llm.rb 15.5.0 → 16.0.0.pre.2
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 +88 -1
- data/README.md +21 -30
- data/data/alibaba.json +1123 -949
- data/data/anthropic.json +296 -242
- data/data/bedrock.json +4178 -3807
- data/data/deepinfra.json +1529 -1358
- data/data/deepseek.json +15 -11
- data/data/google.json +787 -763
- data/data/mistral.json +476 -496
- data/data/moonshot.json +53 -50
- data/data/openai.json +1294 -1165
- data/data/openrouter.json +8722 -8546
- data/data/xai.json +207 -202
- data/data/zai.json +271 -264
- data/docs/deepdive/features/builtin_tools.md +3 -1
- data/docs/deepdive/features/console.md +7 -0
- data/docs/deepdive/fundamentals/schema.md +46 -0
- data/docs/deepdive/fundamentals/tools.md +68 -0
- data/docs/deepdive/protocols/a2a.md +91 -0
- data/docs/deepdive/protocols/mcp.md +50 -0
- data/docs/deepdive/reference/model_registry.md +1 -1
- data/lib/llm/console/input.rb +15 -5
- data/lib/llm/function/async/reactor.rb +10 -0
- data/lib/llm/function/async/task.rb +63 -11
- data/lib/llm/function/fork/job.rb +31 -8
- data/lib/llm/function/ractor/job.rb +59 -20
- data/lib/llm/function/ractor/mailbox.rb +39 -3
- data/lib/llm/function/ractor/task.rb +26 -9
- data/lib/llm/function/window.rb +77 -0
- data/lib/llm/function.rb +2 -1
- data/lib/llm/providers/deepseek/request_adapter/completion.rb +43 -2
- data/lib/llm/tool/param.rb +8 -8
- data/lib/llm/tools/exec.rb +2 -0
- data/lib/llm/version.rb +1 -1
- data/llm.gemspec +4 -4
- metadata +7 -6
data/data/deepinfra.json
CHANGED
|
@@ -7,64 +7,52 @@
|
|
|
7
7
|
"name": "Deep Infra",
|
|
8
8
|
"doc": "https://deepinfra.com/models",
|
|
9
9
|
"models": {
|
|
10
|
-
"
|
|
11
|
-
"id": "
|
|
12
|
-
"name": "
|
|
13
|
-
"description": "
|
|
14
|
-
"family": "
|
|
15
|
-
"attachment":
|
|
10
|
+
"tencent/Hy3": {
|
|
11
|
+
"id": "tencent/Hy3",
|
|
12
|
+
"name": "Hy3",
|
|
13
|
+
"description": "Tencent Hy reasoning model for coding, instruction following, and agent tasks",
|
|
14
|
+
"family": "Hy",
|
|
15
|
+
"attachment": false,
|
|
16
16
|
"reasoning": true,
|
|
17
17
|
"reasoning_options": [],
|
|
18
18
|
"tool_call": true,
|
|
19
19
|
"structured_output": true,
|
|
20
20
|
"temperature": true,
|
|
21
|
-
"release_date": "2026-
|
|
22
|
-
"last_updated": "2026-
|
|
21
|
+
"release_date": "2026-07-06",
|
|
22
|
+
"last_updated": "2026-07-06",
|
|
23
23
|
"modalities": {
|
|
24
24
|
"input": [
|
|
25
|
-
"text"
|
|
26
|
-
"image"
|
|
25
|
+
"text"
|
|
27
26
|
],
|
|
28
27
|
"output": [
|
|
29
28
|
"text"
|
|
30
29
|
]
|
|
31
30
|
},
|
|
32
|
-
"open_weights":
|
|
31
|
+
"open_weights": true,
|
|
33
32
|
"limit": {
|
|
34
|
-
"context":
|
|
35
|
-
"
|
|
33
|
+
"context": 262144,
|
|
34
|
+
"input": 192000,
|
|
35
|
+
"output": 128000
|
|
36
36
|
},
|
|
37
37
|
"cost": {
|
|
38
|
-
"input": 0.
|
|
39
|
-
"output": 0.
|
|
40
|
-
"cache_read": 0.
