@caupulican/pi-ai 0.80.20 → 0.80.22

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.
@@ -2854,30 +2854,13 @@ export const MODELS = {
2854
2854
  reasoning: true,
2855
2855
  input: ["text"],
2856
2856
  cost: {
2857
- input: 0.25,
2858
- output: 0.69,
2857
+ input: 0.35,
2858
+ output: 0.75,
2859
2859
  cacheRead: 0,
2860
2860
  cacheWrite: 0,
2861
2861
  },
2862
2862
  contextWindow: 131072,
2863
- maxTokens: 32768,
2864
- },
2865
- "llama3.1-8b": {
2866
- id: "llama3.1-8b",
2867
- name: "Llama 3.1 8B",
2868
- api: "openai-completions",
2869
- provider: "cerebras",
2870
- baseUrl: "https://api.cerebras.ai/v1",
2871
- reasoning: false,
2872
- input: ["text"],
2873
- cost: {
2874
- input: 0.1,
2875
- output: 0.1,
2876
- cacheRead: 0,
2877
- cacheWrite: 0,
2878
- },
2879
- contextWindow: 32000,
2880
- maxTokens: 8000,
2863
+ maxTokens: 40960,
2881
2864
  },
2882
2865
  "zai-glm-4.7": {
2883
2866
  id: "zai-glm-4.7",
@@ -2885,7 +2868,7 @@ export const MODELS = {
2885
2868
  api: "openai-completions",
2886
2869
  provider: "cerebras",
2887
2870
  baseUrl: "https://api.cerebras.ai/v1",
2888
- reasoning: false,
2871
+ reasoning: true,
2889
2872
  input: ["text"],
2890
2873
  cost: {
2891
2874
  input: 2.25,
@@ -2894,7 +2877,7 @@ export const MODELS = {
2894
2877
  cacheWrite: 0,
2895
2878
  },
2896
2879
  contextWindow: 131072,
2897
- maxTokens: 40000,
2880
+ maxTokens: 40960,
2898
2881
  },
2899
2882
  },
2900
2883
  "cloudflare-ai-gateway": {
@@ -4023,6 +4006,25 @@ export const MODELS = {
4023
4006
  },
4024
4007
  },
4025
4008
  "github-copilot": {
4009
+ "claude-fable-5": {
4010
+ id: "claude-fable-5",
4011
+ name: "Claude Fable 5",
4012
+ api: "openai-completions",
4013
+ provider: "github-copilot",
4014
+ baseUrl: "https://api.individual.githubcopilot.com",
4015
+ headers: { "User-Agent": "GitHubCopilotChat/0.35.0", "Editor-Version": "vscode/1.107.0", "Editor-Plugin-Version": "copilot-chat/0.35.0", "Copilot-Integration-Id": "vscode-chat" },
4016
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false },
4017
+ reasoning: true,
4018
+ input: ["text", "image"],
4019
+ cost: {
4020
+ input: 10,
4021
+ output: 50,
4022
+ cacheRead: 1,
4023
+ cacheWrite: 12.5,
4024
+ },
4025
+ contextWindow: 1000000,
4026
+ maxTokens: 128000,
4027
+ },
4026
4028
  "claude-haiku-4.5": {
4027
4029
  id: "claude-haiku-4.5",
4028
4030
  name: "Claude Haiku 4.5 (latest)",
@@ -4955,77 +4957,9 @@ export const MODELS = {
4955
4957
  },
4956
4958
  },
4957
4959
  "groq": {
4958
- "deepseek-r1-distill-llama-70b": {
4959
- id: "deepseek-r1-distill-llama-70b",
4960
- name: "DeepSeek R1 Distill Llama 70B",
4961
- api: "openai-completions",
4962
- provider: "groq",
4963
- baseUrl: "https://api.groq.com/openai/v1",
4964
- reasoning: true,
4965
- input: ["text"],
4966
- cost: {
4967
- input: 0.75,
4968
- output: 0.99,
4969
- cacheRead: 0,
4970
- cacheWrite: 0,
4971
- },
4972
- contextWindow: 131072,
4973
- maxTokens: 8192,
4974
- },
4975
- "gemma2-9b-it": {
4976
- id: "gemma2-9b-it",
4977
- name: "Gemma 2 9B",
4978
- api: "openai-completions",
4979
- provider: "groq",
4980
- baseUrl: "https://api.groq.com/openai/v1",
4981
- reasoning: false,
4982
- input: ["text"],
4983
- cost: {
4984
- input: 0.2,
4985
- output: 0.2,
4986
- cacheRead: 0,
4987
- cacheWrite: 0,
4988
- },
4989
- contextWindow: 8192,
4990
- maxTokens: 8192,
4991
- },
4992
