llm.rb 15.0.3 → 15.2.0
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.
- checksums.yaml +4 -4
- data/CHANGELOG.md +433 -3
- data/README.md +188 -71
- data/bin/llm.rb +50 -7
- data/data/alibaba.json +912 -823
- data/data/anthropic.json +234 -187
- data/data/bedrock.json +3702 -2058
- data/data/deepinfra.json +1288 -951
- data/data/deepseek.json +87 -53
- data/data/google.json +670 -670
- data/data/mistral.json +501 -460
- data/data/moonshot.json +43 -248
- data/data/openai.json +1008 -914
- data/data/openrouter.json +14417 -0
- data/data/xai.json +213 -201
- data/data/zai.json +242 -149
- data/docs/deepdive/advanced/compaction.md +5 -5
- data/docs/deepdive/advanced/context.md +8 -6
- data/docs/deepdive/advanced/guard.md +2 -2
- data/docs/deepdive/features/builtin_tools.md +93 -22
- data/docs/deepdive/features/{repl.md → console.md} +28 -28
- data/docs/deepdive/features/database.md +3 -3
- data/docs/deepdive/fundamentals/agents.md +13 -12
- data/docs/deepdive/fundamentals/providers.md +91 -6
- data/docs/deepdive/fundamentals/skills.md +14 -6
- data/docs/deepdive/fundamentals/stream.md +4 -4
- data/docs/deepdive/fundamentals/tools.md +63 -31
- data/docs/deepdive/reference/cost.md +2 -2
- data/docs/deepdive/reference/model_registry.md +2 -2
- data/docs/deepdive/reference/tracer.md +15 -13
- data/docs/deepdive.md +2 -2
- data/lib/llm/active_record/acts_as_agent.rb +9 -5
- data/lib/llm/agent.rb +40 -15
- data/lib/llm/{repl → console}/bar.rb +3 -3
- data/lib/llm/{repl → console}/buffer.rb +24 -9
- data/lib/llm/{repl → console}/color.rb +2 -2
- data/lib/llm/{repl → console}/command.rb +12 -12
- data/lib/llm/{repl → console}/commands/exit.rb +4 -4
- data/lib/llm/{repl → console}/commands/help.rb +1 -1
- data/lib/llm/{repl/commands/compact.rb → console/commands/keep.rb} +11 -9
- data/lib/llm/{repl → console}/commands/model.rb +2 -2
- data/lib/llm/{repl → console}/input/cache.rb +2 -2
- data/lib/llm/{repl → console}/input/char.rb +2 -2
- data/lib/llm/{repl → console}/input/row.rb +1 -1
- data/lib/llm/{repl → console}/input.rb +18 -10
- data/lib/llm/console/markdown/parser.rb +78 -0
- data/lib/llm/{repl → console}/markdown/table.rb +8 -5
- data/lib/llm/{repl → console}/markdown.rb +13 -30
- data/lib/llm/console/node.rb +69 -0
- data/lib/llm/{repl → console}/status.rb +11 -11
- data/lib/llm/{repl → console}/stream.rb +36 -9
- data/lib/llm/{repl → console}/walker.rb +1 -1
- data/lib/llm/{repl → console}/window.rb +17 -17
- data/lib/llm/{repl.rb → console.rb} +39 -19
- data/lib/llm/context/deserializer.rb +2 -1
- data/lib/llm/context.rb +29 -13
- data/lib/llm/cost.rb +13 -0
- data/lib/llm/function/async/reactor.rb +20 -1
- data/lib/llm/function/fork/task.rb +14 -10
- data/lib/llm/function.rb +1 -1
- data/lib/llm/json_adapter.rb +40 -28
- data/lib/llm/message.rb +7 -0
- data/lib/llm/provider.rb +31 -10
- data/lib/llm/providers/alibaba.rb +1 -1
- data/lib/llm/providers/anthropic.rb +1 -1
- data/lib/llm/providers/bedrock/models.rb +2 -2
- data/lib/llm/providers/bedrock.rb +1 -1
- data/lib/llm/providers/deepseek.rb +1 -1
- data/lib/llm/providers/google.rb +1 -1
- data/lib/llm/providers/ollama.rb +1 -1
- data/lib/llm/providers/openai/responses.rb +2 -1
- data/lib/llm/providers/openai.rb +2 -1
- data/lib/llm/providers/openrouter.rb +87 -0
- data/lib/llm/schema/leaf.rb +34 -2
- data/lib/llm/schema.rb +4 -2
- data/lib/llm/sequel/agent.rb +9 -5
- data/lib/llm/skill.rb +7 -1
- data/lib/llm/stream.rb +8 -3
- data/lib/llm/tool/param.rb +5 -1
- data/lib/llm/tool.rb +5 -0
- data/lib/llm/tools/bundle.rb +53 -0
- data/lib/llm/tools/edit-file.rb +7 -2
- data/lib/llm/tools/exec.rb +78 -0
- data/lib/llm/tools/git.rb +27 -26
- data/lib/llm/tools/mkdir.rb +12 -19
- data/lib/llm/tools/read_file.rb +69 -9
- data/lib/llm/tools/rg.rb +20 -24
- data/lib/llm/tools/ruby.rb +17 -25
- data/lib/llm/tools/utils.rb +75 -2
- data/lib/llm/tools/write_file.rb +4 -1
- data/lib/llm/tracer/logger.rb +2 -2
- data/lib/llm/tracer/pretty_logger.rb +4 -4
- data/lib/llm/tracer/telemetry.rb +2 -2
- data/lib/llm/tracer.rb +33 -0
- data/lib/llm/transport/curb.rb +5 -3
- data/lib/llm/transport/http.rb +5 -2
- data/lib/llm/transport/persistent_http.rb +6 -4
- data/lib/llm/transport/utils.rb +8 -6
- data/lib/llm/version.rb +1 -1
- data/lib/llm.rb +18 -12
- data/llm.gemspec +8 -8
- metadata +80 -37
- data/lib/llm/repl/node.rb +0 -44
- data/lib/llm/tools/shell.rb +0 -55
data/data/deepinfra.json
CHANGED
|
@@ -7,19 +7,423 @@
|
|
|
7
7
|
"name": "Deep Infra",
|
|
8
8
|
"doc": "https://deepinfra.com/models",
|
|
9
9
|
"models": {
|
|
10
|
-
"
|
|
11
|
-
"id": "
|
|
12
|
-
"name": "
|
|
13
|
-
"description": "
|
|
14
|
-
"family": "
|
|
10
|
+
"ByteDance/Seed-2.0-mini": {
|
|
11
|
+
"id": "ByteDance/Seed-2.0-mini",
|
|
12
|
+
"name": "Seed 2.0 Mini",
|
|
13
|
+
"description": "Lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks",
|
|
14
|
+
"family": "seed",
|
|
15
|
+
"attachment": true,
|
|
16
|
+
"reasoning": true,
|
|
17
|
+
"reasoning_options": [],
|
|
18
|
+
"tool_call": true,
|
|
19
|
+
"structured_output": true,
|
|
20
|
+
"temperature": true,
|
|
21
|
+
"release_date": "2026-02-14",
|
|
22
|
+
"last_updated": "2026-02-14",
|
|
23
|
+
"modalities": {
|
|
24
|
+
"input": [
|
|
25
|
+
"text",
|
|
26
|
+
"image"
|
|
27
|
+
],
|
|
28
|
+
"output": [
|
|
29
|
+
"text"
|
|
30
|
+
]
|
|
31
|
+
},
|
|
32
|
+
"open_weights": false,
|
|
33
|
+
"limit": {
|
|
34
|
+
"context": 256000,
|
|
35
|
+
"output": 32000
|
|
36
|
+
},
|
|
37
|
+
"cost": {
|
|
38
|
+
"input": 0.1,
|
|
39
|
+
"output": 0.4,
|
|
40
|
+
"cache_read": 0.02,
|
|
41
|
+
"tiers": [
|
|
42
|
+
{
|
|
43
|
+
"input": 0.2,
|
|
44
|
+
"output": 0.8,
|
|
45
|
+
"cache_read": 0.2,
|
|
46
|
+
"tier": {
|
|
47
|
+
"type": "context",
|
|
48
|
+
"size": 128000
|
|
49
|
+
}
|
|
50
|
+
}
|
|
51
|
+
]
|
|
52
|
+
}
|
|
53
|
+
},
|
|
54
|
+
"ByteDance/Seed-2.0-code": {
|
|
55
|
+
"id": "ByteDance/Seed-2.0-code",
|
|
56
|
+
"name": "Seed 2.0 Code",
|
|
57
|
+
"description": "ByteDance Seed coding model for multimodal software engineering and long-running agents",
|
|
58
|
+
"family": "seed",
|
|
59
|
+
"attachment": true,
|
|
60
|
+
"reasoning": true,
|
|
61
|
+
"reasoning_options": [
|
|
62
|
+
{
|
|
63
|
+
"type": "effort",
|
|
64
|
+
"values": [
|
|
65
|
+
"none",
|
|
66
|
+
"low",
|
|
67
|
+
"medium",
|
|
68
|
+
"high"
|
|
69
|
+
]
|
|
70
|
+
}
|
|
71
|
+
],
|
|
72
|
+
"tool_call": true,
|
|
73
|
+
"structured_output": true,
|
|
74
|
+
"temperature": true,
|
|
75
|
+
"release_date": "2026-02-14",
|
|
76
|
+
"last_updated": "2026-02-14",
|
|
77
|
+
"modalities": {
|
|
78
|
+
"input": [
|
|
79
|
+
"text",
|
|
80
|
+
"image"
|
|
81
|
+
],
|
|
82
|
+
"output": [
|
|
83
|
+
"text"
|
|
84
|
+
]
|
|
85
|
+
},
|
|
86
|
+
"open_weights": false,
|
|
87
|
+
"limit": {
|
|
88
|
+
"context": 256000,
|
|
89
|
+
"output": 131072
|
|
90
|
+
},
|
|
91
|
+
"cost": {
|
|
92
|
+
"input": 0.5,
|
|
93
|
+
"output": 3,
|
|
94
|
+
"cache_read": 0.1,
|
|
95
|
+
"tiers": [
|
|
96
|
+
{
|
|
97
|
+
"input": 1,
|
|
98
|
+
"output": 6,
|
|
99
|
+
"cache_read": 0.2,
|
|
100
|
+
"tier": {
|
|
101
|
+
"type": "context",
|
|
102
|
+
"size": 128000
|
|
103
|
+
}
|
|
104
|
+
}
|
|
105
|
+
]
|
|
106
|
+
}
|
|
107
|
+
},
|
|
108
|
+
"ByteDance/Seed-2.0-pro": {
|
|
109
|
+
"id": "ByteDance/Seed-2.0-pro",
|
|
110
|
+
"name": "Seed 2.0 Pro",
|
|
111
|
+
"description": "Flagship ByteDance Seed 2.0 model for complex multimodal reasoning and long-horizon agent workflows",
|
|
112
|
+
"family": "seed",
|
|
113
|
+
"attachment": true,
|
|
114
|
+
"reasoning": true,
|
|
115
|
+
"reasoning_options": [],
|
|
116
|
+
"tool_call": true,
|
|
117
|
+
"structured_output": true,
|
|
118
|
+
"temperature": true,
|
|
119
|
+
"release_date": "2026-02-14",
|
|
120
|
+
"last_updated": "2026-02-14",
|
|
121
|
+
"modalities": {
|
|
122
|
+
"input": [
|
|
123
|
+
"text",
|
|
124
|
+
"image"
|
|
125
|
+
],
|
|
126
|
+
"output": [
|
|
127
|
+
"text"
|
|
128
|
+
]
|
|
129
|
+
},
|
|
130
|
+
"open_weights": false,
|
|
131
|
+
"limit": {
|
|
132
|
+
"context": 256000,
|
|
133
|
+
"output": 128000
|
|
134
|
+
},
|
|
135
|
+
"cost": {
|
|
136
|
+
"input": 0.5,
|
|
137
|
+
"output": 3,
|
|
138
|
+
"cache_read": 0.1,
|
|
139
|
+
"tiers": [
|
|
140
|
+
{
|
|
141
|
+
"input": 1,
|
|
142
|
+
"output": 6,
|
|
143
|
+
"cache_read": 0.2,
|
|
144
|
+
"tier": {
|
|
145
|
+
"type": "context",
|
|
146
|
+
"size": 128000
|
|
147
|
+
}
|
|
148
|
+
}
|
|
149
|
+
]
|
|
150
|
+
}
|
|
151
|
+
},
|
|
152
|
+
"stepfun-ai/Step-3.7-Flash": {
|
|
153
|
+
"id": "stepfun-ai/Step-3.7-Flash",
|
|
154
|
+
"name": "Step 3.7 Flash",
|
|
155
|
+
"description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts",
|
|
156
|
+
"attachment": true,
|
|
157
|
+
"reasoning": true,
|
|
158
|
+
"reasoning_options": [],
|
|
159
|
+
"tool_call": true,
|
|
160
|
+
"temperature": true,
|
|
161
|
+
"knowledge": "2026-03-01",
|
|
162
|
+
"release_date": "2026-05-29",
|
|
163
|
+
"last_updated": "2026-05-29",
|
|
164
|
+
"modalities": {
|
|
165
|
+
"input": [
|
|
166
|
+
"text",
|
|
167
|
+
"image",
|
|
168
|
+
"video"
|
|
169
|
+
],
|
|
170
|
+
"output": [
|
|
171
|
+
"text"
|
|
172
|
+
]
|
|
173
|
+
},
|
|
174
|
+
"open_weights": true,
|
|
175
|
+
"limit": {
|
|
176
|
+
"context": 262144,
|
|
177
|
+
"output": 256000
|
|
178
|
+
},
|
|
179
|
+
"cost": {
|
|
180
|
+
"input": 0.2,
|
|
181
|
+
"output": 1.15,
|
|
182
|
+
"cache_read": 0.04
|
|
183
|
+
}
|
|
184
|
+
},
|
|
185
|
+
"deepseek-ai/DeepSeek-V3": {
|
|
186
|
+
"id": "deepseek-ai/DeepSeek-V3",
|
|
187
|
+
"name": "DeepSeek-V3",
|
|
188
|
+
"description": "Open DeepSeek MoE chat model for coding, math, and general reasoning",
|
|
189
|
+
"family": "deepseek",
|
|
190
|
+
"attachment": false,
|
|
191
|
+
"reasoning": false,
|
|
192
|
+
"tool_call": true,
|
|
193
|
+
"structured_output": true,
|
|
194
|
+
"temperature": true,
|
|
195
|
+
"release_date": "2024-12-26",
|
|
196
|
+
"last_updated": "2024-12-26",
|
|
197
|
+
"modalities": {
|
|
198
|
+
"input": [
|
|
199
|
+
"text"
|
|
200
|
+
],
|
|
201
|
+
"output": [
|
|
202
|
+
"text"
|
|
203
|
+
]
|
|
204
|
+
},
|
|
205
|
+
"open_weights": true,
|
|
206
|
+
"limit": {
|
|
207
|
+
"context": 163840,
|
|
208
|
+
"output": 8192
|
|
209
|
+
},
|
|
210
|
+
"cost": {
|
|
211
|
+
"input": 0.32,
|
|
212
|
+
"output": 0.89
|
|
213
|
+
}
|
|
214
|
+
},
|
|
215
|
+
"deepseek-ai/DeepSeek-V4-Flash": {
|
|
216
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash",
|
|
217
|
+
"name": "DeepSeek V4 Flash",
|
|
218
|
+
"description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work",
|
|
219
|
+
"family": "deepseek-flash",
|
|
220
|
+
"attachment": false,
|
|
221
|
+
"reasoning": true,
|
|
222
|
+
"reasoning_options": [
|
|
223
|
+
{
|
|
224
|
+
"type": "toggle"
|
|
225
|
+
},
|
|
226
|
+
{
|
|
227
|
+
"type": "effort",
|
|
228
|
+
"values": [
|
|
229
|
+
"low",
|
|
230
|
+
"medium",
|
|
231
|
+
"high",
|
|
232
|
+
"xhigh"
|
|
233
|
+
]
|
|
234
|
+
}
|
|
235
|
+
],
|
|
236
|
+
"tool_call": true,
|
|
237
|
+
"interleaved": {
|
|
238
|
+
"field": "reasoning_content"
|
|
239
|
+
},
|
|
240
|
+
"structured_output": true,
|
|
241
|
+
"temperature": true,
|
|
242
|
+
"knowledge": "2025-05",
|
|
243
|
+
"release_date": "2026-04-24",
|
|
244
|
+
"last_updated": "2026-04-24",
|
|
245
|
+
"modalities": {
|
|
246
|
+
"input": [
|
|
247
|
+
"text"
|
|
248
|
+
],
|
|
249
|
+
"output": [
|
|
250
|
+
"text"
|
|
251
|
+
]
|
|
252
|
+
},
|
|
253
|
+
"open_weights": true,
|
|
254
|
+
"limit": {
|
|
255
|
+
"context": 1048576,
|
|
256
|
+
"output": 16384
|
|
257
|
+
},
|
|
258
|
+
"cost": {
|
|
259
|
+
"input": 0.09,
|
|
260
|
+
"output": 0.18,
|
|
261
|
+
"cache_read": 0.018
|
|
262
|
+
}
|
|
263
|
+
},
|
|
264
|
+
"deepseek-ai/DeepSeek-V3-0324": {
|
|
265
|
+
"id": "deepseek-ai/DeepSeek-V3-0324",
|
|
266
|
+
"name": "DeepSeek V3 0324",
|
|
267
|
+
"description": "March 2025 checkpoint of DeepSeek-V3 with improved reasoning and coding",