|
|
41
|
-
|
|
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
|
-
}
|
|
38
|
+
"input": 0.13,
|
|
39
|
+
"output": 0.53,
|
|
40
|
+
"cache_read": 0.033
|
|
41
|
+
},
|
|
42
|
+
"canonical_model_id": "tencent/hy3"
|
|
53
43
|
},
|
|
54
|
-
"
|
|
55
|
-
"id": "
|
|
56
|
-
"name": "
|
|
57
|
-
"description": "
|
|
58
|
-
"family": "
|
|
59
|
-
"attachment":
|
|
44
|
+
"tencent/Hy4-preview": {
|
|
45
|
+
"id": "tencent/Hy4-preview",
|
|
46
|
+
"name": "Hy4 preview",
|
|
47
|
+
"description": "A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.",
|
|
48
|
+
"family": "Hy",
|
|
49
|
+
"attachment": false,
|
|
60
50
|
"reasoning": true,
|
|
61
51
|
"reasoning_options": [
|
|
62
52
|
{
|
|
63
53
|
"type": "effort",
|
|
64
54
|
"values": [
|
|
65
55
|
"none",
|
|
66
|
-
"low",
|
|
67
|
-
"medium",
|
|
68
56
|
"high"
|
|
69
57
|
]
|
|
70
58
|
}
|
|
@@ -72,100 +60,72 @@
|
|
|
72
60
|
"tool_call": true,
|
|
73
61
|
"structured_output": true,
|
|
74
62
|
"temperature": true,
|
|
75
|
-
"release_date": "2026-
|
|
76
|
-
"last_updated": "2026-
|
|
63
|
+
"release_date": "2026-08-28",
|
|
64
|
+
"last_updated": "2026-08-28",
|
|
77
65
|
"modalities": {
|
|
78
66
|
"input": [
|
|
79
|
-
"text"
|
|
80
|
-
"image"
|
|
67
|
+
"text"
|
|
81
68
|
],
|
|
82
69
|
"output": [
|
|
83
70
|
"text"
|
|
84
71
|
]
|
|
85
72
|
},
|
|
86
|
-
"open_weights":
|
|
73
|
+
"open_weights": true,
|
|
87
74
|
"limit": {
|
|
88
|
-
"context":
|
|
89
|
-
"output":
|
|
75
|
+
"context": 1048576,
|
|
76
|
+
"output": 64000
|
|
90
77
|
},
|
|
91
78
|
"cost": {
|
|
92
|
-
"input": 0.
|
|
93
|
-
"output":
|
|
94
|
-
"cache_read": 0.
|
|
95
|
-
|
|
96
|
-
|
|
97
|
-
"input": 1,
|
|
98
|
-
"output": 6,
|
|
99
|
-
"cache_read": 0.2,
|
|
100
|
-
"tier": {
|
|
101
|
-
"type": "context",
|
|
102
|
-
"size": 128000
|
|
103
|
-
}
|
|
104
|
-
}
|
|
105
|
-
]
|
|
106
|
-
}
|
|
79
|
+
"input": 0.834,
|
|
80
|
+
"output": 2.501,
|
|
81
|
+
"cache_read": 0.042
|
|
82
|
+
},
|
|
83
|
+
"canonical_model_id": "tencent/hy4-preview"
|
|
107
84
|
},
|
|
108
|
-
"
|
|
109
|
-
"id": "
|
|
110
|
-
"name": "
|
|
111
|
-
"description": "
|
|
112
|
-
"family": "
|
|
113
|
-
"attachment":
|
|
114
|
-
"reasoning":
|
|
115
|
-
"reasoning_options": [],
|
|
85
|
+
"meta-llama/Llama-3.3-70B-Instruct-Turbo": {
|
|
86
|
+
"id": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
87
|
+
"name": "Llama 3.3 70B Turbo",
|
|
88
|
+
"description": "Compact Llama instruction model for fast chat and local deployment",
|
|
89
|
+
"family": "llama",
|
|
90
|
+
"attachment": false,
|
|
91
|
+
"reasoning": false,
|
|
116
92
|
"tool_call": true,
|
|
117
93
|
"structured_output": true,
|
|
118
|
-
"
|
|
119
|
-
"
|
|
120
|
-
"last_updated": "2026-02-14",
|
|
94
|
+
"release_date": "2024-12-06",
|
|
95
|
+
"last_updated": "2024-12-06",
|
|
121
96
|
"modalities": {
|
|
122
97
|
"input": [
|
|
123
|
-
"text"
|
|
124
|
-
"image"
|
|
98
|
+
"text"
|
|
125
99
|
],
|
|
126
100
|
"output": [
|
|
127
101
|
"text"
|
|
128
102
|
]
|
|
129
103
|
},
|
|
130
|
-
"open_weights":
|
|
104
|
+
"open_weights": true,
|
|
131
105
|
"limit": {
|
|
132
|
-
"context":
|
|
133
|
-
"output":
|
|
106
|
+
"context": 131072,
|
|
107
|
+
"output": 16384
|
|
134
108
|
},
|
|
135
109
|
"cost": {
|
|
136
|
-
"input": 0.