- "groq/compound": {
4993
- id: "groq/compound",
4994
- name: "Compound",
4995
- api: "openai-completions",
4996
- provider: "groq",
4997
- baseUrl: "https://api.groq.com/openai/v1",
4998
- reasoning: true,
4999
- input: ["text"],
5000
- cost: {
5001
- input: 0,
5002
- output: 0,
5003
- cacheRead: 0,
5004
- cacheWrite: 0,
5005
- },
5006
- contextWindow: 131072,
5007
- maxTokens: 8192,
5008
- },
5009
- "groq/compound-mini": {
5010
- id: "groq/compound-mini",
5011
- name: "Compound Mini",
5012
- api: "openai-completions",
5013
- provider: "groq",
5014
- baseUrl: "https://api.groq.com/openai/v1",
5015
- reasoning: true,
5016
- input: ["text"],
5017
- cost: {
5018
- input: 0,
5019
- output: 0,
5020
- cacheRead: 0,
5021
- cacheWrite: 0,
5022
- },
5023
- contextWindow: 131072,
5024
- maxTokens: 8192,
5025
- },
5026
4960
  "llama-3.1-8b-instant": {
5027
4961
  id: "llama-3.1-8b-instant",
5028
- name: "Llama 3.1 8B Instant",
4962
+ name: "Llama 3.1 8B",
5029
4963
  api: "openai-completions",
5030
4964
  provider: "groq",
5031
4965
  baseUrl: "https://api.groq.com/openai/v1",
@@ -5042,7 +4976,7 @@ export const MODELS = {
5042
4976
  },
5043
4977
  "llama-3.3-70b-versatile": {
5044
4978
  id: "llama-3.3-70b-versatile",
5045
- name: "Llama 3.3 70B Versatile",
4979
+ name: "Llama 3.3 70B",
5046
4980
  api: "openai-completions",
5047
4981
  provider: "groq",
5048
4982
  baseUrl: "https://api.groq.com/openai/v1",
@@ -5057,60 +4991,9 @@ export const MODELS = {
5057
4991
  contextWindow: 131072,
5058
4992
  maxTokens: 32768,
5059
4993
  },
5060
- "llama3-70b-8192": {
5061
- id: "llama3-70b-8192",
5062
- name: "Llama 3 70B",
5063
- api: "openai-completions",
5064
- provider: "groq",
5065
- baseUrl: "https://api.groq.com/openai/v1",
5066
- reasoning: false,
5067
- input: ["text"],
5068
- cost: {
5069
- input: 0.59,
5070
- output: 0.79,
5071
- cacheRead: 0,
5072
- cacheWrite: 0,
5073
- },
5074
- contextWindow: 8192,
5075
- maxTokens: 8192,
5076
- },
5077
- "llama3-8b-8192": {
5078
- id: "llama3-8b-8192",
5079
- name: "Llama 3 8B",
5080
- api: "openai-completions",
5081
- provider: "groq",
5082
- baseUrl: "https://api.groq.com/openai/v1",
5083
- reasoning: false,
5084
- input: ["text"],
5085
- cost: {
5086
- input: 0.05,
5087
- output: 0.08,
5088
- cacheRead: 0,
5089
- cacheWrite: 0,
5090
- },
5091
- contextWindow: 8192,
5092
- maxTokens: 8192,
5093
- },
5094
- "meta-llama/llama-4-maverick-17b-128e-instruct": {
5095
- id: "meta-llama/llama-4-maverick-17b-128e-instruct",
5096
- name: "Llama 4 Maverick 17B",
5097
- api: "openai-completions",
5098
- provider: "groq",
5099
- baseUrl: "https://api.groq.com/openai/v1",
5100
- reasoning: false,
5101
- input: ["text", "image"],
5102
- cost: {
5103
- input: 0.2,
5104
- output: 0.6,
5105
- cacheRead: 0,
5106
- cacheWrite: 0,
5107
- },
5108
- contextWindow: 131072,
5109
- maxTokens: 8192,
5110
- },
5111
4994
  "meta-llama/llama-4-scout-17b-16e-instruct": {
5112
4995
  id: "meta-llama/llama-4-scout-17b-16e-instruct",
5113
- name: "Llama 4 Scout 17B",
4996
+ name: "Llama 4 Scout 17B 16E",
5114
4997
  api: "openai-completions",
5115
4998
  provider: "groq",
5116
4999
  baseUrl: "https://api.groq.com/openai/v1",
@@ -5125,57 +5008,6 @@ export const MODELS = {
5125
5008
  contextWindow: 131072,
5126
5009
  maxTokens: 8192,
5127
5010
  },
5128
- "mistral-saba-24b": {
5129
- id: "mistral-saba-24b",