|
|
268
|
+
"family": "deepseek",
|
|
269
|
+
"attachment": false,
|
|
270
|
+
"reasoning": false,
|
|
271
|
+
"tool_call": true,
|
|
272
|
+
"structured_output": true,
|
|
273
|
+
"temperature": true,
|
|
274
|
+
"release_date": "2025-03-24",
|
|
275
|
+
"last_updated": "2025-03-24",
|
|
276
|
+
"modalities": {
|
|
277
|
+
"input": [
|
|
278
|
+
"text"
|
|
279
|
+
],
|
|
280
|
+
"output": [
|
|
281
|
+
"text"
|
|
282
|
+
]
|
|
283
|
+
},
|
|
284
|
+
"open_weights": true,
|
|
285
|
+
"limit": {
|
|
286
|
+
"context": 163840,
|
|
287
|
+
"output": 163840
|
|
288
|
+
},
|
|
289
|
+
"cost": {
|
|
290
|
+
"input": 0.24,
|
|
291
|
+
"output": 0.9,
|
|
292
|
+
"cache_read": 0.135
|
|
293
|
+
}
|
|
294
|
+
},
|
|
295
|
+
"deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {
|
|
296
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
|
|
297
|
+
"name": "DeepSeek V4 Flash Vision Exp",
|
|
298
|
+
"description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
|
|
299
|
+
"family": "deepseek-flash",
|
|
300
|
+
"attachment": true,
|
|
301
|
+
"reasoning": true,
|
|
302
|
+
"reasoning_options": [
|
|
303
|
+
{
|
|
304
|
+
"type": "effort",
|
|
305
|
+
"values": [
|
|
306
|
+
"none",
|
|
307
|
+
"low",
|
|
308
|
+
"high",
|
|
309
|
+
"max"
|
|
310
|
+
]
|
|
311
|
+
}
|
|
312
|
+
],
|
|
313
|
+
"tool_call": true,
|
|
314
|
+
"structured_output": true,
|
|
315
|
+
"temperature": true,
|
|
316
|
+
"release_date": "2026-08-21",
|
|
317
|
+
"last_updated": "2026-08-21",
|
|
318
|
+
"modalities": {
|
|
319
|
+
"input": [
|
|
320
|
+
"text",
|
|
321
|
+
"image"
|
|
322
|
+
],
|
|
323
|
+
"output": [
|
|
324
|
+
"text"
|
|
325
|
+
]
|
|
326
|
+
},
|
|
327
|
+
"open_weights": false,
|
|
328
|
+
"limit": {
|
|
329
|
+
"context": 1048576,
|
|
330
|
+
"output": 384000
|
|
331
|
+
},
|
|
332
|
+
"cost": {
|
|
333
|
+
"input": 0.44,
|
|
334
|
+
"output": 1.32,
|
|
335
|
+
"cache_read": 0.014
|
|
336
|
+
}
|
|
337
|
+
},
|
|
338
|
+
"deepseek-ai/DeepSeek-V4-Flash-0731": {
|
|
339
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash-0731",
|
|
340
|
+
"name": "DeepSeek V4 Flash 0731",
|
|
341
|
+
"description": "Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding",
|
|
342
|
+
"family": "deepseek-flash",
|
|
343
|
+
"attachment": false,
|
|
344
|
+
"reasoning": true,
|
|
345
|
+
"reasoning_options": [
|
|
346
|
+
{
|
|
347
|
+
"type": "toggle"
|
|
348
|
+
}
|
|
349
|
+
],
|
|
350
|
+
"tool_call": true,
|
|
351
|
+
"structured_output": true,
|
|
352
|
+
"temperature": true,
|
|
353
|
+
"knowledge": "2025-05",
|
|
354
|
+
"release_date": "2026-07-31",
|
|
355
|
+
"last_updated": "2026-07-31",
|
|
356
|
+
"modalities": {
|
|
357
|
+
"input": [
|
|
358
|
+
"text"
|
|
359
|
+
],
|
|
360
|
+
"output": [
|
|
361
|
+
"text"
|
|
362
|
+
]
|
|
363
|
+
},
|
|
364
|
+
"open_weights": true,
|
|
365
|
+
"limit": {
|
|
366
|
+
"context": 1048576,
|
|
367
|
+
"output": 384000
|
|
368
|
+
},
|
|
369
|
+
"cost": {
|
|
370
|
+
"input": 0.06,
|
|
371
|
+
"output": 0.18,
|
|
372
|
+
"cache_read": 0.015
|
|
373
|
+
}
|
|
374
|
+
},
|
|
375
|
+
"deepseek-ai/DeepSeek-V3.1": {
|
|
376
|
+
"id": "deepseek-ai/DeepSeek-V3.1",
|
|
377
|
+
"name": "DeepSeek-V3.1",
|
|
378
|
+
"description": "Hybrid-reasoning DeepSeek model with thinking and non-thinking modes",
|
|
379
|
+
"family": "deepseek",
|
|
380
|
+
"attachment": false,
|
|
381
|
+
"reasoning": true,
|
|
382
|
+
"reasoning_options": [
|
|
383
|
+
{
|
|
384
|
+
"type": "toggle"
|
|
385
|
+
}
|
|
386
|
+
],
|
|
387
|
+
"tool_call": true,
|
|
388
|
+
"structured_output": true,
|
|
389
|
+
"temperature": true,
|
|
390
|
+
"release_date": "2025-08-21",
|
|
391
|
+
"last_updated": "2025-08-21",
|
|
392
|
+
"modalities": {
|
|
393
|
+
"input": [
|
|
394
|
+
"text"
|
|
395
|
+
],
|
|
396
|
+
"output": [
|
|
397
|
+
"text"
|
|
398
|
+
]
|
|
399
|
+
},
|
|
400
|
+
"open_weights": true,
|
|
401
|
+
"limit": {
|
|
402
|
+
"context": 163840,
|
|
403
|
+
"output": 8192
|
|
404
|
+
},
|
|
405
|
+
"cost": {
|
|
406
|
+
"input": 0.25,
|
|
407
|
+
"output": 0.95,
|
|
408
|
+
"cache_read": 0.13
|
|
409
|
+
}
|
|
410
|
+
},
|
|
411
|
+
"deepseek-ai/DeepSeek-R1-0528": {
|
|
412
|
+
"id": "deepseek-ai/DeepSeek-R1-0528",
|
|
413
|
+
"name": "DeepSeek-R1-0528",
|
|
414
|
+
"description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools",
|
|
15
415
|
"attachment": false,
|
|
16
416
|
"reasoning": true,
|
|
17
417
|
"reasoning_options": [],
|
|
18
418
|
"tool_call": true,
|
|
419
|
+
"interleaved": {
|
|
420
|
+
"field": "reasoning_content"
|
|
421
|
+
},
|
|
19
422
|
"structured_output": true,
|
|
20
423
|
"temperature": true,
|
|
21
|
-
"
|
|
22
|
-
"
|
|
424
|
+
"knowledge": "2024-07",
|
|
425
|
+
"release_date": "2025-05-28",
|
|
426
|
+
"last_updated": "2025-05-28",
|
|
23
427
|
"modalities": {
|
|
24
428
|
"input": [
|
|
25
429
|
"text"
|
|
@@ -28,44 +432,46 @@
|
|
|
28
432
|
"text"
|
|
29
433
|
]
|
|
30
434
|
},
|
|
31
|
-
"open_weights":
|
|
435
|
+
"open_weights": false,
|
|
32
436
|
"limit": {
|
|
33
|
-
"context":
|
|
437
|
+
"context": 163840,
|
|
34
438
|
"output": 64000
|
|
35
439
|
},
|
|
36
440
|
"cost": {
|
|
37
|
-
"input": 0.
|
|
38
|
-
"output":
|
|
39
|
-
"cache_read": 0.
|
|
441
|
+
"input": 0.5,
|
|
442
|
+
"output": 2.15,
|
|
443
|
+
"cache_read": 0.35
|
|
40
444
|
}
|
|
41
445
|
},
|
|
42
|
-
"
|
|
43
|
-
"id": "
|
|
44
|
-
"name": "
|
|
45
|
-
"description": "
|
|
46
|
-
"family": "
|
|
446
|
+
"deepseek-ai/DeepSeek-V4.1-Flash": {
|
|
447
|
+
"id": "deepseek-ai/DeepSeek-V4.1-Flash",
|
|
448
|
+
"name": "DeepSeek V4.1 Flash",
|
|
449
|
+
"description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
|
|
450
|
+
"family": "deepseek-flash",
|
|
47
451
|
"attachment": true,
|
|
48
452
|
"reasoning": true,
|
|
49
453
|
"reasoning_options": [
|
|
50
454
|
{
|
|
51
|
-
"type": "
|
|
455
|
+
"type": "effort",
|
|
456
|
+
"values": [
|
|
457
|
+
"none",
|
|
458
|
+
"low",
|
|
459
|
+
"high",
|
|
460
|
+
"xhigh",
|
|
461
|
+
"max"
|
|
462
|
+
]
|
|
52
463
|
}
|
|
53
464
|
],
|
|
54
465
|
"tool_call": true,
|
|
55
|
-
"interleaved": {
|
|
56
|
-
"field": "reasoning_content"
|
|
57
|
-
},
|
|
58
466
|
"structured_output": true,
|
|
59
467
|
"temperature": true,
|
|
60
|
-
"knowledge": "
|
|
61
|
-
"release_date": "2026-
|
|
62
|
-
"last_updated": "2026-
|
|
468
|
+
"knowledge": "2025-05",
|
|
469
|
+
"release_date": "2026-09-10",
|
|
470
|
+
"last_updated": "2026-09-10",
|
|
63
471
|
"modalities": {
|
|
64
472
|
"input": [
|
|
65
473
|
"text",
|
|
66
|
-
"image"
|
|
67
|
-
"audio",
|
|
68
|
-
"video"
|
|
474
|
+
"image"
|
|
69
475
|
],
|
|
70
476
|
"output": [
|
|
71
477
|
"text"
|
|
@@ -73,40 +479,42 @@
|
|
|
73
479
|
},
|
|
74
480
|
"open_weights": true,
|
|
75
481
|
"limit": {
|
|
76
|
-
"context":
|
|
77
|
-
"output":
|
|
482
|
+
"context": 1048576,
|
|
483
|
+
"output": 384000
|
|
78
484
|
},
|
|
79
485
|
"cost": {
|
|
80
|
-
"input": 0.
|
|
81
|
-
"output": 2,
|
|
82
|
-
"cache_read": 0.
|
|
486
|
+
"input": 0.3,
|
|
487
|
+
"output": 1.2,
|
|
488
|
+
"cache_read": 0.006
|
|
83
489
|
}
|
|
84
490
|
},
|
|
85
|
-
"
|
|
86
|
-
"id": "
|
|
87
|
-
"name": "
|
|
88
|
-
"description": "
|
|
89
|
-
"family": "
|
|
90
|
-
"attachment":
|
|
491
|
+
"deepseek-ai/DeepSeek-V4-Pro-0813": {
|
|
492
|
+
"id": "deepseek-ai/DeepSeek-V4-Pro-0813",
|
|
493
|
+
"name": "DeepSeek V4 Pro 0813",
|
|
494
|
+
"description": "DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes",
|
|
495
|
+
"family": "deepseek-thinking",
|
|
496
|
+
"attachment": false,
|
|
91
497
|
"reasoning": true,
|
|
92
498
|
"reasoning_options": [
|
|
93
499
|
{
|
|
94
500
|
"type": "toggle"
|
|
501
|
+
},
|
|
502
|
+
{
|
|
503
|
+
"type": "effort",
|
|
504
|
+
"values": [
|
|
505
|
+
"high",
|
|
506
|
+
"max"
|
|
507
|
+
]
|
|
95
508
|
}
|
|
96
509
|
],
|
|
97
510
|
"tool_call": true,
|
|
98
|
-
"interleaved": {
|
|
99
|
-
"field": "reasoning_content"
|
|
100
|
-
},
|
|
101
511
|
"structured_output": true,
|
|
102
512
|
"temperature": true,
|
|
103
|
-
"
|
|
104
|
-
"
|
|
105
|
-
"last_updated": "2026-04-22",
|
|
513
|
+
"release_date": "2026-08-12",
|
|
514
|
+
"last_updated": "2026-08-22",
|
|
106
515
|
"modalities": {
|
|
107
516
|
"input": [
|
|
108
|
-
"text"
|
|
109
|
-
"audio"
|
|
517
|
+
"text"
|
|
110
518
|
],
|
|
111
519
|
"output": [
|
|
112
520
|
"text"
|
|
@@ -115,26 +523,34 @@
|
|
|
115
523
|
"open_weights": true,
|
|
116
524
|
"limit": {
|
|
117
525
|
"context": 1048576,
|
|
118
|
-
"output":
|
|
526
|
+
"output": 384000
|
|
119
527
|
},
|
|
120
528
|
"cost": {
|
|
121
|
-
"input": 1,
|
|
122
|
-
"output":
|
|
123
|
-
"cache_read": 0.
|
|
529
|
+
"input": 1.3,
|
|
530
|
+
"output": 2.6,
|
|
531
|
+
"cache_read": 0.1
|
|
124
532
|
}
|
|
125
533
|
},
|
|
126
|
-
"
|
|
127
|
-
"id": "
|
|
128
|
-
"name": "
|
|
129
|
-
"description": "
|
|
130
|
-
"family": "minimax",
|
|
534
|
+
"deepseek-ai/DeepSeek-V3.2": {
|
|
535
|
+
"id": "deepseek-ai/DeepSeek-V3.2",
|
|
536
|
+
"name": "DeepSeek-V3.2",
|
|
537
|
+
"description": "DeepSeek chat model for instruction following, coding, and analysis",
|
|
131
538
|
"attachment": false,
|
|
132
539
|
"reasoning": true,
|
|
133
|
-
"reasoning_options": [
|
|
540
|
+
"reasoning_options": [
|
|
541
|
+
{
|
|
542
|
+
"type": "toggle"
|
|
543
|
+
}
|
|
544
|
+
],
|
|
134
545
|
"tool_call": true,
|
|
546
|
+
"interleaved": {
|
|
547
|
+
"field": "reasoning_content"
|
|
548
|
+
},
|
|
549
|
+
"structured_output": true,
|
|
135
550
|
"temperature": true,
|
|
136
|
-
"
|
|
137
|
-
"
|
|
551
|
+
"knowledge": "2024-12",
|
|
552
|
+
"release_date": "2025-12-02",
|
|
553
|
+
"last_updated": "2025-12-02",
|
|
138
554
|
"modalities": {
|
|
139
555
|
"input": [
|
|
140
556
|
"text"
|
|
@@ -143,35 +559,50 @@
|
|
|
143
559
|
"text"
|
|
144
560
|
]
|
|
145
561
|
},
|
|
146
|
-
"open_weights":
|
|
562
|
+
"open_weights": false,
|
|
147
563
|
"limit": {
|
|
148
|
-
"context":
|
|
149
|
-
"output":
|
|
564
|
+
"context": 163840,
|
|
565
|
+
"output": 64000
|
|
150
566
|
},
|
|
151
567
|
"cost": {
|
|
152
|
-
"input": 0.
|
|
153
|
-
"output":
|
|
154
|
-
"cache_read": 0.
|
|
568
|
+
"input": 0.26,
|
|
569
|
+
"output": 0.38,
|
|
570
|
+
"cache_read": 0.13
|
|
155
571
|
}
|
|
156
572
|
},
|
|
157
|
-
"
|
|
158
|
-
"id": "
|
|
159
|
-
"name": "
|
|
160
|
-
"description": "
|
|
161
|
-
"family": "
|
|
162
|
-
"attachment":
|
|
573
|
+
"deepseek-ai/DeepSeek-V4-Pro": {
|
|
574
|
+
"id": "deepseek-ai/DeepSeek-V4-Pro",
|
|
575
|
+
"name": "DeepSeek V4 Pro",
|
|
576
|
+
"description": "Open MoE flagship with million-token context for coding and long agent runs",
|
|
577
|
+
"family": "deepseek-thinking",
|
|
578
|
+
"attachment": false,
|
|
163
579
|
"reasoning": true,
|
|
164
|
-
"reasoning_options": [
|
|
580
|
+
"reasoning_options": [
|
|
581
|
+
{
|
|
582
|
+
"type": "toggle"
|
|
583
|
+
},
|
|
584
|
+
{
|
|
585
|
+
"type": "effort",
|
|
586
|
+
"values": [
|
|
587
|
+
"low",
|
|
588
|
+
"medium",
|
|
589
|
+
"high",
|
|
590
|
+
"xhigh"
|
|
591
|
+
]
|
|
592
|
+
}
|
|
593
|
+
],
|
|
165
594
|
"tool_call": true,
|
|
595
|
+
"interleaved": {
|
|
596
|
+
"field": "reasoning_content"
|
|
597
|
+
},
|
|
166
598
|
"structured_output": true,
|
|
167
599
|
"temperature": true,
|
|
168
|
-
"
|
|
169
|
-
"
|
|
600
|
+
"knowledge": "2025-05",
|
|
601
|
+
"release_date": "2026-04-24",
|
|
602
|
+
"last_updated": "2026-04-24",
|
|
170
603
|
"modalities": {
|
|
171
604
|
"input": [
|
|
172
|
-
"text"
|
|
173
|
-
"image",
|
|
174
|
-
"video"
|
|
605
|
+
"text"
|
|
175
606
|
],
|
|
176
607
|
"output": [
|
|
177
608
|
"text"
|
|
@@ -179,34 +610,34 @@
|
|
|
179
610
|
},
|
|
180
611
|
"open_weights": true,
|
|
181
612
|
"limit": {
|
|
182
|
-
"context":
|
|
183
|
-
"output":
|
|
613
|
+
"context": 1048576,
|
|
614
|
+
"output": 16384
|
|
184
615
|
},
|
|
185
616
|
"cost": {
|
|
186
|
-
"input":
|
|
187
|
-
"output":
|
|
188
|
-
"cache_read": 0.