|
|
137
|
-
"output":
|
|
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
|
-
]
|
|
110
|
+
"input": 0.1,
|
|
111
|
+
"output": 0.32
|
|
150
112
|
}
|
|
151
113
|
},
|
|
152
|
-
"
|
|
153
|
-
"id": "
|
|
154
|
-
"name": "
|
|
155
|
-
"description": "
|
|
114
|
+
"meta-llama/Llama-4-Scout-17B-16E-Instruct": {
|
|
115
|
+
"id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
116
|
+
"name": "Llama 4 Scout 17B",
|
|
117
|
+
"description": "Open multimodal Llama model for long-context analysis and efficient agents",
|
|
118
|
+
"family": "llama",
|
|
156
119
|
"attachment": true,
|
|
157
|
-
"reasoning":
|
|
158
|
-
"reasoning_options": [],
|
|
120
|
+
"reasoning": false,
|
|
159
121
|
"tool_call": true,
|
|
160
|
-
"
|
|
161
|
-
"
|
|
162
|
-
"
|
|
163
|
-
"last_updated": "2026-05-29",
|
|
122
|
+
"structured_output": true,
|
|
123
|
+
"release_date": "2025-04-05",
|
|
124
|
+
"last_updated": "2025-04-05",
|
|
164
125
|
"modalities": {
|
|
165
126
|
"input": [
|
|
166
127
|
"text",
|
|
167
|
-
"image"
|
|
168
|
-
"video"
|
|
128
|
+
"image"
|
|
169
129
|
],
|
|
170
130
|
"output": [
|
|
171
131
|
"text"
|
|
@@ -173,30 +133,29 @@
|
|
|
173
133
|
},
|
|
174
134
|
"open_weights": true,
|
|
175
135
|
"limit": {
|
|
176
|
-
"context":
|
|
177
|
-
"output":
|
|
136
|
+
"context": 327680,
|
|
137
|
+
"output": 16384
|
|
178
138
|
},
|
|
179
139
|
"cost": {
|
|
180
|
-
"input": 0.
|
|
181
|
-
"output":
|
|
182
|
-
"cache_read": 0.04
|
|
140
|
+
"input": 0.1,
|
|
141
|
+
"output": 0.3
|
|
183
142
|
}
|
|
184
143
|
},
|
|
185
|
-
"
|
|
186
|
-
"id": "
|
|
187
|
-
"name": "
|
|
188
|
-
"description": "Open
|
|
189
|
-
"family": "
|
|
190
|
-
"attachment":
|
|
144
|
+
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": {
|
|
145
|
+
"id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
146
|
+
"name": "Llama 4 Maverick 17B FP8",
|
|
147
|
+
"description": "Open multimodal Llama model for strong reasoning and fast responses",
|
|
148
|
+
"family": "llama",
|
|
149
|
+
"attachment": true,
|
|
191
150
|
"reasoning": false,
|
|
192
|
-
"tool_call":
|
|
151
|
+
"tool_call": false,
|
|
193
152
|
"structured_output": true,
|
|
194
|
-
"
|
|
195
|
-
"
|
|
196
|
-
"last_updated": "2024-12-26",
|
|
153
|
+
"release_date": "2025-04-05",
|
|
154
|
+
"last_updated": "2025-04-05",
|
|
197
155
|
"modalities": {
|
|
198
156
|
"input": [
|
|
199
|
-
"text"
|
|
157
|
+
"text",
|
|
158
|
+
"image"
|
|
200
159
|
],
|
|
201
160
|
"output": [
|
|
202
161
|
"text"
|
|
@@ -204,47 +163,37 @@
|
|
|
204
163
|
},
|
|
205
164
|
"open_weights": true,
|
|
206
165
|
"limit": {
|
|
207
|
-
"context":
|
|
208
|
-
"output":
|
|
166
|
+
"context": 1048576,
|
|
167
|
+
"output": 16384
|
|
209
168
|
},
|
|
210
169
|
"cost": {
|
|
211
|
-
"input": 0.