5130
- name: "Mistral Saba 24B",
5131
- api: "openai-completions",
5132
- provider: "groq",
5133
- baseUrl: "https://api.groq.com/openai/v1",
5134
- reasoning: false,
5135
- input: ["text"],
5136
- cost: {
5137
- input: 0.79,
5138
- output: 0.79,
5139
- cacheRead: 0,
5140
- cacheWrite: 0,
5141
- },
5142
- contextWindow: 32768,
5143
- maxTokens: 32768,
5144
- },
5145
- "moonshotai/kimi-k2-instruct": {
5146
- id: "moonshotai/kimi-k2-instruct",
5147
- name: "Kimi K2 Instruct",
5148
- api: "openai-completions",
5149
- provider: "groq",
5150
- baseUrl: "https://api.groq.com/openai/v1",
5151
- reasoning: false,
5152
- input: ["text"],
5153
- cost: {
5154
- input: 1,
5155
- output: 3,
5156
- cacheRead: 0,
5157
- cacheWrite: 0,
5158
- },
5159
- contextWindow: 131072,
5160
- maxTokens: 16384,
5161
- },
5162
- "moonshotai/kimi-k2-instruct-0905": {
5163
- id: "moonshotai/kimi-k2-instruct-0905",
5164
- name: "Kimi K2 Instruct 0905",
5165
- api: "openai-completions",
5166
- provider: "groq",
5167
- baseUrl: "https://api.groq.com/openai/v1",
5168
- reasoning: false,
5169
- input: ["text"],
5170
- cost: {
5171
- input: 1,
5172
- output: 3,
5173
- cacheRead: 0.5,
5174
- cacheWrite: 0,
5175
- },
5176
- contextWindow: 262144,
5177
- maxTokens: 16384,
5178
- },
5179
5011
  "openai/gpt-oss-120b": {
5180
5012
  id: "openai/gpt-oss-120b",
5181
5013
  name: "GPT OSS 120B",
@@ -5227,26 +5059,9 @@ export const MODELS = {
5227
5059
  contextWindow: 131072,
5228
5060
  maxTokens: 65536,
5229
5061
  },
5230
- "qwen-qwq-32b": {
5231
- id: "qwen-qwq-32b",
5232
- name: "Qwen QwQ 32B",
5233
- api: "openai-completions",
5234
- provider: "groq",
5235
- baseUrl: "https://api.groq.com/openai/v1",
5236
- reasoning: true,
5237
- input: ["text"],
5238
- cost: {
5239
- input: 0.29,
5240
- output: 0.39,
5241
- cacheRead: 0,
5242
- cacheWrite: 0,
5243
- },
5244
- contextWindow: 131072,
5245
- maxTokens: 16384,
5246
- },
5247
5062
  "qwen/qwen3-32b": {
5248
5063
  id: "qwen/qwen3-32b",
5249
- name: "Qwen3 32B",
5064
+ name: "Qwen3-32B",
5250
5065
  api: "openai-completions",
5251
5066
  provider: "groq",
5252
5067
  baseUrl: "https://api.groq.com/openai/v1",
@@ -7617,7 +7432,7 @@ export const MODELS = {
7617
7432
  cost: {
7618
7433
  input: 0.14,
7619
7434
  output: 0.28,
7620
- cacheRead: 0.03,
7435
+ cacheRead: 0.028,
7621
7436
  cacheWrite: 0,
7622
7437
  },
7623
7438
  contextWindow: 1000000,
@@ -7642,6 +7457,25 @@ export const MODELS = {
7642
7457
  contextWindow: 200000,
7643
7458
  maxTokens: 128000,
7644
7459
  },
7460
+ "deepseek-v4-pro": {
7461
+ id: "deepseek-v4-pro",
7462
+ name: "DeepSeek V4 Pro",
7463
+ api: "openai-completions",
7464
+ provider: "opencode",
7465
+ baseUrl: "https://opencode.ai/zen/v1",
7466
+ compat: { "requiresReasoningContentOnAssistantMessages": true, "thinkingFormat": "deepseek" },
7467
+ reasoning: true,
7468
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "max" },
7469
+ input: ["text"],
7470
+ cost: {
7471
+ input: 1.74,
7472
+ output: 3.84,
7473
+ cacheRead: 0.145,
7474
+ cacheWrite: 0,
7475
+ },
7476
+ contextWindow: 1000000,
7477
+ maxTokens: 384000,
7478
+ },
7645
7479
  "gemini-3-flash": {
7646
7480
  id: "gemini-3-flash",
7647
7481
  name: "Gemini 3 Flash",
@@ -8371,16 +8205,16 @@ export const MODELS = {
8371
8205
  },
8372
8206
  "minimax-m3": {
8373
8207
  id: "minimax-m3",
8374
- name: "MiniMax M3",