|
|
617
|
+
"input": 1.3,
|
|
618
|
+
"output": 2.6,
|
|
619
|
+
"cache_read": 0.1
|
|
189
620
|
}
|
|
190
621
|
},
|
|
191
|
-
"
|
|
192
|
-
"id": "
|
|
193
|
-
"name": "
|
|
194
|
-
"description": "
|
|
195
|
-
"family": "
|
|
196
|
-
"attachment":
|
|
622
|
+
"nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning": {
|
|
623
|
+
"id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning",
|
|
624
|
+
"name": "Nemotron 3 Nano Omni 30B A3B Reasoning",
|
|
625
|
+
"description": "Open Nemotron omni model combining reasoning with text, vision, and audio",
|
|
626
|
+
"family": "nemotron",
|
|
627
|
+
"attachment": true,
|
|
197
628
|
"reasoning": true,
|
|
198
629
|
"reasoning_options": [],
|
|
199
630
|
"tool_call": true,
|
|
200
|
-
"
|
|
201
|
-
"field": "reasoning_content"
|
|
202
|
-
},
|
|
631
|
+
"structured_output": true,
|
|
203
632
|
"temperature": true,
|
|
204
|
-
"
|
|
205
|
-
"
|
|
206
|
-
"last_updated": "2026-02-12",
|
|
633
|
+
"release_date": "2026-04-28",
|
|
634
|
+
"last_updated": "2026-04-28",
|
|
207
635
|
"modalities": {
|
|
208
636
|
"input": [
|
|
209
|
-
"text"
|
|
637
|
+
"text",
|
|
638
|
+
"image",
|
|
639
|
+
"video",
|
|
640
|
+
"audio"
|
|
210
641
|
],
|
|
211
642
|
"output": [
|
|
212
643
|
"text"
|
|
@@ -214,14 +645,13 @@
|
|
|
214
645
|
},
|
|
215
646
|
"open_weights": true,
|
|
216
647
|
"limit": {
|
|
217
|
-
"context":
|
|
218
|
-
"output":
|
|
648
|
+
"context": 262144,
|
|
649
|
+
"output": 65536
|
|
219
650
|
},
|
|
220
651
|
"status": "deprecated",
|
|
221
652
|
"cost": {
|
|
222
|
-
"input": 0.
|
|
223
|
-
"output":
|
|
224
|
-
"cache_read": 0.03
|
|
653
|
+
"input": 0.2,
|
|
654
|
+
"output": 0.8
|
|
225
655
|
}
|
|
226
656
|
},
|
|
227
657
|
"nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": {
|
|
@@ -283,32 +713,72 @@
|
|
|
283
713
|
"open_weights": true,
|
|
284
714
|
"limit": {
|
|
285
715
|
"context": 262144,
|
|
286
|
-
"output": 262144
|
|
716
|
+
"output": 262144
|
|
717
|
+
},
|
|
718
|
+
"cost": {
|
|
719
|
+
"input": 0.05,
|
|
720
|
+
"output": 0.2,
|
|
721
|
+
"cache_read": 0.025
|
|
722
|
+
}
|
|
723
|
+
},
|
|
724
|
+
"google/gemma-4-31B-it": {
|
|
725
|
+
"id": "google/gemma-4-31B-it",
|
|
726
|
+
"name": "Gemma 4 31B IT",
|
|
727
|
+
"description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
|
|
728
|
+
"family": "gemma",
|
|
729
|
+
"attachment": true,
|
|
730
|
+
"reasoning": true,
|
|
731
|
+
"reasoning_options": [
|
|
732
|
+
{
|
|
733
|
+
"type": "toggle"
|
|
734
|
+
}
|
|
735
|
+
],
|
|
736
|
+
"tool_call": true,
|
|
737
|
+
"structured_output": true,
|
|
738
|
+
"temperature": true,
|
|
739
|
+
"release_date": "2026-04-02",
|
|
740
|
+
"last_updated": "2026-04-02",
|
|
741
|
+
"modalities": {
|
|
742
|
+
"input": [
|
|
743
|
+
"text",
|
|
744
|
+
"image",
|
|
745
|
+
"video"
|
|
746
|
+
],
|
|
747
|
+
"output": [
|
|
748
|
+
"text"
|
|
749
|
+
]
|
|
750
|
+
},
|
|
751
|
+
"open_weights": true,
|
|
752
|
+
"limit": {
|
|
753
|
+
"context": 262144,
|
|
754
|
+
"output": 32768
|
|
287
755
|
},
|
|
288
756
|
"cost": {
|
|
289
|
-
"input": 0.
|
|
290
|
-
"output": 0.
|
|
291
|
-
"cache_read": 0.025
|
|
757
|
+
"input": 0.13,
|
|
758
|
+
"output": 0.38
|
|
292
759
|
}
|
|
293
760
|
},
|
|
294
|
-
"
|
|
295
|
-
"id": "
|
|
296
|
-
"name": "
|
|
297
|
-
"description": "Open
|
|
298
|
-
"family": "
|
|
761
|
+
"google/gemma-4-E4B-it": {
|
|
762
|
+
"id": "google/gemma-4-E4B-it",
|
|
763
|
+
"name": "Gemma 4 E4B IT",
|
|
764
|
+
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
|
|
765
|
+
"family": "gemma",
|
|
299
766
|
"attachment": true,
|
|
300
767
|
"reasoning": true,
|
|
301
|
-
"reasoning_options": [
|
|
768
|
+
"reasoning_options": [
|
|
769
|
+
{
|
|
770
|
+
"type": "toggle"
|
|
771
|
+
}
|
|
772
|
+
],
|
|
302
773
|
"tool_call": true,
|
|
303
774
|
"structured_output": true,
|
|
304
775
|
"temperature": true,
|
|
305
|
-
"release_date": "2026-04-
|
|
306
|
-
"last_updated": "2026-04-
|
|
776
|
+
"release_date": "2026-04-02",
|
|
777
|
+
"last_updated": "2026-04-02",
|
|
307
778
|
"modalities": {
|
|
308
779
|
"input": [
|
|
309
780
|
"text",
|
|
310
781
|
"image",
|
|
311
|
-
"video",
|
|
312
782
|
"audio"
|
|
313
783
|
],
|
|
314
784
|
"output": [
|
|
@@ -317,13 +787,12 @@
|
|
|
317
787
|
},
|
|
318
788
|
"open_weights": true,
|
|
319
789
|
"limit": {
|
|
320
|
-
"context":
|
|
321
|
-
"output":
|
|
790
|
+
"context": 131072,
|
|
791
|
+
"output": 8192
|
|
322
792
|
},
|
|
323
|
-
"status": "deprecated",
|
|
324
793
|
"cost": {
|
|
325
|
-
"input": 0.
|
|
326
|
-
"output": 0.
|
|
794
|
+
"input": 0.02,
|
|
795
|
+
"output": 0.1
|
|
327
796
|
}
|
|
328
797
|
},
|
|
329
798
|
"google/gemma-4-26B-A4B-it": {
|
|
@@ -362,12 +831,12 @@
|
|
|
362
831
|
"output": 0.34
|
|
363
832
|
}
|
|
364
833
|
},
|
|
365
|
-
"
|
|
366
|
-
"id": "
|
|
367
|
-
"name": "
|
|
368
|
-
"description": "
|
|
369
|
-
"family": "
|
|
370
|
-
"attachment":
|
|
834
|
+
"zai-org/GLM-5.1": {
|
|
835
|
+
"id": "zai-org/GLM-5.1",
|
|
836
|
+
"name": "GLM-5.1",
|
|
837
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
838
|
+
"family": "glm",
|
|
839
|
+
"attachment": false,
|
|
371
840
|
"reasoning": true,
|
|
372
841
|
"reasoning_options": [
|
|
373
842
|
{
|
|
@@ -375,15 +844,17 @@
|
|
|
375
844
|
}
|
|
376
845
|
],
|
|
377
846
|
"tool_call": true,
|
|
847
|
+
"interleaved": {
|
|
848
|
+
"field": "reasoning_content"
|
|
849
|
+
},
|
|
378
850
|
"structured_output": true,
|
|
379
851
|
"temperature": true,
|
|
380
|
-
"
|
|
381
|
-
"
|
|
852
|
+
"knowledge": "2025-04",
|
|
853
|
+
"release_date": "2026-04-07",
|
|
854
|
+
"last_updated": "2026-04-07",
|
|
382
855
|
"modalities": {
|
|
383
856
|
"input": [
|
|
384
|
-
"text"
|
|
385
|
-
"image",
|
|
386
|
-
"audio"
|
|
857
|
+
"text"
|
|
387
858
|
],
|
|
388
859
|
"output": [
|
|
389
860
|
"text"
|
|
@@ -391,36 +862,40 @@
|
|
|
391
862
|
},
|
|
392
863
|
"open_weights": true,
|
|
393
864
|
"limit": {
|
|
394
|
-
"context":
|
|
395
|
-
"output":
|
|
865
|
+
"context": 202752,
|
|
866
|
+
"output": 16384
|
|
396
867
|
},
|
|
397
868
|
"cost": {
|
|
398
|
-
"input":
|
|
399
|
-
"output":
|
|
869
|
+
"input": 1.05,
|
|
870
|
+
"output": 3.5,
|
|
871
|
+
"cache_read": 0.205
|
|
400
872
|
}
|
|
401
873
|
},
|
|
402
|
-
"
|
|
403
|
-
"id": "
|
|
404
|
-
"name": "
|
|
405
|
-
"description": "
|
|
406
|
-
"family": "
|
|
407
|
-
"attachment":
|
|
874
|
+
"zai-org/GLM-5.3": {
|
|
875
|
+
"id": "zai-org/GLM-5.3",
|
|
876
|
+
"name": "GLM-5.3",
|
|
877
|
+
"description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
|
|
878
|
+
"family": "glm",
|
|
879
|
+
"attachment": false,
|
|
408
880
|
"reasoning": true,
|
|
409
881
|
"reasoning_options": [
|
|
410
882
|
{
|
|
411
|
-
"type": "
|
|
883
|
+
"type": "effort",
|
|
884
|
+
"values": [
|
|
885
|
+
"low",
|
|
886
|
+
"high",
|
|
887
|
+
"max"
|
|
888
|
+
]
|
|
412
889
|
}
|
|
413
890
|
],
|
|
414
891
|
"tool_call": true,
|
|
415
892
|
"structured_output": true,
|
|
416
893
|
"temperature": true,
|
|
417
|
-
"release_date": "2026-
|
|
418
|
-
"last_updated": "2026-
|
|
894
|
+
"release_date": "2026-08-14",
|
|
895
|
+
"last_updated": "2026-08-14",
|
|
419
896
|
"modalities": {
|
|
420
897
|
"input": [
|
|
421
|
-
"text"
|
|
422
|
-
"image",
|
|
423
|
-
"video"
|
|
898
|
+
"text"
|
|
424
899
|
],
|
|
425
900
|
"output": [
|
|
426
901
|
"text"
|
|
@@ -428,162 +903,145 @@
|
|
|
428
903
|
},
|
|
429
904
|
"open_weights": true,
|
|
430
905
|
"limit": {
|
|
431
|
-
"context":
|
|
432
|
-
"output":
|
|
906
|
+
"context": 1048576,
|
|
907
|
+
"output": 131072
|
|
433
908
|
},
|
|
434
909
|
"cost": {
|
|
435
|
-
"input":
|
|
436
|
-
"output":
|
|
910
|
+
"input": 1.2,
|
|
911
|
+
"output": 4,
|
|
912
|
+
"cache_read": 0.12
|
|
437
913
|
}
|
|
438
914
|
},
|
|
439
|
-
"
|
|
440
|
-
"id": "
|
|
441
|
-
"name": "
|
|
442
|
-
"description": "
|
|
443
|
-
"family": "
|
|
444
|
-
"attachment":
|
|
915
|
+
"zai-org/GLM-5.2": {
|
|
916
|
+
"id": "zai-org/GLM-5.2",
|
|
917
|
+
"name": "GLM-5.2",
|
|
918
|
+
"description": "Open flagship GLM for long-horizon coding agents and million-token context work",
|
|
919
|
+
"family": "glm",
|
|
920
|
+
"attachment": false,
|
|
445
921
|
"reasoning": true,
|
|
446
922
|
"reasoning_options": [
|
|
923
|
+
{
|
|
924
|
+
"type": "toggle"
|
|
925
|
+
},
|
|
447
926
|
{
|
|
448
927
|
"type": "effort",
|
|
449
928
|
"values": [
|
|
450
|
-
"none",
|
|
451
929
|
"low",
|
|
452
930
|
"medium",
|
|
453
|
-
"high"
|
|
931
|
+
"high",
|
|
932
|
+
"xhigh"
|
|
454
933
|
]
|
|
455
934
|
}
|
|
456
935
|
],
|
|
457
936
|
"tool_call": true,
|
|
937
|
+
"interleaved": {
|
|
938
|
+
"field": "reasoning_content"
|
|
939
|
+
},
|
|
458
940
|
"structured_output": true,
|
|
459
941
|
"temperature": true,
|
|
460
|
-
"release_date": "2026-
|
|
461
|
-
"last_updated": "2026-
|
|
942
|
+
"release_date": "2026-06-13",
|
|
943
|
+
"last_updated": "2026-06-13",
|
|
462
944
|
"modalities": {
|
|
463
945
|
"input": [
|
|
464
|
-
"text"
|
|
465
|
-
"image"
|
|
946
|
+
"text"
|
|
466
947
|
],
|
|
467
948
|
"output": [
|
|
468
949
|
"text"
|
|
469
950
|
]
|
|
470
951
|
},
|
|
471
|
-
"open_weights":
|
|
952
|
+
"open_weights": true,
|
|
472
953
|
"limit": {
|
|
473
|
-
"context":
|
|
474
|
-
"output":
|
|
954
|
+
"context": 1048576,
|
|
955
|
+
"output": 32768
|
|
475
956
|
},
|
|
476
957
|
"cost": {
|
|
477
|
-
"input": 0.
|
|
478
|
-
"output":
|
|
479
|
-
"cache_read": 0.
|
|
480
|
-
"tiers": [
|
|
481
|
-
{
|
|
482
|
-
"input": 1,
|
|
483
|
-
"output": 6,
|
|
484
|
-
"cache_read": 0.2,
|
|
485
|
-
"tier": {
|
|
486
|
-
"type": "context",
|
|
487
|
-
"size": 128000
|
|
488
|
-
}
|
|
489
|
-
}
|
|
490
|
-
]
|
|
958
|
+
"input": 0.75,
|
|
959
|
+
"output": 2.4,
|
|
960
|
+
"cache_read": 0.14
|
|
491
961
|
}
|
|
492
962
|
},
|
|
493
|
-
"
|
|
494
|
-
"id": "
|
|
495
|
-
"name": "
|
|
496
|
-
"description": "
|
|
497
|
-
"family": "
|
|
498
|
-
"attachment":
|
|
963
|
+
"zai-org/GLM-4.7-Flash": {
|
|
964
|
+
"id": "zai-org/GLM-4.7-Flash",
|
|
965
|
+
"name": "GLM-4.7-Flash",
|
|
966
|
+
"description": "Efficient GLM model for fast reasoning, coding, and agent workflows",
|
|
967
|
+
"family": "glm-flash",
|
|
968
|
+
"attachment": false,
|
|
499
969
|
"reasoning": true,
|
|
500
970
|
"reasoning_options": [],
|
|
501
971
|
"tool_call": true,
|
|
972
|
+
"interleaved": {
|
|
973
|
+
"field": "reasoning_content"
|
|
974
|
+
},
|
|
502
975
|
"structured_output": true,
|
|
503
976
|
"temperature": true,
|
|
504
|
-
"
|
|
505
|
-
"
|
|
977
|
+
"knowledge": "2025-04",
|
|
978
|
+
"release_date": "2026-01-19",
|
|
979
|
+
"last_updated": "2026-01-19",
|
|
506
980
|
"modalities": {
|
|
507
981
|
"input": [
|
|
508
|
-
"text"
|
|
509
|
-
"image"
|
|
982
|
+
"text"
|
|
510
983
|
],
|
|
511
984
|
"output": [
|
|
512
985
|
"text"
|
|
513
986
|
]
|
|
514
987
|
},
|
|
515
|
-
"open_weights":
|
|
988
|
+
"open_weights": true,
|
|
516
989
|
"limit": {
|
|
517
|
-
"context":
|
|
518
|
-
"output":
|
|
990
|
+
"context": 202752,
|
|
991
|
+
"output": 16384
|
|
519
992
|
},
|
|
520
993
|
"cost": {
|
|
521
|
-
"input": 0.
|
|
522
|
-
"output":
|
|
523
|
-
"cache_read": 0.