|
|
212
|
-
"output": 0.
|
|
170
|
+
"input": 0.2,
|
|
171
|
+
"output": 0.8
|
|
213
172
|
}
|
|
214
173
|
},
|
|
215
|
-
"
|
|
216
|
-
"id": "
|
|
217
|
-
"name": "
|
|
218
|
-
"description": "
|
|
219
|
-
"family": "
|
|
220
|
-
"attachment":
|
|
174
|
+
"XiaomiMiMo/MiMo-V2.6-Pro": {
|
|
175
|
+
"id": "XiaomiMiMo/MiMo-V2.6-Pro",
|
|
176
|
+
"name": "MiMo-V2.6-Pro",
|
|
177
|
+
"description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution",
|
|
178
|
+
"family": "mimo",
|
|
179
|
+
"attachment": true,
|
|
221
180
|
"reasoning": true,
|
|
222
181
|
"reasoning_options": [
|
|
223
182
|
{
|
|
224
183
|
"type": "toggle"
|
|
225
|
-
},
|
|
226
|
-
{
|
|
227
|
-
"type": "effort",
|
|
228
|
-
"values": [
|
|
229
|
-
"low",
|
|
230
|
-
"medium",
|
|
231
|
-
"high",
|
|
232
|
-
"xhigh"
|
|
233
|
-
]
|
|
234
184
|
}
|
|
235
185
|
],
|
|
236
186
|
"tool_call": true,
|
|
237
|
-
"interleaved": {
|
|
238
|
-
"field": "reasoning_content"
|
|
239
|
-
},
|
|
240
187
|
"structured_output": true,
|
|
241
188
|
"temperature": true,
|
|
242
|
-
"
|
|
243
|
-
"
|
|
244
|
-
"last_updated": "2026-04-24",
|
|
189
|
+
"release_date": "2026-09-22",
|
|
190
|
+
"last_updated": "2026-09-22",
|
|
245
191
|
"modalities": {
|
|
246
192
|
"input": [
|
|
247
|
-
"text"
|
|
193
|
+
"text",
|
|
194
|
+
"image",
|
|
195
|
+
"audio",
|
|
196
|
+
"video"
|
|
248
197
|
],
|
|
249
198
|
"output": [
|
|
250
199
|
"text"
|
|
@@ -253,29 +202,40 @@
|
|
|
253
202
|
"open_weights": true,
|
|
254
203
|
"limit": {
|
|
255
204
|
"context": 1048576,
|
|
256
|
-
"output":
|
|
205
|
+
"output": 131072
|
|
257
206
|
},
|
|
258
207
|
"cost": {
|
|
259
|
-
"input": 0.
|
|
260
|
-
"output": 0.
|
|
261
|
-
"cache_read": 0.