8208
+ name: "MiniMax M3 (3x usage)",
8375
8209
  api: "anthropic-messages",
8376
8210
  provider: "opencode-go",
8377
8211
  baseUrl: "https://opencode.ai/zen/go",
8378
8212
  reasoning: true,
8379
8213
  input: ["text", "image"],
8380
8214
  cost: {
8381
- input: 0.3,
8382
- output: 1.2,
8383
- cacheRead: 0.06,
8215
+ input: 0.1,
8216
+ output: 0.4,
8217
+ cacheRead: 0.02,
8384
8218
  cacheWrite: 0,
8385
8219
  },
8386
8220
  contextWindow: 512000,
@@ -9021,7 +8855,7 @@ export const MODELS = {
9021
8855
  cacheRead: 0.135,
9022
8856
  cacheWrite: 0,
9023
8857
  },
9024
- contextWindow: 163840,
8858
+ contextWindow: 131072,
9025
8859
  maxTokens: 16384,
9026
8860
  },
9027
8861
  "deepseek/deepseek-chat-v3.1": {
@@ -9137,13 +8971,13 @@ export const MODELS = {
9137
8971
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": "xhigh" },
9138
8972
  input: ["text"],
9139
8973
  cost: {
9140
- input: 0.0983,
9141
- output: 0.1966,
9142
- cacheRead: 0.019700000000000002,
8974
+ input: 0.098,
8975
+ output: 0.196,
8976
+ cacheRead: 0.02,
9143
8977
  cacheWrite: 0,
9144
8978
  },
9145
8979
  contextWindow: 1048576,
9146
- maxTokens: 131072,
8980
+ maxTokens: 4096,
9147
8981
  },
9148
8982
  "deepseek/deepseek-v4-pro": {
9149
8983
  id: "deepseek/deepseek-v4-pro",
@@ -9463,12 +9297,12 @@ export const MODELS = {
9463
9297
  input: ["text", "image"],
9464
9298
  cost: {
9465
9299
  input: 0.12,
9466
- output: 0.36,
9300
+ output: 0.35,
9467
9301
  cacheRead: 0.09,
9468
9302
  cacheWrite: 0,
9469
9303
  },
9470
9304
  contextWindow: 262144,
9471
- maxTokens: 8192,
9305
+ maxTokens: 262144,
9472
9306
  },
9473
9307
  "google/gemma-4-31b-it:free": {
9474
9308
  id: "google/gemma-4-31b-it:free",
@@ -9769,9 +9603,9 @@ export const MODELS = {
9769
9603
  reasoning: true,
9770
9604
  input: ["text"],
9771
9605
  cost: {
9772
- input: 0.27,
9773
- output: 1.08,
9774
- cacheRead: 0.054,
9606
+ input: 0.25,
9607
+ output: 1,
9608
+ cacheRead: 0.049999999999999996,
9775
9609
  cacheWrite: 0,
9776
9610
  },
9777
9611
  contextWindow: 204800,
@@ -10161,27 +9995,26 @@ export const MODELS = {
10161
9995
  reasoning: true,
10162
9996
  input: ["text", "image"],
10163
9997
  cost: {
10164
- input: 0.6799999999999999,
10165
- output: 3.41,
10166
- cacheRead: 0.33999999999999997,
9998
+ input: 0.67,
9999
+ output: 3.39,
10000
+ cacheRead: 0.14,
10167
10001
  cacheWrite: 0,
10168
10002
  },
10169
10003
  contextWindow: 262144,
10170
- maxTokens: 262142,
10004
+ maxTokens: 262144,
10171
10005
  },
10172
- "moonshotai/kimi-k2.6:free": {
10173
- id: "moonshotai/kimi-k2.6:free",
10174
- name: "MoonshotAI: Kimi K2.6 (free)",
10006
+ "moonshotai/kimi-k2.7-code": {
10007
+ id: "moonshotai/kimi-k2.7-code",
10008
+ name: "MoonshotAI: Kimi K2.7 Code",
10175
10009
  api: "openai-completions",
10176
10010
  provider: "openrouter",
10177
10011
  baseUrl: "https://openrouter.ai/api/v1",
10178
- compat: { "supportsDeveloperRole": false },
10179
10012
  reasoning: true,
10180
10013
  input: ["text", "image"],
10181
10014
  cost: {
10182
- input: 0,
10183
- output: 0,
10184
- cacheRead: 0,
10015
+ input: 0.95,
10016
+ output: 4,
10017
+ cacheRead: 0.19,
10185
10018
  cacheWrite: 0,
10186
10019
  },
10187
10020
  contextWindow: 262144,
@@ -10357,23 +10190,6 @@ export const MODELS = {
10357
10190
  contextWindow: 128000,
10358
10191
  maxTokens: 128000,