|
|
524
|
-
"tiers": [
|
|
525
|
-
{
|
|
526
|
-
"input": 1,
|
|
527
|
-
"output": 6,
|
|
528
|
-
"cache_read": 0.2,
|
|
529
|
-
"tier": {
|
|
530
|
-
"type": "context",
|
|
531
|
-
"size": 128000
|
|
532
|
-
}
|
|
533
|
-
}
|
|
534
|
-
]
|
|
994
|
+
"input": 0.06,
|
|
995
|
+
"output": 0.4,
|
|
996
|
+
"cache_read": 0.01
|
|
535
997
|
}
|
|
536
998
|
},
|
|
537
|
-
"
|
|
538
|
-
"id": "
|
|
539
|
-
"name": "
|
|
540
|
-
"description": "
|
|
541
|
-
"family": "
|
|
542
|
-
"attachment":
|
|
999
|
+
"zai-org/GLM-4.7": {
|
|
1000
|
+
"id": "zai-org/GLM-4.7",
|
|
1001
|
+
"name": "GLM-4.7",
|
|
1002
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1003
|
+
"family": "glm",
|
|
1004
|
+
"attachment": false,
|
|
543
1005
|
"reasoning": true,
|
|
544
|
-
"reasoning_options": [
|
|
1006
|
+
"reasoning_options": [
|
|
1007
|
+
{
|
|
1008
|
+
"type": "toggle"
|
|
1009
|
+
}
|
|
1010
|
+
],
|
|
545
1011
|
"tool_call": true,
|
|
1012
|
+
"interleaved": {
|
|
1013
|
+
"field": "reasoning_content"
|
|
1014
|
+
},
|
|
546
1015
|
"structured_output": true,
|
|
547
1016
|
"temperature": true,
|
|
548
|
-
"
|
|
549
|
-
"
|
|
1017
|
+
"knowledge": "2025-04",
|
|
1018
|
+
"release_date": "2025-12-22",
|
|
1019
|
+
"last_updated": "2025-12-22",
|
|
550
1020
|
"modalities": {
|
|
551
1021
|
"input": [
|
|
552
|
-
"text"
|
|
553
|
-
"image"
|
|
1022
|
+
"text"
|
|
554
1023
|
],
|
|
555
1024
|
"output": [
|
|
556
1025
|
"text"
|
|
557
1026
|
]
|
|
558
1027
|
},
|
|
559
|
-
"open_weights":
|
|
1028
|
+
"open_weights": true,
|
|
560
1029
|
"limit": {
|
|
561
|
-
"context":
|
|
562
|
-
"output":
|
|
1030
|
+
"context": 202752,
|
|
1031
|
+
"output": 16384
|
|
563
1032
|
},
|
|
564
1033
|
"cost": {
|
|
565
|
-
"input": 0.
|
|
566
|
-
"output":
|
|
567
|
-
"cache_read": 0.
|
|
568
|
-
"tiers": [
|
|
569
|
-
{
|
|
570
|
-
"input": 0.2,
|
|
571
|
-
"output": 0.8,
|
|
572
|
-
"cache_read": 0.2,
|
|
573
|
-
"tier": {
|
|
574
|
-
"type": "context",
|
|
575
|
-
"size": 128000
|
|
576
|
-
}
|
|
577
|
-
}
|
|
578
|
-
]
|
|
1034
|
+
"input": 0.4,
|
|
1035
|
+
"output": 1.75,
|
|
1036
|
+
"cache_read": 0.08
|
|
579
1037
|
}
|
|
580
|
-
},
|
|
581
|
-
"
|
|
582
|
-
"id": "
|
|
583
|
-
"name": "
|
|
584
|
-
"description": "
|
|
585
|
-
"family": "
|
|
586
|
-
"attachment":
|
|
1038
|
+
},
|
|
1039
|
+
"zai-org/GLM-5": {
|
|
1040
|
+
"id": "zai-org/GLM-5",
|
|
1041
|
+
"name": "GLM-5",
|
|
1042
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1043
|
+
"family": "glm",
|
|
1044
|
+
"attachment": false,
|
|
587
1045
|
"reasoning": true,
|
|
588
1046
|
"reasoning_options": [
|
|
589
1047
|
{
|
|
@@ -596,14 +1054,12 @@
|
|
|
596
1054
|
},
|
|
597
1055
|
"structured_output": true,
|
|
598
1056
|
"temperature": true,
|
|
599
|
-
"knowledge": "2025-
|
|
600
|
-
"release_date": "2026-
|
|
601
|
-
"last_updated": "2026-
|
|
1057
|
+
"knowledge": "2025-12",
|
|
1058
|
+
"release_date": "2026-02-12",
|
|
1059
|
+
"last_updated": "2026-02-12",
|
|
602
1060
|
"modalities": {
|
|
603
1061
|
"input": [
|
|
604
|
-
"text"
|
|
605
|
-
"image",
|
|
606
|
-
"video"
|
|
1062
|
+
"text"
|
|
607
1063
|
],
|
|
608
1064
|
"output": [
|
|
609
1065
|
"text"
|
|
@@ -611,21 +1067,21 @@
|
|
|
611
1067
|
},
|
|
612
1068
|
"open_weights": true,
|
|
613
1069
|
"limit": {
|
|
614
|
-
"context":
|
|
615
|
-
"output":
|
|
1070
|
+
"context": 202752,
|
|
1071
|
+
"output": 16384
|
|
616
1072
|
},
|
|
617
1073
|
"cost": {
|
|
618
|
-
"input": 0.
|
|
619
|
-
"output": 2.
|
|
620
|
-
"cache_read": 0.
|
|
1074
|
+
"input": 0.6,
|
|
1075
|
+
"output": 2.08,
|
|
1076
|
+
"cache_read": 0.12
|
|
621
1077
|
}
|
|
622
1078
|
},
|
|
623
|
-
"
|
|
624
|
-
"id": "
|
|
625
|
-
"name": "
|
|
626
|
-
"description": "
|
|
627
|
-
"family": "
|
|
628
|
-
"attachment":
|
|
1079
|
+
"zai-org/GLM-4.6": {
|
|
1080
|
+
"id": "zai-org/GLM-4.6",
|
|
1081
|
+
"name": "GLM-4.6",
|
|
1082
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1083
|
+
"family": "glm",
|
|
1084
|
+
"attachment": false,
|
|
629
1085
|
"reasoning": true,
|
|
630
1086
|
"reasoning_options": [
|
|
631
1087
|
{
|
|
@@ -637,15 +1093,13 @@
|
|
|
637
1093
|
"field": "reasoning_content"
|
|
638
1094
|
},
|
|
639
1095
|
"structured_output": true,
|
|
640
|
-
"temperature":
|
|
641
|
-
"knowledge": "2025-
|
|
642
|
-
"release_date": "
|
|
643
|
-
"last_updated": "
|
|
1096
|
+
"temperature": true,
|
|
1097
|
+
"knowledge": "2025-04",
|
|
1098
|
+
"release_date": "2025-09-30",
|
|
1099
|
+
"last_updated": "2025-09-30",
|
|
644
1100
|
"modalities": {
|
|
645
1101
|
"input": [
|
|
646
|
-
"text"
|
|
647
|
-
"image",
|
|
648
|
-
"video"
|
|
1102
|
+
"text"
|
|
649
1103
|
],
|
|
650
1104
|
"output": [
|
|
651
1105
|
"text"
|
|
@@ -653,20 +1107,20 @@
|
|
|
653
1107
|
},
|
|
654
1108
|
"open_weights": true,
|
|
655
1109
|
"limit": {
|
|
656
|
-
"context":
|
|
657
|
-
"output":
|
|
1110
|
+
"context": 202752,
|
|
1111
|
+
"output": 131072
|
|
658
1112
|
},
|
|
659
1113
|
"cost": {
|
|
660
|
-
"input": 0.
|
|
661
|
-
"output":
|
|
662
|
-
"cache_read": 0.
|
|
1114
|
+
"input": 0.5,
|
|
1115
|
+
"output": 2,
|
|
1116
|
+
"cache_read": 0.1
|
|
663
1117
|
}
|
|
664
1118
|
},
|
|
665
|
-
"
|
|
666
|
-
"id": "
|
|
667
|
-
"name": "
|
|
668
|
-
"description": "
|
|
669
|
-
"family": "
|
|
1119
|
+
"zai-org/GLM-5.3-Flash": {
|
|
1120
|
+
"id": "zai-org/GLM-5.3-Flash",
|
|
1121
|
+
"name": "GLM-5.3-Flash",
|
|
1122
|
+
"description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
|
|
1123
|
+
"family": "glm",
|
|
670
1124
|
"attachment": true,
|
|
671
1125
|
"reasoning": true,
|
|
672
1126
|
"reasoning_options": [
|
|
@@ -681,13 +1135,15 @@
|
|
|
681
1135
|
],
|
|
682
1136
|
"tool_call": true,
|
|
683
1137
|
"structured_output": true,
|
|
684
|
-
"temperature":
|
|
685
|
-
"release_date": "2026-
|
|
686
|
-
"last_updated": "2026-
|
|
1138
|
+
"temperature": true,
|
|
1139
|
+
"release_date": "2026-08-26",
|
|
1140
|
+
"last_updated": "2026-08-26",
|
|
687
1141
|
"modalities": {
|
|
688
1142
|
"input": [
|
|
689
1143
|
"text",
|
|
690
|
-
"image"
|
|
1144
|
+
"image",
|
|
1145
|
+
"video",
|
|
1146
|
+
"pdf"
|
|
691
1147
|
],
|
|
692
1148
|
"output": [
|
|
693
1149
|
"text"
|
|
@@ -699,78 +1155,29 @@
|
|
|
699
1155
|
"output": 131072
|
|
700
1156
|
},
|
|
701
1157
|
"cost": {
|
|
702
|
-
"input":
|
|
703
|
-
"output":
|
|
704
|
-
"cache_read": 0.
|
|
1158
|
+
"input": 0.15,
|
|
1159
|
+
"output": 0.5,
|
|
1160
|
+
"cache_read": 0.03
|
|
705
1161
|
}
|
|
706
1162
|
},
|
|
707
|
-
"
|
|
708
|
-
"id": "
|
|
709
|
-
"name": "
|
|
710
|
-
"description": "
|
|
711
|
-
"family": "
|
|
1163
|
+
"thinkingmachines/Inkling-Small": {
|
|
1164
|
+
"id": "thinkingmachines/Inkling-Small",
|
|
1165
|
+
"name": "Inkling Small",
|
|
1166
|
+
"description": "Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio",
|
|
1167
|
+
"family": "ling",
|
|
712
1168
|
"attachment": true,
|
|
713
1169
|
"reasoning": true,
|
|
714
|
-
"reasoning_options": [
|
|
715
|
-
{
|
|
716
|
-
"type": "toggle"
|
|
717
|
-
}
|
|
718
|
-
],
|
|
1170
|
+
"reasoning_options": [],
|
|
719
1171
|
"tool_call": true,
|
|
720
|
-
"interleaved": {
|
|
721
|
-
"field": "reasoning_content"
|
|
722
|
-
},
|
|
723
1172
|
"structured_output": true,
|
|
724
1173
|
"temperature": true,
|
|
725
|
-
"
|
|
726
|
-
"
|
|
727
|
-
"last_updated": "2026-04-21",
|
|
1174
|
+
"release_date": "2026-07-30",
|
|
1175
|
+
"last_updated": "2026-07-30",
|
|
728
1176
|
"modalities": {
|
|
729
1177
|
"input": [
|
|
730
1178
|
"text",
|
|
731
1179
|
"image",
|
|
732
|
-
"
|
|
733
|
-
],
|
|
734
|
-
"output": [
|
|
735
|
-
"text"
|
|
736
|
-
]
|
|
737
|
-
},
|
|
738
|
-
"open_weights": true,
|
|
739
|
-
"limit": {
|
|
740
|
-
"context": 262144,
|
|
741
|
-
"output": 16384
|
|
742
|
-
},
|
|
743
|
-
"cost": {
|
|
744
|
-
"input": 0.75,
|
|
745
|
-
"output": 3.5,
|
|
746
|
-
"cache_read": 0.15
|
|
747
|
-
}
|
|
748
|
-
},
|
|
749
|
-
"openai/gpt-oss-120b": {
|
|
750
|
-
"id": "openai/gpt-oss-120b",
|
|
751
|
-
"name": "GPT OSS 120B",
|
|
752
|
-
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
753
|
-
"family": "gpt-oss",
|
|
754
|
-
"attachment": false,
|
|
755
|
-
"reasoning": true,
|
|
756
|
-
"reasoning_options": [
|
|
757
|
-
{
|
|
758
|
-
"type": "effort",
|
|
759
|
-
"values": [
|
|
760
|
-
"low",
|
|
761
|
-
"medium",
|
|
762
|
-
"high"
|
|
763
|
-
]
|
|
764
|
-
}
|
|
765
|
-
],
|
|
766
|
-
"tool_call": true,
|
|
767
|
-
"structured_output": true,
|
|
768
|
-
"temperature": true,
|
|
769
|
-
"release_date": "2025-08-05",
|
|
770
|
-
"last_updated": "2025-08-05",
|
|
771
|
-
"modalities": {
|
|
772
|
-
"input": [
|
|
773
|
-
"text"
|
|
1180
|
+
"audio"
|
|
774
1181
|
],
|
|
775
1182
|
"output": [
|
|
776
1183
|
"text"
|
|
@@ -778,39 +1185,32 @@
|
|
|
778
1185
|
},
|
|
779
1186
|
"open_weights": true,
|
|
780
1187
|
"limit": {
|
|
781
|
-
"context":
|
|
782
|
-
"output":
|
|
1188
|
+
"context": 524288,
|
|
1189
|
+
"output": 1048576
|
|
783
1190
|
},
|
|
784
1191
|
"cost": {
|
|
785
|
-
"input": 0.
|
|
786
|
-
"output":
|
|
1192
|
+
"input": 0.45,
|
|
1193
|
+
"output": 1.2,
|
|
1194
|
+
"cache_read": 0.1
|
|
787
1195
|
}
|
|
788
1196
|
},
|
|
789
|
-
"
|
|
790
|
-
"id": "
|
|
791
|
-
"name": "
|
|
792
|
-
"description": "
|
|
793
|
-
"family": "
|
|
794
|
-
"attachment":
|
|
1197
|
+
"thinkingmachines/Inkling": {
|
|
1198
|
+
"id": "thinkingmachines/Inkling",
|
|
1199
|
+
"name": "Inkling",
|
|
1200
|
+
"description": "Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio",
|
|
1201
|
+
"family": "ling",
|
|
1202
|
+
"attachment": true,
|
|
795
1203
|
"reasoning": true,
|
|
796
|
-
"reasoning_options": [
|
|
797
|
-
{
|
|
798
|
-
"type": "effort",
|
|
799
|
-
"values": [
|
|
800
|
-
"low",
|
|
801
|
-
"medium",
|
|
802
|
-
"high"
|
|
803
|
-
]
|
|
804
|
-
}
|
|
805
|
-
],
|
|
1204
|
+
"reasoning_options": [],
|
|
806
1205
|
"tool_call": true,
|
|
807
|
-
"structured_output": true,
|
|
808
1206
|
"temperature": true,
|
|
809
|
-
"release_date": "
|
|
810
|
-
"last_updated": "
|
|
1207
|
+
"release_date": "2026-07-15",
|
|
1208
|
+
"last_updated": "2026-07-15",
|
|
811
1209
|
"modalities": {
|
|
812
1210
|
"input": [
|
|
813
|
-
"text"
|
|
1211
|
+
"text",
|
|
1212
|
+
"image",
|
|
1213
|
+
"audio"
|
|
814
1214
|
],
|
|
815
1215
|
"output": [
|
|
816
1216
|
"text"
|
|
@@ -818,44 +1218,27 @@
|
|
|
818
1218
|
},
|
|
819
1219
|
"open_weights": true,
|
|
820
1220
|
"limit": {
|
|
821
|
-
"context":
|
|
822
|
-
"output":
|
|
1221
|
+
"context": 524288,
|
|
1222
|
+
"output": 1048576
|
|
823
1223
|
},
|
|
824
1224
|
"cost": {
|
|
825
|
-
"input": 0.
|
|
826
|
-
"output":
|
|
1225
|
+
"input": 0.95,
|
|
1226
|
+
"output": 4.05,
|
|
1227
|
+
"cache_read": 0.16
|
|
827
1228
|
}
|
|
828
1229
|
},
|
|
829
|
-
"
|
|
830
|
-
"id": "
|
|
831
|
-
"name": "
|
|
832
|
-
"description": "
|
|
833
|
-
"family": "
|
|
1230
|
+
"Qwen/Qwen3.7-Max": {
|
|
1231
|
+
"id": "Qwen/Qwen3.7-Max",
|
|
1232
|
+
"name": "Qwen3.7 Max",
|
|
1233
|
+
"description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
|
|
1234
|
+
"family": "qwen",
|
|
834
1235
|
"attachment": false,
|
|
835
|
-
"reasoning":
|
|
836
|
-
"reasoning_options": [
|
|
837
|
-
{
|
|
838
|
-
"type": "toggle"
|
|
839
|
-
},
|
|
840
|
-
{
|
|
841
|
-
"type": "effort",
|
|
842
|
-
"values": [
|
|
843
|
-
"low",
|
|
844
|
-
"medium",
|
|
845
|
-
"high",
|
|
846
|
-
"xhigh"
|
|
847
|
-
]
|
|
848
|
-
}
|
|
849
|
-
],
|
|
1236
|
+
"reasoning": false,
|
|
850
1237
|
"tool_call": true,
|
|
851
|
-
"interleaved": {
|
|
852
|
-
"field": "reasoning_content"
|
|
853
|
-
},
|
|
854
1238
|
"structured_output": true,
|
|
855
1239
|
"temperature": true,
|
|
856
|
-
"
|
|
857
|
-
"
|
|
858
|
-
"last_updated": "2026-04-24",
|
|
1240
|
+
"release_date": "2026-05-21",
|
|
1241
|
+
"last_updated": "2026-05-21",
|
|
859
1242
|
"modalities": {
|
|
860
1243
|
"input": [
|
|
861
1244
|
"text"
|
|
@@ -864,38 +1247,66 @@
|
|
|
864
1247
|
"text"
|
|
865
1248
|
]
|
|
866
1249
|
},
|
|
867
|
-
"open_weights":
|
|
1250
|
+
"open_weights": false,
|
|
868
1251
|
"limit": {
|
|
869
|
-
"context":
|
|
870
|
-
"output":
|
|
1252
|
+
"context": 256000,
|
|
1253
|
+
"output": 65536
|
|
871
1254
|
},
|
|
872
1255
|
"cost": {
|
|
873
|
-
"input":
|
|
874
|
-
"output":
|
|
875
|
-
"cache_read": 0.
