input: 0.15,
12807
+ output: 1.2,
12808
+ cacheRead: 0,
12809
+ cacheWrite: 0,
12810
+ },
12811
+ contextWindow: 131072,
12812
+ maxTokens: 32768,
12813
+ },
12730
12814
  "alibaba/qwen3-vl-thinking": {
12731
12815
  id: "alibaba/qwen3-vl-thinking",
12732
12816
  name: "Qwen3 VL 235B A22B Thinking",
@@ -13113,17 +13197,17 @@ export const MODELS = {
13113
13197
  reasoning: false,
13114
13198
  input: ["text"],
13115
13199
  cost: {
13116
- input: 0.77,
13117
- output: 0.77,
13118
- cacheRead: 0,
13200
+ input: 0.27,
13201
+ output: 1.12,
13202
+ cacheRead: 0.135,
13119
13203
  cacheWrite: 0,
13120
13204
  },
13121
13205
  contextWindow: 163840,
13122
- maxTokens: 16384,
13206
+ maxTokens: 163840,
13123
13207
  },
13124
13208
  "deepseek/deepseek-v3.1": {
13125
13209
  id: "deepseek/deepseek-v3.1",
13126
- name: "DeepSeek-V3.1",
13210
+ name: "DeepSeek V3.1",
13127
13211
  api: "anthropic-messages",
13128
13212
  provider: "vercel-ai-gateway",
13129
13213
  baseUrl: "https://ai-gateway.vercel.sh",
@@ -13161,8 +13245,8 @@ export const MODELS = {
13161
13245
  api: "anthropic-messages",
13162
13246
  provider: "vercel-ai-gateway",
13163
13247
  baseUrl: "https://ai-gateway.vercel.sh",
13164
- reasoning: false,
13165
- input: ["text"],
13248
+ reasoning: true,
13249
+ input: ["text", "image"],
13166
13250
  cost: {
13167
13251
  input: 0.28,
13168
13252
  output: 0.42,
@@ -13178,8 +13262,8 @@ export const MODELS = {
13178
13262
  api: "anthropic-messages",
13179
13263
  provider: "vercel-ai-gateway",
13180
13264
  baseUrl: "https://ai-gateway.vercel.sh",
13181
- reasoning: false,
13182
- input: ["text"],
13265
+ reasoning: true,
13266
+ input: ["text", "image"],
13183
13267
  cost: {
13184
13268
  input: 0.62,
13185
13269
  output: 1.85,
@@ -13196,7 +13280,7 @@ export const MODELS = {
13196
13280
  provider: "vercel-ai-gateway",
13197
13281
  baseUrl: "https://ai-gateway.vercel.sh",
13198
13282
  reasoning: true,
13199
- input: ["text"],
13283
+ input: ["text", "image"],
13200
13284
  cost: {
13201
13285
  input: 0.14,
13202
13286
  output: 0.28,
@@ -13213,7 +13297,7 @@ export const MODELS = {
13213
13297
  provider: "vercel-ai-gateway",
13214
13298
  baseUrl: "https://ai-gateway.vercel.sh",
13215
13299
  reasoning: true,
13216
- input: ["text"],
13300
+ input: ["text", "image"],
13217
13301
  cost: {
13218
13302
  input: 0.435,
13219
13303
  output: 0.87,
@@ -13706,7 +13790,7 @@ export const MODELS = {
13706
13790
  provider: "vercel-ai-gateway",
13707
13791
  baseUrl: "https://ai-gateway.vercel.sh",
13708
13792
  reasoning: true,
13709
- input: ["text"],
13793
+ input: ["text", "image"],
13710
13794
  cost: {
13711
13795
  input: 0.6,
13712
13796
  output: 2.4,
@@ -13750,6 +13834,23 @@ export const MODELS = {
13750
13834
  contextWindow: 204800,
13751
13835
  maxTokens: 131100,
13752
13836
  },
13837
+ "minimax/minimax-m3": {
13838
+ id: "minimax/minimax-m3",
13839
+ name: "MiniMax M3",
13840
+ api: "anthropic-messages",
13841
+ provider: "vercel-ai-gateway",
13842
+ baseUrl: "https://ai-gateway.vercel.sh",
13843
+ reasoning: true,
13844
+ input: ["text", "image"],
13845
+ cost: {
13846
+ input: 0.3,
13847
+ output: 1.2,
13848
+ cacheRead: 0.06,
13849
+ cacheWrite: 0,
13850
+ },
13851
+ contextWindow: 1000000,
13852
+ maxTokens: 1000000,
13853
+ },
13753
13854
  "mistral/codestral": {
13754
13855
  id: "mistral/codestral",
13755
13856
  name: "Mistral Codestral",
@@ -13886,6 +13987,23 @@ export const MODELS = {
13886
13987
  contextWindow: 256000,
13887
13988
  maxTokens: 256000,
13888
13989
  },
13990
+ "mistral/mistral-nemo": {
13991
+ id: "mistral/mistral-nemo",
13992
+ name: "Mistral Nemo 12B",