10359
10192
  },
10360
- "nvidia/nemotron-nano-9b-v2": {
10361
- id: "nvidia/nemotron-nano-9b-v2",
10362
- name: "NVIDIA: Nemotron Nano 9B V2",
10363
- api: "openai-completions",
10364
- provider: "openrouter",
10365
- baseUrl: "https://openrouter.ai/api/v1",
10366
- reasoning: true,
10367
- input: ["text"],
10368
- cost: {
10369
- input: 0.04,
10370
- output: 0.16,
10371
- cacheRead: 0,
10372
- cacheWrite: 0,
10373
- },
10374
- contextWindow: 131072,
10375
- maxTokens: 16384,
10376
- },
10377
10193
  "nvidia/nemotron-nano-9b-v2:free": {
10378
10194
  id: "nvidia/nemotron-nano-9b-v2:free",
10379
10195
  name: "NVIDIA: Nemotron Nano 9B V2 (free)",
@@ -12095,13 +11911,13 @@ export const MODELS = {
12095
11911
  reasoning: true,
12096
11912
  input: ["text", "image"],
12097
11913
  cost: {
12098
- input: 0.28900000000000003,
12099
- output: 2.4,
11914
+ input: 0.28700000000000003,
11915
+ output: 3.1,
12100
11916
  cacheRead: 0,
12101
11917
  cacheWrite: 0,
12102
11918
  },
12103
11919
  contextWindow: 262144,
12104
- maxTokens: 131072,
11920
+ maxTokens: 262140,
12105
11921
  },
12106
11922
  "qwen/qwen3.6-35b-a3b": {
12107
11923
  id: "qwen/qwen3.6-35b-a3b",
@@ -12112,13 +11928,13 @@ export const MODELS = {
12112
11928
  reasoning: true,
12113
11929
  input: ["text", "image"],
12114
11930
  cost: {
12115
- input: 0.14,
11931
+ input: 0.15,
12116
11932
  output: 1,
12117
- cacheRead: 0,
11933
+ cacheRead: 0.049999999999999996,
12118
11934
  cacheWrite: 0,
12119
11935
  },
12120
11936
  contextWindow: 262144,
12121
- maxTokens: 262140,
11937
+ maxTokens: 262144,
12122
11938
  },
12123
11939
  "qwen/qwen3.6-flash": {
12124
11940
  id: "qwen/qwen3.6-flash",
@@ -12197,10 +12013,10 @@ export const MODELS = {
12197
12013
  reasoning: true,
12198
12014
  input: ["text", "image"],
12199
12015
  cost: {
12200
- input: 0.39999999999999997,
12201
- output: 1.5999999999999999,
12202
- cacheRead: 0.08,
12203
- cacheWrite: 0.5,
12016
+ input: 0.32,
12017
+ output: 1.28,
12018
+ cacheRead: 0.064,
12019
+ cacheWrite: 0.39999999999999997,
12204
12020
  },
12205
12021
  contextWindow: 1000000,
12206
12022
  maxTokens: 65536,
@@ -12741,13 +12557,13 @@ export const MODELS = {
12741
12557
  reasoning: true,
12742
12558
  input: ["text", "image"],
12743
12559
  cost: {
12744
- input: 0.6799999999999999,
12745
- output: 3.41,
12746
- cacheRead: 0.33999999999999997,
12560
+ input: 0.67,
12561
+ output: 3.39,
12562
+ cacheRead: 0.14,
12747
12563
  cacheWrite: 0,
12748
12564
  },
12749
12565
  contextWindow: 262144,
12750
- maxTokens: 262142,
12566
+ maxTokens: 262144,
12751
12567
  },
12752
12568
  "~openai/gpt-latest": {
12753
12569
  id: "~openai/gpt-latest",
@@ -12785,25 +12601,6 @@ export const MODELS = {
12785
12601
  },
12786
12602
  },
12787
12603
  "together": {
12788
- "MiniMaxAI/MiniMax-M2.5": {
12789
- id: "MiniMaxAI/MiniMax-M2.5",
12790
- name: "MiniMax-M2.5",
12791
- api: "openai-completions",
12792
- provider: "together",
12793
- baseUrl: "https://api.together.ai/v1",
12794
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12795
- reasoning: true,
12796
- thinkingLevelMap: { "off": null, "minimal": null, "low": null, "medium": null },
12797
- input: ["text"],
12798
- cost: {
12799
- input: 0.3,
12800
- output: 1.2,
12801
- cacheRead: 0.06,
12802
- cacheWrite: 0,
12803
- },
12804
- contextWindow: 204800,
12805
- maxTokens: 131072,