|
|
1256
|
+
"input": 2.5,
|
|
1257
|
+
"output": 7.5,
|
|
1258
|
+
"cache_read": 0.5,
|
|
1259
|
+
"tiers": [
|
|
1260
|
+
{
|
|
1261
|
+
"input": 5,
|
|
1262
|
+
"output": 15,
|
|
1263
|
+
"cache_read": 1,
|
|
1264
|
+
"tier": {
|
|
1265
|
+
"type": "context",
|
|
1266
|
+
"size": 32000
|
|
1267
|
+
}
|
|
1268
|
+
},
|
|
1269
|
+
{
|
|
1270
|
+
"input": 6.25,
|
|
1271
|
+
"output": 18.5,
|
|
1272
|
+
"cache_read": 1.25,
|
|
1273
|
+
"tier": {
|
|
1274
|
+
"type": "context",
|
|
1275
|
+
"size": 128000
|
|
1276
|
+
}
|
|
1277
|
+
}
|
|
1278
|
+
]
|
|
876
1279
|
}
|
|
877
1280
|
},
|
|
878
|
-
"
|
|
879
|
-
"id": "
|
|
880
|
-
"name": "
|
|
881
|
-
"description": "
|
|
882
|
-
"family": "
|
|
883
|
-
"attachment":
|
|
1281
|
+
"Qwen/Qwen3.8-27B": {
|
|
1282
|
+
"id": "Qwen/Qwen3.8-27B",
|
|
1283
|
+
"name": "Qwen3.8 27B",
|
|
1284
|
+
"description": "Dense 27B vision-language model for coding, agent tasks, and image and video understanding",
|
|
1285
|
+
"family": "qwen",
|
|
1286
|
+
"attachment": true,
|
|
884
1287
|
"reasoning": true,
|
|
885
1288
|
"reasoning_options": [
|
|
886
1289
|
{
|
|
887
1290
|
"type": "toggle"
|
|
1291
|
+
},
|
|
1292
|
+
{
|
|
1293
|
+
"type": "effort",
|
|
1294
|
+
"values": [
|
|
1295
|
+
"low",
|
|
1296
|
+
"medium",
|
|
1297
|
+
"xhigh"
|
|
1298
|
+
]
|
|
888
1299
|
}
|
|
889
1300
|
],
|
|
890
1301
|
"tool_call": true,
|
|
891
1302
|
"structured_output": true,
|
|
892
1303
|
"temperature": true,
|
|
893
|
-
"
|
|
894
|
-
"
|
|
895
|
-
"last_updated": "2026-07-31",
|
|
1304
|
+
"release_date": "2026-08-14",
|
|
1305
|
+
"last_updated": "2026-08-14",
|
|
896
1306
|
"modalities": {
|
|
897
1307
|
"input": [
|
|
898
|
-
"text"
|
|
1308
|
+
"text",
|
|
1309
|
+
"image"
|
|
899
1310
|
],
|
|
900
1311
|
"output": [
|
|
901
1312
|
"text"
|
|
@@ -903,70 +1314,62 @@
|
|
|
903
1314
|
},
|
|
904
1315
|
"open_weights": true,
|
|
905
1316
|
"limit": {
|
|
906
|
-
"context":
|
|
907
|
-
"output":
|
|
1317
|
+
"context": 262144,
|
|
1318
|
+
"output": 32768
|
|
908
1319
|
},
|
|
909
1320
|
"cost": {
|
|
910
|
-
"input": 0.
|
|
911
|
-
"output":
|
|
912
|
-
"cache_read": 0.
|
|
1321
|
+
"input": 0.4,
|
|
1322
|
+
"output": 3,
|
|
1323
|
+
"cache_read": 0.04
|
|
913
1324
|
}
|
|
914
1325
|
},
|
|
915
|
-
"
|
|
916
|
-
"id": "
|
|
917
|
-
"name": "
|
|
918
|
-
"description": "
|
|
919
|
-
"
|
|
1326
|
+
"Qwen/Qwen3.5-27B": {
|
|
1327
|
+
"id": "Qwen/Qwen3.5-27B",
|
|
1328
|
+
"name": "Qwen3.5 27B",
|
|
1329
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1330
|
+
"family": "qwen",
|
|
1331
|
+
"attachment": true,
|
|
920
1332
|
"reasoning": true,
|
|
921
|
-
"reasoning_options": [
|
|
922
|
-
{
|
|
923
|
-
"type": "toggle"
|
|
924
|
-
}
|
|
925
|
-
],
|
|
1333
|
+
"reasoning_options": [],
|
|
926
1334
|
"tool_call": true,
|
|
927
|
-
"interleaved": {
|
|
928
|
-
"field": "reasoning_content"
|
|
929
|
-
},
|
|
930
1335
|
"structured_output": true,
|
|
931
1336
|
"temperature": true,
|
|
932
|
-
"
|
|
933
|
-
"
|
|
934
|
-
"last_updated": "2025-12-02",
|
|
1337
|
+
"release_date": "2026-02-23",
|
|
1338
|
+
"last_updated": "2026-02-23",
|
|
935
1339
|
"modalities": {
|
|
936
1340
|
"input": [
|
|
937
|
-
"text"
|
|
1341
|
+
"text",
|
|
1342
|
+
"image",
|
|
1343
|
+
"video",
|
|
1344
|
+
"audio"
|
|
938
1345
|
],
|
|
939
1346
|
"output": [
|
|
940
1347
|
"text"
|
|
941
1348
|
]
|
|
942
1349
|
},
|
|
943
|
-
"open_weights":
|
|
1350
|
+
"open_weights": true,
|
|
944
1351
|
"limit": {
|
|
945
|
-
"context":
|
|
946
|
-
"output":
|
|
1352
|
+
"context": 262144,
|
|
1353
|
+
"output": 65536
|
|
947
1354
|
},
|
|
948
1355
|
"cost": {
|
|
949
1356
|
"input": 0.26,
|
|
950
|
-
"output":
|
|
951
|
-
"cache_read": 0.13
|
|
1357
|
+
"output": 2.6
|
|
952
1358
|
}
|
|
953
1359
|
},
|
|
954
|
-
"
|
|
955
|
-
"id": "
|
|
956
|
-
"name": "
|
|
957
|
-
"description": "
|
|
1360
|
+
"Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": {
|
|
1361
|
+
"id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo",
|
|
1362
|
+
"name": "Qwen3 Coder 480B A35B Instruct Turbo",
|
|
1363
|
+
"description": "Qwen coding model for software agents, repository edits, and code reasoning",
|
|
1364
|
+
"family": "qwen",
|
|
958
1365
|
"attachment": false,
|
|
959
|
-
"reasoning":
|
|
960
|
-
"reasoning_options": [],
|
|
1366
|
+
"reasoning": false,
|
|
961
1367
|
"tool_call": true,
|
|
962
|
-
"interleaved": {
|
|
963
|
-
"field": "reasoning_content"
|
|
964
|
-
},
|
|
965
1368
|
"structured_output": true,
|
|
966
1369
|
"temperature": true,
|
|
967
|
-
"knowledge": "
|
|
968
|
-
"release_date": "2025-
|
|
969
|
-
"last_updated": "2025-
|
|
1370
|
+
"knowledge": "2025-04",
|
|
1371
|
+
"release_date": "2025-07-23",
|
|
1372
|
+
"last_updated": "2025-07-23",
|
|
970
1373
|
"modalities": {
|
|
971
1374
|
"input": [
|
|
972
1375
|
"text"
|
|
@@ -975,68 +1378,69 @@
|
|
|
975
1378
|
"text"
|
|
976
1379
|
]
|
|
977
1380
|
},
|
|
978
|
-
"open_weights":
|
|
1381
|
+
"open_weights": true,
|
|
979
1382
|
"limit": {
|
|
980
|
-
"context":
|
|
981
|
-
"output":
|
|
1383
|
+
"context": 262144,
|
|
1384
|
+
"output": 66536
|
|
982
1385
|
},
|
|
983
1386
|
"cost": {
|
|
984
|
-
"input": 0.
|
|
985
|
-
"output":
|
|
986
|
-
"cache_read": 0.
|
|
1387
|
+
"input": 0.3,
|
|
1388
|
+
"output": 1,
|
|
1389
|
+
"cache_read": 0.1
|
|
987
1390
|
}
|
|
988
1391
|
},
|
|
989
|
-
"
|
|
990
|
-
"id": "
|
|
991
|
-
"name": "
|
|
992
|
-
"description": "
|
|
993
|
-
"family": "
|
|
994
|
-
"attachment":
|
|
995
|
-
"reasoning":
|
|
996
|
-
"reasoning_options": [
|
|
997
|
-
{
|
|
998
|
-
"type": "toggle"
|
|
999
|
-
}
|
|
1000
|
-
],
|
|
1392
|
+
"Qwen/Qwen3.8-Max": {
|
|
1393
|
+
"id": "Qwen/Qwen3.8-Max",
|
|
1394
|
+
"name": "Qwen3.8 Max",
|
|
1395
|
+
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
|
|
1396
|
+
"family": "qwen",
|
|
1397
|
+
"attachment": true,
|
|
1398
|
+
"reasoning": false,
|
|
1001
1399
|
"tool_call": true,
|
|
1002
1400
|
"structured_output": true,
|
|
1003
1401
|
"temperature": true,
|
|
1004
|
-
"release_date": "
|
|
1005
|
-
"last_updated": "
|
|
1402
|
+
"release_date": "2026-08-03",
|
|
1403
|
+
"last_updated": "2026-08-03",
|
|
1006
1404
|
"modalities": {
|
|
1007
1405
|
"input": [
|
|
1008
|
-
"text"
|
|
1406
|
+
"text",
|
|
1407
|
+
"image",
|
|
1408
|
+
"video",
|
|
1409
|
+
"pdf"
|
|
1009
1410
|
],
|
|
1010
1411
|
"output": [
|
|
1011
1412
|
"text"
|
|
1012
1413
|
]
|
|
1013
1414
|
},
|
|
1014
|
-
"open_weights":
|
|
1415
|
+
"open_weights": false,
|
|
1015
1416
|
"limit": {
|
|
1016
|
-
"context":
|
|
1017
|
-
"output":
|
|
1417
|
+
"context": 256000,
|
|
1418
|
+
"output": 131072
|
|
1018
1419
|
},
|
|
1019
1420
|
"cost": {
|
|
1020
|
-
"input":
|
|
1021
|
-
"output":
|
|
1022
|
-
"cache_read": 0.
|
|
1421
|
+
"input": 1.65,
|
|
1422
|
+
"output": 4.951,
|
|
1423
|
+
"cache_read": 0.206
|
|
1023
1424
|
}
|
|
1024
1425
|
},
|
|
1025
|
-
"
|
|
1026
|
-
"id": "
|
|
1027
|
-
"name": "
|
|
1028
|
-
"description": "
|
|
1029
|
-
"family": "
|
|
1030
|
-
"attachment":
|
|
1031
|
-
"reasoning":
|
|
1426
|
+
"Qwen/Qwen3.5-9B": {
|
|
1427
|
+
"id": "Qwen/Qwen3.5-9B",
|
|
1428
|
+
"name": "Qwen3.5 9B",
|
|
1429
|
+
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1430
|
+
"family": "qwen",
|
|
1431
|
+
"attachment": true,
|
|
1432
|
+
"reasoning": true,
|
|
1433
|
+
"reasoning_options": [],
|
|
1032
1434
|
"tool_call": true,
|
|
1033
1435
|
"structured_output": true,
|
|
1034
1436
|
"temperature": true,
|
|
1035
|
-
"release_date": "
|
|
1036
|
-
"last_updated": "
|
|
1437
|
+
"release_date": "2026-02-23",
|
|
1438
|
+
"last_updated": "2026-02-23",
|
|
1037
1439
|
"modalities": {
|
|
1038
1440
|
"input": [
|
|
1039
|
-
"text"
|
|
1441
|
+
"text",
|
|
1442
|
+
"image",
|
|
1443
|
+
"video"
|
|
1040
1444
|
],
|
|
1041
1445
|
"output": [
|
|
1042
1446
|
"text"
|
|
@@ -1044,27 +1448,27 @@
|
|
|
1044
1448
|
},
|
|
1045
1449
|
"open_weights": true,
|
|
1046
1450
|
"limit": {
|
|
1047
|
-
"context":
|
|
1048
|
-
"output":
|
|
1451
|
+
"context": 262144,
|
|
1452
|
+
"output": 65536
|
|
1049
1453
|
},
|
|
1050
1454
|
"cost": {
|
|
1051
|
-
"input": 0.
|
|
1052
|
-
"output": 0.
|
|
1053
|
-
"cache_read": 0.135
|
|
1455
|
+
"input": 0.1,
|
|
1456
|
+
"output": 0.15
|
|
1054
1457
|
}
|
|
1055
1458
|
},
|
|
1056
|
-
"
|
|
1057
|
-
"id": "
|
|
1058
|
-
"name": "
|
|
1059
|
-
"description": "
|
|
1060
|
-
"family": "
|
|
1459
|
+
"Qwen/Qwen3-30B-A3B": {
|
|
1460
|
+
"id": "Qwen/Qwen3-30B-A3B",
|
|
1461
|
+
"name": "Qwen3 30B A3B",
|
|
1462
|
+
"description": "Sparse MoE Qwen model with 3B active parameters for efficient chat and reasoning",
|
|
1463
|
+
"family": "qwen",
|
|
1061
1464
|
"attachment": false,
|
|
1062
|
-
"reasoning":
|
|
1465
|
+
"reasoning": true,
|
|
1466
|
+
"reasoning_options": [],
|
|
1063
1467
|
"tool_call": true,
|
|
1064
1468
|
"structured_output": true,
|
|
1065
1469
|
"temperature": true,
|
|
1066
|
-
"release_date": "
|
|
1067
|
-
"last_updated": "
|
|
1470
|
+
"release_date": "2025-04-28",
|
|
1471
|
+
"last_updated": "2025-04-28",
|
|
1068
1472
|
"modalities": {
|
|
1069
1473
|
"input": [
|
|
1070
1474
|
"text"
|
|
@@ -1075,47 +1479,33 @@
|
|
|
1075
1479
|
},
|
|
1076
1480
|
"open_weights": true,
|
|
1077
1481
|
"limit": {
|
|
1078
|
-
"context":
|
|
1079
|
-
"output":
|
|
1482
|
+
"context": 40960,
|
|
1483
|
+
"output": 16384
|
|
1080
1484
|
},
|
|
1081
1485
|
"cost": {
|
|
1082
|
-
"input": 0.
|
|
1083
|
-
"output": 0.
|
|
1486
|
+
"input": 0.12,
|
|
1487
|
+
"output": 0.5
|
|
1084
1488
|
}
|
|
1085
1489
|
},
|
|
1086
|
-
"
|
|
1087
|
-
"id": "
|
|
1088
|
-
"name": "
|
|
1089
|
-
"description": "
|
|
1090
|
-
"family": "
|
|
1091
|
-
"attachment":
|
|
1490
|
+
"Qwen/Qwen3.5-122B-A10B": {
|
|
1491
|
+
"id": "Qwen/Qwen3.5-122B-A10B",
|
|
1492
|
+
"name": "Qwen3.5 122B-A10B",
|
|
1493
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1494
|
+
"family": "qwen",
|
|
1495
|
+
"attachment": true,
|
|
1092
1496
|
"reasoning": true,
|
|
1093
|
-
"reasoning_options": [
|
|
1094
|
-
{
|
|
1095
|
-
"type": "toggle"
|
|
1096
|
-
},
|
|
1097
|
-
{
|
|
1098
|
-
"type": "effort",
|
|
1099
|
-
"values": [
|
|
1100
|
-
"low",
|
|
1101
|
-
"medium",
|
|
1102
|
-
"high",
|
|
1103
|
-
"xhigh"
|
|
1104
|
-
]
|
|
1105
|
-
}
|
|
1106
|
-
],
|
|
1497
|
+
"reasoning_options": [],
|
|
1107
1498
|
"tool_call": true,
|
|
1108
|
-
"interleaved": {
|
|
1109
|
-
"field": "reasoning_content"
|
|
1110
|
-
},
|
|
1111
1499
|
"structured_output": true,
|
|
1112
1500
|
"temperature": true,
|
|
1113
|
-
"
|
|
1114
|
-
"
|
|
1115
|
-
"last_updated": "2026-04-24",
|
|
1501
|
+
"release_date": "2026-02-23",
|
|
1502
|
+
"last_updated": "2026-02-23",
|
|
1116
1503
|
"modalities": {
|
|
1117
1504
|
"input": [
|
|
1118
|
-
"text"
|
|
1505
|
+
"text",
|
|
1506
|
+
"image",
|
|
1507
|
+
"video",
|
|
1508
|
+
"audio"
|
|
1119
1509
|
],
|
|
1120
1510
|
"output": [
|
|
1121
1511
|
"text"
|
|
@@ -1123,32 +1513,39 @@
|
|
|
1123
1513
|
},
|
|
1124
1514
|
"open_weights": true,
|
|
1125
1515
|
"limit": {
|
|
1126
|
-
"context":
|
|
1127
|
-
"output":
|
|
1128
|
-
},
|
|
1129
|
-
"cost": {
|
|
1130
|
-
"input": 0.