13993
+ api: "anthropic-messages",
13994
+ provider: "vercel-ai-gateway",
13995
+ baseUrl: "https://ai-gateway.vercel.sh",
13996
+ reasoning: false,
13997
+ input: ["text"],
13998
+ cost: {
13999
+ input: 0.02,
14000
+ output: 0.04,
14001
+ cacheRead: 0,
14002
+ cacheWrite: 0,
14003
+ },
14004
+ contextWindow: 131072,
14005
+ maxTokens: 131072,
14006
+ },
13889
14007
  "mistral/mistral-small": {
13890
14008
  id: "mistral/mistral-small",
13891
14009
  name: "Mistral Small",
@@ -14039,6 +14157,23 @@ export const MODELS = {
14039
14157
  contextWindow: 262000,
14040
14158
  maxTokens: 262000,
14041
14159
  },
14160
+ "nvidia/nemotron-3-super-120b-a12b": {
14161
+ id: "nvidia/nemotron-3-super-120b-a12b",
14162
+ name: "NVIDIA Nemotron 3 Super 120B A12B",
14163
+ api: "anthropic-messages",
14164
+ provider: "vercel-ai-gateway",
14165
+ baseUrl: "https://ai-gateway.vercel.sh",
14166
+ reasoning: true,
14167
+ input: ["text"],
14168
+ cost: {
14169
+ input: 0.15,
14170
+ output: 0.65,
14171
+ cacheRead: 0,
14172
+ cacheWrite: 0,
14173
+ },
14174
+ contextWindow: 256000,
14175
+ maxTokens: 32000,
14176
+ },
14042
14177
  "nvidia/nemotron-nano-12b-v2-vl": {
14043
14178
  id: "nvidia/nemotron-nano-12b-v2-vl",
14044
14179
  name: "Nvidia Nemotron Nano 12B V2 VL",
@@ -14199,7 +14334,7 @@ export const MODELS = {
14199
14334
  provider: "vercel-ai-gateway",
14200
14335
  baseUrl: "https://ai-gateway.vercel.sh",
14201
14336
  reasoning: true,
14202
- input: ["text"],
14337
+ input: ["text", "image"],
14203
14338
  cost: {
14204
14339
  input: 1.25,
14205
14340
  output: 10,
@@ -14561,6 +14696,23 @@ export const MODELS = {
14561
14696
  contextWindow: 1000000,
14562
14697
  maxTokens: 128000,
14563
14698
  },
14699
+ "openai/gpt-oss-120b": {
14700
+ id: "openai/gpt-oss-120b",
14701
+ name: "GPT OSS 120B",
14702
+ api: "anthropic-messages",
14703
+ provider: "vercel-ai-gateway",
14704
+ baseUrl: "https://ai-gateway.vercel.sh",
14705
+ reasoning: true,
14706
+ input: ["text"],
14707
+ cost: {
14708
+ input: 0.35,
14709
+ output: 0.75,
14710
+ cacheRead: 0.25,
14711
+ cacheWrite: 0,
14712
+ },
14713
+ contextWindow: 131072,
14714
+ maxTokens: 131000,
14715
+ },
14564
14716
  "openai/gpt-oss-20b": {
14565
14717
  id: "openai/gpt-oss-20b",
14566
14718
  name: "GPT OSS 20B",
@@ -14731,6 +14883,23 @@ export const MODELS = {
14731
14883
  contextWindow: 200000,
14732
14884
  maxTokens: 8000,
14733
14885
  },
14886
+ "stepfun/step-3.5-flash": {
14887
+ id: "stepfun/step-3.5-flash",
14888
+ name: "StepFun 3.5 Flash",
14889
+ api: "anthropic-messages",
14890
+ provider: "vercel-ai-gateway",
14891
+ baseUrl: "https://ai-gateway.vercel.sh",
14892
+ reasoning: true,
14893
+ input: ["text"],
14894
+ cost: {
14895
+ input: 0.09,
14896
+ output: 0.3,
14897
+ cacheRead: 0,
14898
+ cacheWrite: 0.02,
14899
+ },
14900
+ contextWindow: 262114,
14901
+ maxTokens: 262114,
14902
+ },
14734
14903
  "stepfun/step-3.7-flash": {
14735
14904
  id: "stepfun/step-3.7-flash",
14736
14905
  name: "Step 3.7 Flash",
@@ -14992,7 +15161,7 @@ export const MODELS = {
14992
15161
  api: "anthropic-messages",
14993
15162
  provider: "vercel-ai-gateway",
14994
15163
  baseUrl: "https://ai-gateway.vercel.sh",
14995
- reasoning: false,
15164
+ reasoning: true,
14996
15165
  input: ["text", "image"],
14997
15166
  cost: {
14998
15167
  input: 0.6,
@@ -15146,7 +15315,7 @@ export const MODELS = {
15146
15315
  provider: "vercel-ai-gateway",
15147
15316
  baseUrl: "https://ai-gateway.vercel.sh",
15148
15317
  reasoning: true,
15149
- input: ["text"],
15318
+ input: ["text", "image"],
15150
15319
  cost: {
15151
15320
  input: 1.4,
15152
15321
  output: 4.4,