12806
- },
12807
12604
  "MiniMaxAI/MiniMax-M2.7": {
12808
12605
  id: "MiniMaxAI/MiniMax-M2.7",
12809
12606
  name: "MiniMax-M2.7",
@@ -12823,28 +12620,27 @@ export const MODELS = {
12823
12620
  contextWindow: 202752,
12824
12621
  maxTokens: 131072,
12825
12622
  },
12826
- "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12827
- id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12828
- name: "Qwen3 235B A22B Instruct 2507 FP8",
12623
+ "Qwen/Qwen2.5-7B-Instruct-Turbo": {
12624
+ id: "Qwen/Qwen2.5-7B-Instruct-Turbo",
12625
+ name: "Qwen 2.5 7B Instruct Turbo",
12829
12626
  api: "openai-completions",
12830
12627
  provider: "together",
12831
12628
  baseUrl: "https://api.together.ai/v1",
12832
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12833
- reasoning: true,
12834
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12629
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12630
+ reasoning: false,
12835
12631
  input: ["text"],
12836
12632
  cost: {
12837
- input: 0.2,
12838
- output: 0.6,
12633
+ input: 0.3,
12634
+ output: 0.3,
12839
12635
  cacheRead: 0,
12840
12636
  cacheWrite: 0,
12841
12637
  },
12842
- contextWindow: 262144,
12843
- maxTokens: 262144,
12638
+ contextWindow: 32768,
12639
+ maxTokens: 32768,
12844
12640
  },
12845
- "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {
12846
- id: "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8",
12847
- name: "Qwen3 Coder 480B A35B Instruct",
12641
+ "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {
12642
+ id: "Qwen/Qwen3-235B-A22B-Instruct-2507-tput",
12643
+ name: "Qwen3 235B A22B Instruct 2507 FP8",
12848
12644
  api: "openai-completions",
12849
12645
  provider: "together",
12850
12646
  baseUrl: "https://api.together.ai/v1",
@@ -12852,36 +12648,36 @@ export const MODELS = {
12852
12648
  reasoning: false,
12853
12649
  input: ["text"],
12854
12650
  cost: {
12855
- input: 2,
12856
- output: 2,
12651
+ input: 0.2,
12652
+ output: 0.6,
12857
12653
  cacheRead: 0,
12858
12654
  cacheWrite: 0,
12859
12655
  },
12860
12656
  contextWindow: 262144,
12861
12657
  maxTokens: 262144,
12862
12658
  },
12863
- "Qwen/Qwen3-Coder-Next-FP8": {
12864
- id: "Qwen/Qwen3-Coder-Next-FP8",
12865
- name: "Qwen3 Coder Next FP8",
12659
+ "Qwen/Qwen3.5-397B-A17B": {
12660
+ id: "Qwen/Qwen3.5-397B-A17B",
12661
+ name: "Qwen3.5 397B A17B",
12866
12662
  api: "openai-completions",
12867
12663
  provider: "together",
12868
12664
  baseUrl: "https://api.together.ai/v1",
12869
12665
  compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12870
12666
  reasoning: true,
12871
12667
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12872
- input: ["text"],
12668
+ input: ["text", "image"],
12873
12669
  cost: {
12874
- input: 0.5,
12875
- output: 1.2,
12670
+ input: 0.6,
12671
+ output: 3.6,
12876
12672
  cacheRead: 0,
12877
12673
  cacheWrite: 0,
12878
12674
  },
12879
12675
  contextWindow: 262144,
12880
- maxTokens: 262144,
12676
+ maxTokens: 130000,
12881
12677
  },
12882
- "Qwen/Qwen3.5-397B-A17B": {
12883
- id: "Qwen/Qwen3.5-397B-A17B",
12884
- name: "Qwen3.5 397B A17B",
12678
+ "Qwen/Qwen3.5-9B": {
12679
+ id: "Qwen/Qwen3.5-9B",
12680
+ name: "Qwen3.5 9B",
12885
12681
  api: "openai-completions",
12886
12682
  provider: "together",
12887
12683
  baseUrl: "https://api.together.ai/v1",
@@ -12890,13 +12686,13 @@ export const MODELS = {