|
|
1131
|
-
"output":
|
|
1132
|
-
"cache_read": 0.018
|
|
1516
|
+
"context": 262144,
|
|
1517
|
+
"output": 65536
|
|
1518
|
+
},
|
|
1519
|
+
"cost": {
|
|
1520
|
+
"input": 0.29,
|
|
1521
|
+
"output": 2.4
|
|
1133
1522
|
}
|
|
1134
1523
|
},
|
|
1135
|
-
"
|
|
1136
|
-
"id": "
|
|
1137
|
-
"name": "
|
|
1138
|
-
"description": "
|
|
1139
|
-
"
|
|
1524
|
+
"Qwen/Qwen3.8-2.4T-A95B": {
|
|
1525
|
+
"id": "Qwen/Qwen3.8-2.4T-A95B",
|
|
1526
|
+
"name": "Qwen3.8 2.4T A95B",
|
|
1527
|
+
"description": "Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows",
|
|
1528
|
+
"family": "qwen",
|
|
1529
|
+
"attachment": false,
|
|
1140
1530
|
"reasoning": true,
|
|
1141
|
-
"reasoning_options": [
|
|
1531
|
+
"reasoning_options": [
|
|
1532
|
+
{
|
|
1533
|
+
"type": "effort",
|
|
1534
|
+
"values": [
|
|
1535
|
+
"low",
|
|
1536
|
+
"medium",
|
|
1537
|
+
"xhigh"
|
|
1538
|
+
]
|
|
1539
|
+
}
|
|
1540
|
+
],
|
|
1142
1541
|
"tool_call": true,
|
|
1542
|
+
"structured_output": true,
|
|
1143
1543
|
"temperature": true,
|
|
1144
|
-
"
|
|
1145
|
-
"
|
|
1146
|
-
"last_updated": "2026-05-29",
|
|
1544
|
+
"release_date": "2026-08-12",
|
|
1545
|
+
"last_updated": "2026-08-12",
|
|
1147
1546
|
"modalities": {
|
|
1148
1547
|
"input": [
|
|
1149
|
-
"text"
|
|
1150
|
-
"image",
|
|
1151
|
-
"video"
|
|
1548
|
+
"text"
|
|
1152
1549
|
],
|
|
1153
1550
|
"output": [
|
|
1154
1551
|
"text"
|
|
@@ -1157,12 +1554,12 @@
|
|
|
1157
1554
|
"open_weights": true,
|
|
1158
1555
|
"limit": {
|
|
1159
1556
|
"context": 262144,
|
|
1160
|
-
"output":
|
|
1557
|
+
"output": 131072
|
|
1161
1558
|
},
|
|
1162
1559
|
"cost": {
|
|
1163
|
-
"input":
|
|
1164
|
-
"output":
|
|
1165
|
-
"cache_read": 0.
|
|
1560
|
+
"input": 2,
|
|
1561
|
+
"output": 6,
|
|
1562
|
+
"cache_read": 0.2
|
|
1166
1563
|
}
|
|
1167
1564
|
},
|
|
1168
1565
|
"Qwen/Qwen3-235B-A22B-Instruct-2507": {
|
|
@@ -1195,25 +1592,23 @@
|
|
|
1195
1592
|
"output": 0.55
|
|
1196
1593
|
}
|
|
1197
1594
|
},
|
|
1198
|
-
"Qwen/Qwen3
|
|
1199
|
-
"id": "Qwen/Qwen3
|
|
1200
|
-
"name": "Qwen3
|
|
1201
|
-
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1595
|
+
"Qwen/Qwen3-VL-235B-A22B-Instruct": {
|
|
1596
|
+
"id": "Qwen/Qwen3-VL-235B-A22B-Instruct",
|
|
1597
|
+
"name": "Qwen3 VL 235B A22B Instruct",
|
|
1598
|
+
"description": "Qwen vision-language instruct model for visual reasoning, documents, and agent tasks",
|
|
1202
1599
|
"family": "qwen",
|
|
1203
1600
|
"attachment": true,
|
|
1204
|
-
"reasoning":
|
|
1205
|
-
"reasoning_options": [],
|
|
1601
|
+
"reasoning": false,
|
|
1206
1602
|
"tool_call": true,
|
|
1207
1603
|
"structured_output": true,
|
|
1208
1604
|
"temperature": true,
|
|
1209
|
-
"
|
|
1210
|
-
"
|
|
1605
|
+
"knowledge": "2025-03-31",
|
|
1606
|
+
"release_date": "2025-09-23",
|
|
1607
|
+
"last_updated": "2025-09-23",
|
|
1211
1608
|
"modalities": {
|
|
1212
1609
|
"input": [
|
|
1213
1610
|
"text",
|
|
1214
|
-
"image"
|
|
1215
|
-
"video",
|
|
1216
|
-
"audio"
|
|
1611
|
+
"image"
|
|
1217
1612
|
],
|
|
1218
1613
|
"output": [
|
|
1219
1614
|
"text"
|
|
@@ -1222,31 +1617,30 @@
|
|
|
1222
1617
|
"open_weights": true,
|
|
1223
1618
|
"limit": {
|
|
1224
1619
|
"context": 262144,
|
|
1225
|
-
"output":
|
|
1620
|
+
"output": 32768
|
|
1226
1621
|
},
|
|
1227
1622
|
"cost": {
|
|
1228
|
-
"input": 0.
|
|
1229
|
-
"output":
|
|
1623
|
+
"input": 0.2,
|
|
1624
|
+
"output": 0.88,
|
|
1625
|
+
"cache_read": 0.11
|
|
1230
1626
|
}
|
|
1231
1627
|
},
|
|
1232
|
-
"Qwen/Qwen3
|
|
1233
|
-
"id": "Qwen/Qwen3
|
|
1234
|
-
"name": "Qwen3
|
|
1628
|
+
"Qwen/Qwen3-Next-80B-A3B-Instruct": {
|
|
1629
|
+
"id": "Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
1630
|
+
"name": "Qwen3-Next 80B-A3B Instruct",
|
|
1235
1631
|
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1236
1632
|
"family": "qwen",
|
|
1237
|
-
"attachment":
|
|
1238
|
-
"reasoning":
|
|
1239
|
-
"reasoning_options": [],
|
|
1633
|
+
"attachment": false,
|
|
1634
|
+
"reasoning": false,
|
|
1240
1635
|
"tool_call": true,
|
|
1241
1636
|
"structured_output": true,
|
|
1242
1637
|
"temperature": true,
|
|
1243
|
-
"
|
|
1244
|
-
"
|
|
1638
|
+
"knowledge": "2025-04",
|
|
1639
|
+
"release_date": "2025-09",
|
|
1640
|
+
"last_updated": "2025-09",
|
|
1245
1641
|
"modalities": {
|
|
1246
1642
|
"input": [
|
|
1247
|
-
"text"
|
|
1248
|
-
"image",
|
|
1249
|
-
"video"
|
|
1643
|
+
"text"
|
|
1250
1644
|
],
|
|
1251
1645
|
"output": [
|
|
1252
1646
|
"text"
|
|
@@ -1255,29 +1649,32 @@
|
|
|
1255
1649
|
"open_weights": true,
|
|
1256
1650
|
"limit": {
|
|
1257
1651
|
"context": 262144,
|
|
1258
|
-
"output":
|
|
1652
|
+
"output": 32768
|
|
1259
1653
|
},
|
|
1260
1654
|
"cost": {
|
|
1261
|
-
"input": 0.
|
|
1262
|
-
"output":
|
|
1655
|
+
"input": 0.09,
|
|
1656
|
+
"output": 1.1
|
|
1263
1657
|
}
|
|
1264
1658
|
},
|
|
1265
|
-
"Qwen/Qwen3-
|
|
1266
|
-
"id": "Qwen/Qwen3-
|
|
1267
|
-
"name": "
|
|
1268
|
-
"description": "Qwen
|
|
1659
|
+
"Qwen/Qwen3.5-397B-A17B": {
|
|
1660
|
+
"id": "Qwen/Qwen3.5-397B-A17B",
|
|
1661
|
+
"name": "Qwen 3.5 397B A17B",
|
|
1662
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1269
1663
|
"family": "qwen",
|
|
1270
|
-
"attachment":
|
|
1271
|
-
"reasoning":
|
|
1664
|
+
"attachment": true,
|
|
1665
|
+
"reasoning": true,
|
|
1666
|
+
"reasoning_options": [],
|
|
1272
1667
|
"tool_call": true,
|
|
1273
1668
|
"structured_output": true,
|
|
1274
1669
|
"temperature": true,
|
|
1275
|
-
"knowledge": "2025-
|
|
1276
|
-
"release_date": "
|
|
1277
|
-
"last_updated": "
|
|
1670
|
+
"knowledge": "2025-01",
|
|
1671
|
+
"release_date": "2026-02-01",
|
|
1672
|
+
"last_updated": "2026-04-20",
|
|
1278
1673
|
"modalities": {
|
|
1279
1674
|
"input": [
|
|
1280
|
-
"text"
|
|
1675
|
+
"text",
|
|
1676
|
+
"image",
|
|
1677
|
+
"video"
|
|
1281
1678
|
],
|
|
1282
1679
|
"output": [
|
|
1283
1680
|
"text"
|
|
@@ -1286,17 +1683,17 @@
|
|
|
1286
1683
|
"open_weights": true,
|
|
1287
1684
|
"limit": {
|
|
1288
1685
|
"context": 262144,
|
|
1289
|
-
"output":
|
|
1686
|
+
"output": 81920
|
|
1290
1687
|
},
|
|
1291
1688
|
"cost": {
|
|
1292
|
-
"input": 0.
|
|
1293
|
-
"output":
|
|
1294
|
-
"cache_read": 0.
|
|
1689
|
+
"input": 0.45,
|
|
1690
|
+
"output": 3,
|
|
1691
|
+
"cache_read": 0.22
|
|
1295
1692
|
}
|
|
1296
1693
|
},
|
|
1297
|
-
"Qwen/Qwen3.5-
|
|
1298
|
-
"id": "Qwen/Qwen3.5-
|
|
1299
|
-
"name": "
|
|
1694
|
+
"Qwen/Qwen3.5-35B-A3B": {
|
|
1695
|
+
"id": "Qwen/Qwen3.5-35B-A3B",
|
|
1696
|
+
"name": "Qwen 3.5 35B A3B",
|
|
1300
1697
|
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1301
1698
|
"family": "qwen",
|
|
1302
1699
|
"attachment": true,
|
|
@@ -1305,14 +1702,14 @@
|
|
|
1305
1702
|
"tool_call": true,
|
|
1306
1703
|
"structured_output": true,
|
|
1307
1704
|
"temperature": true,
|
|
1308
|
-
"
|
|
1309
|
-
"
|
|
1705
|
+
"knowledge": "2025-01",
|
|
1706
|
+
"release_date": "2026-02-01",
|
|
1707
|
+
"last_updated": "2026-04-20",
|
|
1310
1708
|
"modalities": {
|
|
1311
1709
|
"input": [
|
|
1312
1710
|
"text",
|
|
1313
1711
|
"image",
|
|
1314
|
-
"video"
|
|
1315
|
-
"audio"
|
|
1712
|
+
"video"
|
|
1316
1713
|
],
|
|
1317
1714
|
"output": [
|
|
1318
1715
|
"text"
|
|
@@ -1321,11 +1718,12 @@
|
|
|
1321
1718
|
"open_weights": true,
|
|
1322
1719
|
"limit": {
|
|
1323
1720
|
"context": 262144,
|
|
1324
|
-
"output":
|
|
1721
|
+
"output": 81920
|
|
1325
1722
|
},
|
|
1326
1723
|
"cost": {
|
|
1327
|
-
"input": 0.
|
|
1328
|
-
"output":
|
|
1724
|
+
"input": 0.14,
|
|
1725
|
+
"output": 1,
|
|
1726
|
+
"cache_read": 0.05
|
|
1329
1727
|
}
|
|
1330
1728
|
},
|
|
1331
1729
|
"Qwen/Qwen3-32B": {
|
|
@@ -1360,94 +1758,9 @@
|
|
|
1360
1758
|
"output": 0.28
|
|
1361
1759
|
}
|
|
1362
1760
|
},
|
|
1363
|
-
"Qwen/Qwen3.
|
|
1364
|
-
"id": "Qwen/Qwen3.
|
|
1365
|
-
"name": "Qwen3.
|
|
1366
|
-
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
|
|
1367
|
-
"family": "qwen",
|
|
1368
|
-
"attachment": true,
|
|
1369
|
-
"reasoning": false,
|
|
1370
|
-
"tool_call": true,
|
|
1371
|
-
"structured_output": true,
|
|
1372
|
-
"temperature": true,
|
|
1373
|
-
"release_date": "2026-08-03",
|
|
1374
|
-
"last_updated": "2026-08-03",
|
|
1375
|
-
"modalities": {
|
|
1376
|
-
"input": [
|
|
1377
|
-
"text",
|
|
1378
|
-
"image",
|
|
1379
|
-
"video",
|
|
1380
|
-
"pdf"
|
|
1381
|
-
],
|
|
1382
|
-
"output": [
|
|
1383
|
-
"text"
|
|
1384
|
-
]
|
|
1385
|
-
},
|
|
1386
|
-
"open_weights": false,
|
|
1387
|
-
"limit": {
|
|
1388
|
-
"context": 256000,
|
|
1389
|
-
"output": 131072
|
|
1390
|
-
},
|
|
1391
|
-
"cost": {
|
|
1392
|
-
"input": 1.65,
|
|
1393
|
-
"output": 4.951,
|
|
1394
|
-
"cache_read": 0.206
|
|
1395
|
-
}
|
|
1396
|
-
},
|
|
1397
|
-
"Qwen/Qwen3.7-Max": {
|
|
1398
|
-
"id": "Qwen/Qwen3.7-Max",
|
|
1399
|
-
"name": "Qwen3.7 Max",
|
|
1400
|
-
"description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
|
|
1401
|
-
"family": "qwen",
|
|
1402
|
-
"attachment": false,
|
|
1403
|
-
"reasoning": false,
|
|
1404
|
-
"tool_call": true,
|
|
1405
|
-
"structured_output": true,
|
|
1406
|
-
"temperature": true,
|
|
1407
|
-
"release_date": "2026-05-21",
|
|
1408
|
-
"last_updated": "2026-05-21",
|
|
1409
|
-
"modalities": {
|
|
1410
|
-
"input": [
|
|
1411
|
-
"text"
|
|
1412
|
-
],
|
|
1413
|
-
"output": [
|
|
1414
|
-
"text"
|
|
1415
|
-
]
|
|
1416
|
-
},
|
|
1417
|
-
"open_weights": false,
|
|
1418
|
-
"limit": {
|
|
1419
|
-
"context": 256000,
|
|
1420
|
-
"output": 65536
|
|
1421
|
-
},
|
|
1422
|
-
"cost": {
|
|
1423
|
-
"input": 2.5,
|
|
1424
|
-
"output": 7.5,
|
|
1425
|
-
"cache_read": 0.5,
|
|
1426
|
-
"tiers": [
|
|
1427
|
-
{
|
|
1428
|
-
"input": 5,
|
|
1429
|
-
"output": 15,
|
|
1430
|
-
"cache_read": 1,
|
|
1431
|
-
"tier": {
|
|
1432
|
-
"type": "context",
|
|
1433
|
-
"size": 32000
|
|
1434
|
-
}
|
|
1435
|
-
},
|
|
1436
|
-
{
|
|
1437
|
-
"input": 6.25,
|
|
1438
|
-
"output": 18.5,
|
|
1439
|
-
"cache_read": 1.25,
|
|
1440
|
-
"tier": {
|
|
1441
|
-
"type": "context",
|
|
1442
|
-
"size": 128000
|
|
1443
|
-
}
|
|
1444
|
-
}
|
|
1445
|
-
]
|
|
1446
|
-
}
|
|
1447
|
-
},
|
|
1448
|
-
"Qwen/Qwen3.5-27B": {
|
|
1449
|
-
"id": "Qwen/Qwen3.5-27B",
|
|
1450
|
-
"name": "Qwen3.5 27B",
|
|
1761
|
+
"Qwen/Qwen3.6-27B": {
|
|
1762
|
+
"id": "Qwen/Qwen3.6-27B",
|
|
1763
|
+
"name": "Qwen3.6 27B",
|
|
1451
1764
|
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1452
1765
|
"family": "qwen",
|
|
1453
1766
|
"attachment": true,
|
|
@@ -1456,8 +1769,8 @@
|
|
|
1456
1769
|
"tool_call": true,
|
|
1457
1770
|
"structured_output": true,
|
|
1458
1771
|
"temperature": true,
|
|
1459
|
-
"release_date": "2026-
|
|
1460
|
-
"last_updated": "2026-
|
|
1772
|
+
"release_date": "2026-04-22",
|
|
1773
|
+
"last_updated": "2026-04-22",
|
|
1461
1774
|
"modalities": {
|
|
1462
1775
|
"input": [
|
|
1463
1776
|
"text",
|
|
@@ -1475,13 +1788,13 @@
|
|
|
1475
1788
|
"output": 65536
|
|
1476
1789
|
},
|
|
1477
1790
|
"cost": {
|
|
1478
|
-
"input": 0.
|
|
1479
|
-
"output": 2
|
|
1791
|
+
"input": 0.32,
|
|
1792
|
+
"output": 3.2
|
|
1480
1793
|
}
|
|
1481
1794
|
},
|
|
1482
|
-
"Qwen/Qwen3.
|
|
1483
|
-
"id": "Qwen/Qwen3.
|
|
1484
|
-
"name": "
|
|
1795
|
+
"Qwen/Qwen3.6-35B-A3B": {
|
|
1796
|
+
"id": "Qwen/Qwen3.6-35B-A3B",
|
|
1797
|
+
"name": "Qwen3.6 35B A3B",
|
|
1485
1798
|
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1486
1799
|
"family": "qwen",
|
|
1487
1800
|
"attachment": true,
|
|
@@ -1490,9 +1803,8 @@
|
|
|
1490
1803
|
"tool_call": true,
|
|
1491
1804
|
"structured_output": true,
|
|
1492
1805
|
"temperature": true,
|
|
1493
|
-
"
|
|
1494
|
-
"
|
|
1495
|
-
"last_updated": "2026-04-20",
|
|
1806
|
+
"release_date": "2026-04-01",
|
|
1807
|
+
"last_updated": "2026-04-01",
|
|
1496
1808
|
"modalities": {
|
|
1497
1809
|
"input": [
|
|
1498
1810
|
"text",
|
|
@@ -1509,9 +1821,8 @@
|
|
|
1509
1821
|
"output": 81920
|
|
1510
1822
|
},
|
|
1511
1823
|
"cost": {
|
|
1512
|
-
"input": 0.