12890
12686
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12891
12687
  input: ["text", "image"],
12892
12688
  cost: {
12893
- input: 0.6,
12894
- output: 3.6,
12689
+ input: 0.17,
12690
+ output: 0.25,
12895
12691
  cacheRead: 0,
12896
12692
  cacheWrite: 0,
12897
12693
  },
12898
12694
  contextWindow: 262144,
12899
- maxTokens: 130000,
12695
+ maxTokens: 65536,
12900
12696
  },
12901
12697
  "Qwen/Qwen3.6-Plus": {
12902
12698
  id: "Qwen/Qwen3.6-Plus",
@@ -12923,9 +12719,8 @@ export const MODELS = {
12923
12719
  api: "openai-completions",
12924
12720
  provider: "together",
12925
12721
  baseUrl: "https://api.together.ai/v1",
12926
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12927
- reasoning: true,
12928
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12722
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false },
12723
+ reasoning: false,
12929
12724
  input: ["text"],
12930
12725
  cost: {
12931
12726
  input: 2.5,
@@ -12936,44 +12731,6 @@ export const MODELS = {
12936
12731
  contextWindow: 1000000,
12937
12732
  maxTokens: 500000,
12938
12733
  },
12939
- "deepseek-ai/DeepSeek-V3": {
12940
- id: "deepseek-ai/DeepSeek-V3",
12941
- name: "DeepSeek-V3",
12942
- api: "openai-completions",
12943
- provider: "together",
12944
- baseUrl: "https://api.together.ai/v1",
12945
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12946
- reasoning: true,
12947
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12948
- input: ["text"],
12949
- cost: {
12950
- input: 1.25,
12951
- output: 1.25,
12952
- cacheRead: 0,
12953
- cacheWrite: 0,
12954
- },
12955
- contextWindow: 131072,
12956
- maxTokens: 131072,
12957
- },
12958
- "deepseek-ai/DeepSeek-V3-1": {
12959
- id: "deepseek-ai/DeepSeek-V3-1",
12960
- name: "DeepSeek V3.1",
12961
- api: "openai-completions",
12962
- provider: "together",
12963
- baseUrl: "https://api.together.ai/v1",
12964
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12965
- reasoning: true,
12966
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12967
- input: ["text"],
12968
- cost: {
12969
- input: 0.6,
12970
- output: 1.7,
12971
- cacheRead: 0,
12972
- cacheWrite: 0,
12973
- },
12974
- contextWindow: 131072,
12975
- maxTokens: 131072,
12976
- },
12977
12734
  "deepseek-ai/DeepSeek-V4-Pro": {
12978
12735
  id: "deepseek-ai/DeepSeek-V4-Pro",
12979
12736
  name: "DeepSeek V4 Pro",
@@ -12985,8 +12742,8 @@ export const MODELS = {
12985
12742
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null, "high": "high", "xhigh": null },
12986
12743
  input: ["text"],
12987
12744
  cost: {
12988
- input: 2.1,
12989
- output: 4.4,
12745
+ input: 1.74,
12746
+ output: 3.48,
12990
12747
  cacheRead: 0.2,
12991
12748
  cacheWrite: 0,
12992
12749
  },
@@ -13022,8 +12779,8 @@ export const MODELS = {
13022
12779
  thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13023
12780
  input: ["text", "image"],
13024
12781
  cost: {
13025
- input: 0.2,
13026
- output: 0.5,
12782
+ input: 0.39,
12783
+ output: 0.97,
13027
12784
  cacheRead: 0,
13028
12785
  cacheWrite: 0,
13029
12786
  },
@@ -13048,25 +12805,6 @@ export const MODELS = {
13048
12805
  contextWindow: 131072,
13049
12806
  maxTokens: 131072,
13050
12807
  },
13051
- "moonshotai/Kimi-K2.5": {
13052
- id: "moonshotai/Kimi-K2.5",
13053
- name: "Kimi K2.5",
13054
- api: "openai-completions",
13055
- provider: "together",
13056
- baseUrl: "https://api.together.ai/v1",
13057
- compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
13058
- reasoning: true,
13059
- thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
13060
- input: ["text", "image"],
13061
- cost: {
13062
- input: 0.5,
13063
- output: 2.8,
13064
- cacheRead: 0,
13065
- cacheWrite: 0,
13066
- },
13067
- contextWindow: 262144,
13068
- maxTokens: 262144,
13069
- },
13070
12808
  "moonshotai/Kimi-K2.6": {
13071
12809
  id: "moonshotai/Kimi-K2.6",
13072
12810
  name: "Kimi K2.6",
@@ -13124,6 +12862,44 @@ export const MODELS = {
13124
12862
  contextWindow: 131072,
13125
12863
  maxTokens: 131072,
13126
12864
  },
12865
+ "openai/gpt-oss-20b": {
12866
+ id: "openai/gpt-oss-20b",
12867
+ name: "GPT OSS 20B",
12868
+ api: "openai-completions",
12869
+ provider: "together",
12870
+ baseUrl: "https://api.together.ai/v1",
12871
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": true, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "openai" },
12872
+ reasoning: true,
12873
+ thinkingLevelMap: { "off": null, "minimal": null },
12874
+ input: ["text"],
12875
+ cost: {
12876
+ input: 0.05,
12877
+ output: 0.2,
12878
+ cacheRead: 0,
12879
+ cacheWrite: 0,
12880
+ },
12881
+ contextWindow: 131072,
12882
+ maxTokens: 131072,
12883
+ },
12884
+ "zai-org/GLM-5": {
12885
+ id: "zai-org/GLM-5",
12886
+ name: "GLM-5",
12887
+ api: "openai-completions",
12888
+ provider: "together",
12889
+ baseUrl: "https://api.together.ai/v1",
12890
+ compat: { "supportsStore": false, "supportsDeveloperRole": false, "supportsReasoningEffort": false, "maxTokensField": "max_tokens", "supportsStrictMode": false, "supportsLongCacheRetention": false, "thinkingFormat": "together" },
12891
+ reasoning: true,
12892
+ thinkingLevelMap: { "minimal": null, "low": null, "medium": null },
12893
+ input: ["text"],
12894
+ cost: {
12895
+ input: 1,
12896
+ output: 3.2,
12897
+ cacheRead: 0,
12898
+ cacheWrite: 0,
12899
+ },
12900
+ contextWindow: 202752,
12901
+ maxTokens: 131072,
12902
+ },
13127
12903
  "zai-org/GLM-5.1": {
13128
12904
  id: "zai-org/GLM-5.1",
13129
12905
  name: "GLM-5.1",
@@ -14699,40 +14475,6 @@ export const MODELS = {
14699
14475
  contextWindow: 262114,
14700
14476
  maxTokens: 262114,
14701
14477
  },
14702
- "moonshotai/kimi-k2-thinking-turbo": {
14703
- id: "moonshotai/kimi-k2-thinking-turbo",
14704
- name: "Kimi K2 Thinking Turbo",
14705
- api: "anthropic-messages",
14706
- provider: "vercel-ai-gateway",
14707
- baseUrl: "https://ai-gateway.vercel.sh",
14708
- reasoning: true,
14709
- input: ["text"],
14710
- cost: {
14711
- input: 1.15,
14712
- output: 8,
14713
- cacheRead: 0.15,
14714
- cacheWrite: 0,
14715
- },
14716
- contextWindow: 262114,
14717
- maxTokens: 262114,
14718
- },
14719
- "moonshotai/kimi-k2-turbo": {
14720
- id: "moonshotai/kimi-k2-turbo",
14721
- name: "Kimi K2 Turbo",
14722
- api: "anthropic-messages",
14723
- provider: "vercel-ai-gateway",
14724
- baseUrl: "https://ai-gateway.vercel.sh",
14725
- reasoning: false,
14726
- input: ["text"],
14727
- cost: {
14728
- input: 1.15,
14729
- output: 8,
14730
- cacheRead: 0.15,
14731
- cacheWrite: 0,
14732
- },
14733
- contextWindow: 256000,
14734
- maxTokens: 16384,
14735
- },
14736
14478
  "moonshotai/kimi-k2.5": {
14737
14479
  id: "moonshotai/kimi-k2.5",
14738
14480
  name: "Kimi K2.5",