|
|
1513
|
-
"output":
|
|
1514
|
-
"cache_read": 0.05
|
|
1824
|
+
"input": 0.1,
|
|
1825
|
+
"output": 0.95
|
|
1515
1826
|
}
|
|
1516
1827
|
},
|
|
1517
1828
|
"Qwen/Qwen3-Max": {
|
|
@@ -1565,20 +1876,23 @@
|
|
|
1565
1876
|
}
|
|
1566
1877
|
]
|
|
1567
1878
|
}
|
|
1568
|
-
},
|
|
1569
|
-
"
|
|
1570
|
-
"id": "
|
|
1571
|
-
"name": "
|
|
1572
|
-
"description": "
|
|
1573
|
-
"family": "
|
|
1879
|
+
},
|
|
1880
|
+
"MiniMaxAI/MiniMax-M2.5": {
|
|
1881
|
+
"id": "MiniMaxAI/MiniMax-M2.5",
|
|
1882
|
+
"name": "MiniMax M2.5",
|
|
1883
|
+
"description": "MiniMax model for chat, coding, office work, and agentic tasks",
|
|
1884
|
+
"family": "minimax",
|
|
1574
1885
|
"attachment": false,
|
|
1575
1886
|
"reasoning": true,
|
|
1576
1887
|
"reasoning_options": [],
|
|
1577
1888
|
"tool_call": true,
|
|
1578
|
-
"
|
|
1889
|
+
"interleaved": {
|
|
1890
|
+
"field": "reasoning_content"
|
|
1891
|
+
},
|
|
1579
1892
|
"temperature": true,
|
|
1580
|
-
"
|
|
1581
|
-
"
|
|
1893
|
+
"knowledge": "2025-06",
|
|
1894
|
+
"release_date": "2026-02-12",
|
|
1895
|
+
"last_updated": "2026-02-12",
|
|
1582
1896
|
"modalities": {
|
|
1583
1897
|
"input": [
|
|
1584
1898
|
"text"
|
|
@@ -1589,28 +1903,29 @@
|
|
|
1589
1903
|
},
|
|
1590
1904
|
"open_weights": true,
|
|
1591
1905
|
"limit": {
|
|
1592
|
-
"context":
|
|
1593
|
-
"output":
|
|
1906
|
+
"context": 196608,
|
|
1907
|
+
"output": 131072
|
|
1594
1908
|
},
|
|
1909
|
+
"status": "deprecated",
|
|
1595
1910
|
"cost": {
|
|
1596
|
-
"input": 0.
|
|
1597
|
-
"output":
|
|
1911
|
+
"input": 0.15,
|
|
1912
|
+
"output": 1.15,
|
|
1913
|
+
"cache_read": 0.03
|
|
1598
1914
|
}
|
|
1599
1915
|
},
|
|
1600
|
-
"
|
|
1601
|
-
"id": "
|
|
1602
|
-
"name": "
|
|
1603
|
-
"description": "
|
|
1604
|
-
"family": "
|
|
1916
|
+
"MiniMaxAI/MiniMax-M3": {
|
|
1917
|
+
"id": "MiniMaxAI/MiniMax-M3",
|
|
1918
|
+
"name": "MiniMax-M3",
|
|
1919
|
+
"description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
|
|
1920
|
+
"family": "minimax",
|
|
1605
1921
|
"attachment": true,
|
|
1606
1922
|
"reasoning": true,
|
|
1607
1923
|
"reasoning_options": [],
|
|
1608
1924
|
"tool_call": true,
|
|
1609
1925
|
"structured_output": true,
|
|
1610
1926
|
"temperature": true,
|
|
1611
|
-
"
|
|
1612
|
-
"
|
|
1613
|
-
"last_updated": "2026-04-20",
|
|
1927
|
+
"release_date": "2026-06-01",
|
|
1928
|
+
"last_updated": "2026-06-01",
|
|
1614
1929
|
"modalities": {
|
|
1615
1930
|
"input": [
|
|
1616
1931
|
"text",
|
|
@@ -1623,33 +1938,30 @@
|
|
|
1623
1938
|
},
|
|
1624
1939
|
"open_weights": true,
|
|
1625
1940
|
"limit": {
|
|
1626
|
-
"context":
|
|
1627
|
-
"output":
|
|
1941
|
+
"context": 524288,
|
|
1942
|
+
"output": 512000
|
|
1628
1943
|
},
|
|
1629
1944
|
"cost": {
|
|
1630
|
-
"input": 0.
|
|
1631
|
-
"output":
|
|
1632
|
-
"cache_read": 0.
|
|
1945
|
+
"input": 0.28,
|
|
1946
|
+
"output": 1.1,
|
|
1947
|
+
"cache_read": 0.056
|
|
1633
1948
|
}
|
|
1634
1949
|
},
|
|
1635
|
-
"
|
|
1636
|
-
"id": "
|
|
1637
|
-
"name": "
|
|
1638
|
-
"description": "
|
|
1639
|
-
"family": "
|
|
1640
|
-
"attachment":
|
|
1950
|
+
"MiniMaxAI/MiniMax-M2.7": {
|
|
1951
|
+
"id": "MiniMaxAI/MiniMax-M2.7",
|
|
1952
|
+
"name": "MiniMax-M2.7",
|
|
1953
|
+
"description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
|
|
1954
|
+
"family": "minimax",
|
|
1955
|
+
"attachment": false,
|
|
1641
1956
|
"reasoning": true,
|
|
1642
1957
|
"reasoning_options": [],
|
|
1643
1958
|
"tool_call": true,
|
|
1644
|
-
"structured_output": true,
|
|
1645
1959
|
"temperature": true,
|
|
1646
|
-
"release_date": "2026-
|
|
1647
|
-
"last_updated": "2026-
|
|
1960
|
+
"release_date": "2026-03-18",
|
|
1961
|
+
"last_updated": "2026-03-18",
|
|
1648
1962
|
"modalities": {
|
|
1649
1963
|
"input": [
|
|
1650
|
-
"text"
|
|
1651
|
-
"image",
|
|
1652
|
-
"video"
|
|
1964
|
+
"text"
|
|
1653
1965
|
],
|
|
1654
1966
|
"output": [
|
|
1655
1967
|
"text"
|
|
@@ -1657,30 +1969,30 @@
|
|
|
1657
1969
|
},
|
|
1658
1970
|
"open_weights": true,
|
|
1659
1971
|
"limit": {
|
|
1660
|
-
"context":
|
|
1661
|
-
"output":
|
|
1972
|
+
"context": 196608,
|
|
1973
|
+
"output": 131072
|
|
1662
1974
|
},
|
|
1663
1975
|
"cost": {
|
|
1664
|
-
"input": 0.
|
|
1665
|
-
"output":
|
|
1976
|
+
"input": 0.25,
|
|
1977
|
+
"output": 1,
|
|
1978
|
+
"cache_read": 0.05
|
|
1666
1979
|
}
|
|
1667
1980
|
},
|
|
1668
|
-
"
|
|
1669
|
-
"id": "
|
|
1670
|
-
"name": "
|
|
1671
|
-
"description": "
|
|
1672
|
-
"family": "
|
|
1673
|
-
"attachment":
|
|
1981
|
+
"meta-llama/Llama-4-Scout-17B-16E-Instruct": {
|
|
1982
|
+
"id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
1983
|
+
"name": "Llama 4 Scout 17B",
|
|
1984
|
+
"description": "Open multimodal Llama model for long-context analysis and efficient agents",
|
|
1985
|
+
"family": "llama",
|
|
1986
|
+
"attachment": true,
|
|
1674
1987
|
"reasoning": false,
|
|
1675
1988
|
"tool_call": true,
|
|
1676
1989
|
"structured_output": true,
|
|
1677
|
-
"
|
|
1678
|
-
"
|
|
1679
|
-
"release_date": "2025-09",
|
|
1680
|
-
"last_updated": "2025-09",
|
|
1990
|
+
"release_date": "2025-04-05",
|
|
1991
|
+
"last_updated": "2025-04-05",
|
|
1681
1992
|
"modalities": {
|
|
1682
1993
|
"input": [
|
|
1683
|
-
"text"
|
|
1994
|
+
"text",
|
|
1995
|
+
"image"
|
|
1684
1996
|
],
|
|
1685
1997
|
"output": [
|
|
1686
1998
|
"text"
|
|
@@ -1688,35 +2000,25 @@
|
|
|
1688
2000
|
},
|
|
1689
2001
|
"open_weights": true,
|
|
1690
2002
|
"limit": {
|
|
1691
|
-
"context":
|
|
1692
|
-
"output":
|
|
2003
|
+
"context": 327680,
|
|
2004
|
+
"output": 16384
|
|
1693
2005
|
},
|
|
1694
2006
|
"cost": {
|
|
1695
|
-
"input": 0.
|
|
1696
|
-
"output":
|
|
2007
|
+
"input": 0.1,
|
|
2008
|
+
"output": 0.3
|
|
1697
2009
|
}
|
|
1698
2010
|
},
|
|
1699
|
-
"
|
|
1700
|
-
"id": "
|
|
1701
|
-
"name": "
|
|
1702
|
-
"description": "
|
|
1703
|
-
"family": "
|
|
2011
|
+
"meta-llama/Llama-3.3-70B-Instruct-Turbo": {
|
|
2012
|
+
"id": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
2013
|
+
"name": "Llama 3.3 70B Turbo",
|
|
2014
|
+
"description": "Compact Llama instruction model for fast chat and local deployment",
|
|
2015
|
+
"family": "llama",
|
|
1704
2016
|
"attachment": false,
|
|
1705
|
-
"reasoning":
|
|
1706
|
-
"reasoning_options": [
|
|
1707
|
-
{
|
|
1708
|
-
"type": "toggle"
|
|
1709
|
-
}
|
|
1710
|
-
],
|
|
2017
|
+
"reasoning": false,
|
|
1711
2018
|
"tool_call": true,
|
|
1712
|
-
"interleaved": {
|
|
1713
|
-
"field": "reasoning_content"
|
|
1714
|
-
},
|
|
1715
2019
|
"structured_output": true,
|
|
1716
|
-
"
|
|
1717
|
-
"
|
|
1718
|
-
"release_date": "2026-04-07",
|
|
1719
|
-
"last_updated": "2026-04-07",
|
|
2020
|
+
"release_date": "2024-12-06",
|
|
2021
|
+
"last_updated": "2024-12-06",
|
|
1720
2022
|
"modalities": {
|
|
1721
2023
|
"input": [
|
|
1722
2024
|
"text"
|
|
@@ -1727,39 +2029,29 @@
|
|
|
1727
2029
|
},
|
|
1728
2030
|
"open_weights": true,
|
|
1729
2031
|
"limit": {
|
|
1730
|
-
"context":
|
|
2032
|
+
"context": 131072,
|
|
1731
2033
|
"output": 16384
|
|
1732
2034
|
},
|
|
1733
2035
|
"cost": {
|
|
1734
|
-
"input": 1
|
|
1735
|
-
"output":
|
|
1736
|
-
"cache_read": 0.205
|
|
2036
|
+
"input": 0.1,
|
|
2037
|
+
"output": 0.32
|
|
1737
2038
|
}
|
|
1738
2039
|
},
|
|
1739
|
-
"
|
|
1740
|
-
"id": "
|
|
1741
|
-
"name": "
|
|
1742
|
-
"description": "
|
|
1743
|
-
"family": "
|
|
1744
|
-
"attachment":
|
|
1745
|
-
"reasoning":
|
|
1746
|
-
"
|
|
1747
|
-
{
|
|
1748
|
-
"type": "toggle"
|
|
1749
|
-
}
|
|
1750
|
-
],
|
|
1751
|
-
"tool_call": true,
|
|
1752
|
-
"interleaved": {
|
|
1753
|
-
"field": "reasoning_content"
|
|
1754
|
-
},
|
|
2040
|
+
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": {
|
|
2041
|
+
"id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
2042
|
+
"name": "Llama 4 Maverick 17B FP8",
|
|
2043
|
+
"description": "Open multimodal Llama model for strong reasoning and fast responses",
|
|
2044
|
+
"family": "llama",
|
|
2045
|
+
"attachment": true,
|
|
2046
|
+
"reasoning": false,
|
|
2047
|
+
"tool_call": false,
|
|
1755
2048
|
"structured_output": true,
|
|
1756
|
-
"
|
|
1757
|
-
"
|
|
1758
|
-
"release_date": "2025-09-30",
|
|
1759
|
-
"last_updated": "2025-09-30",
|
|
2049
|
+
"release_date": "2025-04-05",
|
|
2050
|
+
"last_updated": "2025-04-05",
|
|
1760
2051
|
"modalities": {
|
|
1761
2052
|
"input": [
|
|
1762
|
-
"text"
|
|
2053
|
+
"text",
|
|
2054
|
+
"image"
|
|
1763
2055
|
],
|
|
1764
2056
|
"output": [
|
|
1765
2057
|
"text"
|
|
@@ -1767,36 +2059,36 @@
|
|
|
1767
2059
|
},
|
|
1768
2060
|
"open_weights": true,
|
|
1769
2061
|
"limit": {
|
|
1770
|
-
"context":
|
|
1771
|
-
"output":
|
|
2062
|
+
"context": 1048576,
|
|
2063
|
+
"output": 16384
|
|
1772
2064
|
},
|
|
1773
2065
|
"cost": {
|
|
1774
|
-
"input": 0.
|
|
1775
|
-
"output":
|
|
1776
|
-
"cache_read": 0.1
|
|
2066
|
+
"input": 0.2,
|
|
2067
|
+
"output": 0.8
|
|
1777
2068
|
}
|
|
1778
2069
|
},
|
|
1779
|
-
"
|
|
1780
|
-
"id": "
|
|
1781
|
-
"name": "
|
|
1782
|
-
"description": "
|
|
1783
|
-
"family": "
|
|
2070
|
+
"openai/gpt-oss-20b": {
|
|
2071
|
+
"id": "openai/gpt-oss-20b",
|
|
2072
|
+
"name": "GPT OSS 20B",
|
|
2073
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
2074
|
+
"family": "gpt-oss",
|
|
1784
2075
|
"attachment": false,
|
|
1785
2076
|
"reasoning": true,
|
|
1786
2077
|
"reasoning_options": [
|
|
1787
2078
|
{
|
|
1788
|
-
"type": "
|
|
2079
|
+
"type": "effort",
|
|
2080
|
+
"values": [
|
|
2081
|
+
"low",
|
|
2082
|
+
"medium",
|
|
2083
|
+
"high"
|
|
2084
|
+
]
|
|
1789
2085
|
}
|
|
1790
2086
|
],
|
|
1791
2087
|
"tool_call": true,
|
|
1792
|
-
"interleaved": {
|
|
1793
|
-
"field": "reasoning_content"
|
|
1794
|
-
},
|
|
1795
2088
|
"structured_output": true,
|
|
1796
2089
|
"temperature": true,
|
|
1797
|
-
"
|
|
1798
|
-
"
|
|
1799
|
-
"last_updated": "2026-02-12",
|
|
2090
|
+
"release_date": "2025-08-05",
|
|
2091
|
+
"last_updated": "2025-08-05",
|
|
1800
2092
|
"modalities": {
|
|
1801
2093
|
"input": [
|
|
1802
2094
|
"text"
|
|
@@ -1807,36 +2099,36 @@
|
|
|
1807
2099
|
},
|
|
1808
2100
|
"open_weights": true,
|
|
1809
2101
|
"limit": {
|
|
1810
|
-
"context":
|
|
2102
|
+
"context": 131072,
|
|
1811
2103
|
"output": 16384
|
|
1812
2104
|
},
|
|
1813
2105
|
"cost": {
|
|
1814
|
-
"input": 0.
|
|
1815
|
-
"output":
|
|
1816
|
-
"cache_read": 0.12
|
|
2106
|
+
"input": 0.03,
|
|
2107
|
+
"output": 0.14
|
|
1817
2108
|
}
|
|
1818
2109
|
},
|
|
1819
|
-
"
|
|
1820
|
-
"id": "
|
|
1821
|
-
"name": "
|
|
1822
|
-
"description": "
|
|
1823
|
-
"family": "
|
|
2110
|
+
"openai/gpt-oss-120b": {
|
|
2111
|
+
"id": "openai/gpt-oss-120b",
|
|
2112
|
+
"name": "GPT OSS 120B",
|
|
2113
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
2114
|
+
"family": "gpt-oss",
|
|
1824
2115
|
"attachment": false,
|
|
1825
2116
|
"reasoning": true,
|
|
1826
2117
|
"reasoning_options": [
|
|
1827
2118
|
{
|
|
1828
|
-
"type": "
|
|
2119
|
+
"type": "effort",
|
|
2120
|
+
"values": [
|
|
2121
|
+
"low",
|
|
2122
|
+
"medium",
|
|
2123
|
+
"high"
|
|
2124
|
+
]
|
|
1829
2125
|
}
|
|
1830
2126
|
],
|
|
1831
2127
|
"tool_call": true,
|
|
1832
|
-
"interleaved": {
|
|
1833
|
-
"field": "reasoning_content"
|
|
1834
|
-
},
|
|
1835
2128
|
"structured_output": true,
|
|
1836
2129
|
"temperature": true,
|
|
1837
|
-
"
|
|
1838
|
-
"
|
|
1839
|
-
"last_updated": "2025-12-22",
|
|
2130
|
+
"release_date": "2025-08-05",
|
|
2131
|
+
"last_updated": "2025-08-05",
|
|
1840
2132
|
"modalities": {
|
|
1841
2133
|
"input": [
|
|
1842
2134
|
"text"
|
|
@@ -1847,34 +2139,24 @@
|
|
|
1847
2139
|
},
|
|
1848
2140
|
"open_weights": true,
|
|
1849
2141
|
"limit": {
|
|
1850
|
-
"context":
|
|
2142
|
+
"context": 131072,
|
|
1851
2143
|
"output": 16384
|
|
1852
2144
|
},
|
|
1853
2145
|
"cost": {
|
|
1854
|
-
"input": 0.
|
|
1855
|
-
"output":
|
|
1856
|
-
"cache_read": 0.08
|
|
2146
|
+
"input": 0.037,
|
|
2147
|
+
"output": 0.17
|
|
1857
2148
|
}
|
|
1858
2149
|
},
|
|
1859
|
-
"
|
|
1860
|
-
"id": "
|
|
1861
|
-
"name": "
|
|
1862
|
-
"description": "
|
|
1863
|
-
"family": "
|
|
1864
|
-
"attachment":
|
|
2150
|
+
"moonshotai/Kimi-K2.5": {
|
|
2151
|
+
"id": "moonshotai/Kimi-K2.5",
|
|
2152
|
+
"name": "Kimi K2.5",
|
|
2153
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
2154
|
+
"family": "kimi-k2",
|
|
2155
|
+
"attachment": true,
|
|
1865
2156
|
"reasoning": true,
|
|
1866
2157
|
"reasoning_options": [
|
|
1867
2158
|
{
|
|
1868
2159
|
"type": "toggle"
|
|
1869
|
-
},
|
|
1870
|
-
{
|
|
1871
|
-
"type": "effort",
|
|
1872
|
-
"values": [
|
|
1873
|
-
"low",
|
|
1874
|
-
"medium",
|
|
1875
|
-
"high",
|
|
1876
|
-
"xhigh"
|
|
1877
|
-
]
|
|
1878
2160
|
}
|
|
1879
2161
|
],
|
|
1880
2162
|
"tool_call": true,
|
|
@@ -1883,11 +2165,14 @@
|
|
|
1883
2165
|
},
|
|
1884
2166
|
"structured_output": true,
|
|
1885
2167
|
"temperature": true,
|
|
1886
|
-
"
|
|
1887
|
-
"
|
|
2168
|
+
"knowledge": "2025-01",
|
|
2169
|
+
"release_date": "2026-01-27",
|
|
2170
|
+
"last_updated": "2026-01-27",
|
|
1888
2171
|
"modalities": {
|
|
1889
2172
|
"input": [
|
|
1890
|
-
"text"
|
|
2173
|
+
"text",
|
|
2174
|
+
"image",
|
|
2175
|
+
"video"
|
|
1891
2176
|
],
|
|
1892
2177
|
"output": [
|
|
1893
2178
|
"text"
|
|
@@ -1895,35 +2180,42 @@
|
|
|
1895
2180
|
},
|
|
1896
2181
|
"open_weights": true,
|
|
1897
2182
|
"limit": {
|
|
1898
|
-
"context":
|
|
2183
|
+
"context": 262144,
|
|
1899
2184
|
"output": 32768
|
|
1900
2185
|
},
|
|
2186
|
+
"status": "deprecated",
|
|
1901
2187
|
"cost": {
|
|
1902
|
-
"input": 0.
|
|
1903
|
-
"output": 2.
|
|
1904
|
-
"cache_read": 0.
|
|
2188
|
+
"input": 0.45,
|
|
2189
|
+
"output": 2.25,
|
|
2190
|
+
"cache_read": 0.07
|
|
1905
2191
|
}
|
|
1906
2192
|
},
|
|
1907
|
-
"
|
|
1908
|
-
"id": "
|
|
1909
|
-
"name": "
|
|
1910
|
-
"description": "
|
|
1911
|
-
"family": "
|
|
1912
|
-
"attachment":
|
|
2193
|
+
"moonshotai/Kimi-K2.7-Code": {
|
|
2194
|
+
"id": "moonshotai/Kimi-K2.7-Code",
|
|
2195
|
+
"name": "Kimi K2.7 Code",
|
|
2196
|
+
"description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
|
|
2197
|
+
"family": "kimi-k2",
|
|
2198
|
+
"attachment": true,
|
|
1913
2199
|
"reasoning": true,
|
|
1914
|
-
"reasoning_options": [
|
|
2200
|
+
"reasoning_options": [
|
|
2201
|
+
{
|
|
2202
|
+
"type": "toggle"
|
|
2203
|
+
}
|
|
2204
|
+
],
|
|
1915
2205
|
"tool_call": true,
|
|
1916
2206
|
"interleaved": {
|
|
1917
2207
|
"field": "reasoning_content"
|
|
1918
2208
|
},
|
|
1919
2209
|
"structured_output": true,
|
|
1920
|
-
"temperature":
|
|
1921
|
-
"knowledge": "2025-
|
|
1922
|
-
"release_date": "2026-
|
|
1923
|
-
"last_updated": "2026-
|
|
2210
|
+
"temperature": false,
|
|
2211
|
+
"knowledge": "2025-01",
|
|
2212
|
+
"release_date": "2026-06-12",
|
|
2213
|
+
"last_updated": "2026-06-12",
|
|
1924
2214
|
"modalities": {
|
|
1925
2215
|
"input": [
|
|
1926
|
-
"text"
|
|
2216
|
+
"text",
|
|
2217
|
+
"image",
|
|
2218
|
+
"video"
|
|
1927
2219
|
],
|
|
1928
2220
|
"output": [
|
|
1929
2221
|
"text"
|
|
@@ -1931,32 +2223,41 @@
|
|
|
1931
2223
|
},
|
|
1932
2224
|
"open_weights": true,
|
|
1933
2225
|
"limit": {
|
|
1934
|
-
"context":
|
|
1935
|
-
"output":
|
|
2226
|
+
"context": 262144,
|
|
2227
|
+
"output": 262144
|
|
1936
2228
|
},
|
|
1937
2229
|
"cost": {
|
|
1938
|
-
"input": 0.
|
|
1939
|
-
"output":
|
|
1940
|
-
"cache_read": 0.
|
|
2230
|
+
"input": 0.68,
|
|
2231
|
+
"output": 3.4,
|
|
2232
|
+
"cache_read": 0.136
|
|
1941
2233
|
}
|
|
1942
2234
|
},
|
|
1943
|
-
"
|
|
1944
|
-
"id": "
|
|
1945
|
-
"name": "
|
|
1946
|
-
"description": "
|
|
1947
|
-
"family": "
|
|
2235
|
+
"moonshotai/Kimi-K2.6": {
|
|
2236
|
+
"id": "moonshotai/Kimi-K2.6",
|
|
2237
|
+
"name": "Kimi K2.6",
|
|
2238
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
2239
|
+
"family": "kimi-k2",
|
|
1948
2240
|
"attachment": true,
|
|
1949
2241
|
"reasoning": true,
|
|
1950
|
-
"reasoning_options": [
|
|
2242
|
+
"reasoning_options": [
|
|
2243
|
+
{
|
|
2244
|
+
"type": "toggle"
|
|
2245
|
+
}
|
|
2246
|
+
],
|
|
1951
2247
|
"tool_call": true,
|
|
2248
|
+
"interleaved": {
|
|
2249
|
+
"field": "reasoning_content"
|
|
2250
|
+
},
|
|
2251
|
+
"structured_output": true,
|
|
1952
2252
|
"temperature": true,
|
|
1953
|
-
"
|
|
1954
|
-
"
|
|
2253
|
+
"knowledge": "2024-04",
|
|
2254
|
+
"release_date": "2026-04-21",
|
|
2255
|
+
"last_updated": "2026-04-21",
|
|
1955
2256
|
"modalities": {
|
|
1956
2257
|
"input": [
|
|
1957
2258
|
"text",
|
|
1958
2259
|
"image",
|
|
1959
|
-
"
|
|
2260
|
+
"video"
|
|
1960
2261
|
],
|
|
1961
2262
|
"output": [
|
|
1962
2263
|
"text"
|
|
@@ -1964,33 +2265,41 @@
|
|
|
1964
2265
|
},
|
|
1965
2266
|
"open_weights": true,
|
|
1966
2267
|
"limit": {
|
|
1967
|
-
"context":
|
|
1968
|
-
"output":
|
|
2268
|
+
"context": 262144,
|
|
2269
|
+
"output": 16384
|
|
1969
2270
|
},
|
|
1970
2271
|
"cost": {
|
|
1971
|
-
"input": 0.
|
|
1972
|
-
"output":
|
|
1973
|
-
"cache_read": 0.
|
|
2272
|
+
"input": 0.75,
|
|
2273
|
+
"output": 3.5,
|
|
2274
|
+
"cache_read": 0.15
|
|
1974
2275
|
}
|
|
1975
2276
|
},
|
|
1976
|
-
"
|
|
1977
|
-
"id": "
|
|
1978
|
-
"name": "
|
|
1979
|
-
"description": "Multimodal
|
|
1980
|
-
"family": "
|
|
2277
|
+
"moonshotai/Kimi-K3": {
|
|
2278
|
+
"id": "moonshotai/Kimi-K3",
|
|
2279
|
+
"name": "Kimi K3",
|
|
2280
|
+
"description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
|
|
2281
|
+
"family": "kimi-k3",
|
|
1981
2282
|
"attachment": true,
|
|
1982
2283
|
"reasoning": true,
|
|
1983
|
-
"reasoning_options": [
|
|
2284
|
+
"reasoning_options": [
|
|
2285
|
+
{
|
|
2286
|
+
"type": "effort",
|
|
2287
|
+
"values": [
|
|
2288
|
+
"low",
|
|
2289
|
+
"high",
|
|
2290
|
+
"max"
|
|
2291
|
+
]
|
|
2292
|
+
}
|
|
2293
|
+
],
|
|
1984
2294
|
"tool_call": true,
|
|
1985
2295
|
"structured_output": true,
|
|
1986
|
-
"temperature":
|
|
1987
|
-
"release_date": "2026-07-
|
|
1988
|
-
"last_updated": "2026-07-
|
|
2296
|
+
"temperature": false,
|
|
2297
|
+
"release_date": "2026-07-16",
|
|
2298
|
+
"last_updated": "2026-07-16",
|
|
1989
2299
|
"modalities": {
|
|
1990
2300
|
"input": [
|
|
1991
2301
|
"text",
|
|
1992
|
-
"image"
|
|
1993
|
-
"audio"
|
|
2302
|
+
"image"
|
|
1994
2303
|
],
|
|
1995
2304
|
"output": [
|
|
1996
2305
|
"text"
|
|
@@ -1998,30 +2307,31 @@
|
|
|
1998
2307
|
},
|
|
1999
2308
|
"open_weights": true,
|
|
2000
2309
|
"limit": {
|
|
2001
|
-
"context":
|
|
2002
|
-
"output":
|
|
2310
|
+
"context": 1048576,
|
|
2311
|
+
"output": 131072
|
|
2003
2312
|
},
|
|
2004
2313
|
"cost": {
|
|
2005
|
-
"input":
|
|
2006
|
-
"output":
|
|
2007
|
-
"cache_read": 0.
|
|
2314
|
+
"input": 2.85,
|
|
2315
|
+
"output": 14.25,
|
|
2316
|
+
"cache_read": 0.285
|
|
2008
2317
|
}
|
|
2009
2318
|
},
|
|
2010
|
-
"
|
|
2011
|
-
"id": "
|
|
2012
|
-
"name": "
|
|
2013
|
-
"description": "
|
|
2014
|
-
"family": "
|
|
2015
|
-
"attachment":
|
|
2016
|
-
"reasoning":
|
|
2017
|
-
"
|
|
2319
|
+
"tencent/Hy3": {
|
|
2320
|
+
"id": "tencent/Hy3",
|
|
2321
|
+
"name": "Hy3",
|
|
2322
|
+
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
|
|
2323
|
+
"family": "Hy",
|
|
2324
|
+
"attachment": false,
|
|
2325
|
+
"reasoning": true,
|
|
2326
|
+
"reasoning_options": [],
|
|
2327
|
+
"tool_call": true,
|
|
2018
2328
|
"structured_output": true,
|
|
2019
|
-
"
|
|
2020
|
-
"
|
|
2329
|
+
"temperature": true,
|
|
2330
|
+
"release_date": "2026-07-06",
|
|
2331
|
+
"last_updated": "2026-07-06",
|
|
2021
2332
|
"modalities": {
|
|
2022
2333
|
"input": [
|
|
2023
|
-
"text"
|
|
2024
|
-
"image"
|
|
2334
|
+
"text"
|
|
2025
2335
|
],
|
|
2026
2336
|
"output": [
|
|
2027
2337
|
"text"
|
|
@@ -2029,28 +2339,41 @@
|
|
|
2029
2339
|
},
|
|
2030
2340
|
"open_weights": true,
|
|
2031
2341
|
"limit": {
|
|
2032
|
-
"context":
|
|
2033
|
-
"
|
|
2342
|
+
"context": 262144,
|
|
2343
|
+
"input": 192000,
|
|
2344
|
+
"output": 128000
|
|
2034
2345
|
},
|
|
2035
2346
|
"cost": {
|
|
2036
|
-
"input": 0.
|
|
2037
|
-
"output": 0.
|
|
2347
|
+
"input": 0.14,
|
|
2348
|
+
"output": 0.58,
|
|
2349
|
+
"cache_read": 0.035
|
|
2038
2350
|
}
|
|
2039
2351
|
},
|
|
2040
|
-
"
|
|
2041
|
-
"id": "
|
|
2042
|
-
"name": "
|
|
2043
|
-
"description": "
|
|
2044
|
-
"family": "
|
|
2045
|
-
"attachment":
|
|
2046
|
-
"reasoning":
|
|
2352
|
+
"XiaomiMiMo/MiMo-V2.5-Pro": {
|
|
2353
|
+
"id": "XiaomiMiMo/MiMo-V2.5-Pro",
|
|
2354
|
+
"name": "MiMo-V2.5-Pro",
|
|
2355
|
+
"description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution",
|
|
2356
|
+
"family": "mimo",
|
|
2357
|
+
"attachment": true,
|
|
2358
|
+
"reasoning": true,
|
|
2359
|
+
"reasoning_options": [
|
|
2360
|
+
{
|
|
2361
|
+
"type": "toggle"
|
|
2362
|
+
}
|
|
2363
|
+
],
|
|
2047
2364
|
"tool_call": true,
|
|
2365
|
+
"interleaved": {
|
|
2366
|
+
"field": "reasoning_content"
|
|
2367
|
+
},
|
|
2048
2368
|
"structured_output": true,
|
|
2049
|
-
"
|
|
2050
|
-
"
|
|
2369
|
+
"temperature": true,
|
|
2370
|
+
"knowledge": "2024-12",
|
|
2371
|
+
"release_date": "2026-04-22",
|
|
2372
|
+
"last_updated": "2026-04-22",
|
|
2051
2373
|
"modalities": {
|
|
2052
2374
|
"input": [
|
|
2053
|
-
"text"
|
|
2375
|
+
"text",
|
|
2376
|
+
"audio"
|
|
2054
2377
|
],
|
|
2055
2378
|
"output": [
|
|
2056
2379
|
"text"
|
|
@@ -2058,29 +2381,42 @@
|
|
|
2058
2381
|
},
|
|
2059
2382
|
"open_weights": true,
|
|
2060
2383
|
"limit": {
|
|
2061
|
-
"context":
|
|
2384
|
+
"context": 1048576,
|
|
2062
2385
|
"output": 16384
|
|
2063
2386
|
},
|
|
2064
2387
|
"cost": {
|
|
2065
|
-
"input":
|
|
2066
|
-
"output":
|
|
2388
|
+
"input": 1,
|
|
2389
|
+
"output": 3,
|
|
2390
|
+
"cache_read": 0.2
|
|
2067
2391
|
}
|
|
2068
2392
|
},
|
|
2069
|
-
"
|
|
2070
|
-
"id": "
|
|
2071
|
-
"name": "
|
|
2072
|
-
"description": "Open
|
|
2073
|
-
"family": "
|
|
2393
|
+
"XiaomiMiMo/MiMo-V2.5": {
|
|
2394
|
+
"id": "XiaomiMiMo/MiMo-V2.5",
|
|
2395
|
+
"name": "MiMo-V2.5",
|
|
2396
|
+
"description": "Open MiMo model for multimodal coding agents and long-context automation",
|
|
2397
|
+
"family": "mimo",
|
|
2074
2398
|
"attachment": true,
|
|
2075
|
-
"reasoning":
|
|
2399
|
+
"reasoning": true,
|
|
2400
|
+
"reasoning_options": [
|
|
2401
|
+
{
|
|
2402
|
+
"type": "toggle"
|
|
2403
|
+
}
|
|
2404
|
+
],
|
|
2076
2405
|
"tool_call": true,
|
|
2406
|
+
"interleaved": {
|
|
2407
|
+
"field": "reasoning_content"
|
|
2408
|
+
},
|
|
2077
2409
|
"structured_output": true,
|
|
2078
|
-
"
|
|
2079
|
-
"
|
|
2410
|
+
"temperature": true,
|
|
2411
|
+
"knowledge": "2024-12",
|
|
2412
|
+
"release_date": "2026-04-22",
|
|
2413
|
+
"last_updated": "2026-04-22",
|
|
2080
2414
|
"modalities": {
|
|
2081
2415
|
"input": [
|
|
2082
2416
|
"text",
|
|
2083
|
-
"image"
|
|
2417
|
+
"image",
|
|
2418
|
+
"audio",
|
|
2419
|
+
"video"
|
|
2084
2420
|
],
|
|
2085
2421
|
"output": [
|
|
2086
2422
|
"text"
|
|
@@ -2088,12 +2424,13 @@
|
|
|
2088
2424
|
},
|
|
2089
2425
|
"open_weights": true,
|
|
2090
2426
|
"limit": {
|
|
2091
|
-
"context":
|
|
2427
|
+
"context": 262144,
|
|
2092
2428
|
"output": 16384
|
|
2093
2429
|
},
|
|
2094
2430
|
"cost": {
|
|
2095
|
-
"input": 0.
|
|
2096
|
-
"output":
|
|
2431
|
+
"input": 0.4,
|
|
2432
|
+
"output": 2,
|
|
2433
|
+
"cache_read": 0.08
|
|
2097
2434
|
}
|
|
2098
2435
|
}
|
|
2099
2436
|
}
|