|
|
262
|
-
}
|
|
208
|
+
"input": 0.43,
|
|
209
|
+
"output": 0.87,
|
|
210
|
+
"cache_read": 0.0036
|
|
211
|
+
},
|
|
212
|
+
"canonical_model_id": "xiaomi/mimo-v2.6-pro"
|
|
263
213
|
},
|
|
264
|
-
"
|
|
265
|
-
"id": "
|
|
266
|
-
"name": "
|
|
267
|
-
"description": "
|
|
268
|
-
"family": "
|
|
269
|
-
"attachment":
|
|
270
|
-
"reasoning":
|
|
214
|
+
"XiaomiMiMo/MiMo-V2.5-Pro": {
|
|
215
|
+
"id": "XiaomiMiMo/MiMo-V2.5-Pro",
|
|
216
|
+
"name": "MiMo-V2.5-Pro",
|
|
217
|
+
"description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution",
|
|
218
|
+
"family": "mimo",
|
|
219
|
+
"attachment": true,
|
|
220
|
+
"reasoning": true,
|
|
221
|
+
"reasoning_options": [
|
|
222
|
+
{
|
|
223
|
+
"type": "toggle"
|
|
224
|
+
}
|
|
225
|
+
],
|
|
271
226
|
"tool_call": true,
|
|
227
|
+
"interleaved": {
|
|
228
|
+
"field": "reasoning_content"
|
|
229
|
+
},
|
|
272
230
|
"structured_output": true,
|
|
273
231
|
"temperature": true,
|
|
274
|
-
"
|
|
275
|
-
"
|
|
232
|
+
"knowledge": "2024-12",
|
|
233
|
+
"release_date": "2026-04-22",
|
|
234
|
+
"last_updated": "2026-04-22",
|
|
276
235
|
"modalities": {
|
|
277
236
|
"input": [
|
|
278
|
-
"text"
|
|
237
|
+
"text",
|
|
238
|
+
"audio"
|
|
279
239
|
],
|
|
280
240
|
"output": [
|
|
281
241
|
"text"
|
|
@@ -283,42 +243,39 @@
|
|
|
283
243
|
},
|
|
284
244
|
"open_weights": true,
|
|
285
245
|
"limit": {
|
|
286
|
-
"context":
|
|
287
|
-
"output":
|
|
246
|
+
"context": 1048576,
|
|
247
|
+
"output": 16384
|
|
288
248
|
},
|
|
289
249
|
"cost": {
|
|
290
|
-
"input":
|
|
291
|
-
"output":
|
|
292
|
-
"cache_read": 0.
|
|
293
|
-
}
|
|
250
|
+
"input": 1,
|
|
251
|
+
"output": 3,
|
|
252
|
+
"cache_read": 0.2
|
|
253
|
+
},
|
|
254
|
+
"canonical_model_id": "xiaomi/mimo-v2.5-pro"
|
|
294
255
|
},
|
|
295
|
-
"
|
|
296
|
-
"id": "
|
|
297
|
-
"name": "
|
|
298
|
-
"description": "
|
|
299
|
-
"family": "
|
|
256
|
+
"XiaomiMiMo/MiMo-V2.6-Flash": {
|
|
257
|
+
"id": "XiaomiMiMo/MiMo-V2.6-Flash",
|
|
258
|
+
"name": "MiMo-V2.6-Flash",
|
|
259
|
+
"description": "MiMo Flash model for multimodal coding agents and long-context automation",
|
|
260
|
+
"family": "mimo",
|
|
300
261
|
"attachment": true,
|
|
301
262
|
"reasoning": true,
|
|
302
263
|
"reasoning_options": [
|
|
303
264
|
{
|
|
304
|
-
"type": "
|
|
305
|
-
"values": [
|
|
306
|
-
"none",
|
|
307
|
-
"low",
|
|
308
|
-
"high",
|
|
309
|
-
"max"
|
|
310
|
-
]
|
|
265
|
+
"type": "toggle"
|
|
311
266
|
}
|
|
312
267
|
],
|
|
313
268
|
"tool_call": true,
|
|
314
269
|
"structured_output": true,
|
|
315
270
|
"temperature": true,
|
|
316
|
-
"release_date": "2026-
|
|
317
|
-
"last_updated": "2026-09-
|
|
271
|
+
"release_date": "2026-09-22",
|
|
272
|
+
"last_updated": "2026-09-22",
|
|
318
273
|
"modalities": {
|
|
319
274
|
"input": [
|
|
320
275
|
"text",
|
|
321
|
-
"image"
|
|
276
|
+
"image",
|
|
277
|
+
"audio",
|
|
278
|
+
"video"
|
|
322
279
|
],
|
|
323
280
|
"output": [
|
|
324
281
|
"text"
|
|
@@ -327,20 +284,21 @@
|
|
|
327
284
|
"open_weights": true,
|
|
328
285
|
"limit": {
|
|
329
286
|
"context": 1048576,
|
|
330
|
-
"output":
|
|
287
|
+
"output": 131072
|
|
331
288
|
},
|
|
332
289
|
"cost": {
|
|
333
|
-
"input": 0.
|
|
334
|
-
"output":
|
|
335
|
-
"cache_read": 0.
|
|
336
|
-
}
|
|
290
|
+
"input": 0.14,
|
|
291
|
+
"output": 0.28,
|
|
292
|
+
"cache_read": 0.0028
|
|
293
|
+
},
|
|
294
|
+
"canonical_model_id": "xiaomi/mimo-v2.6-flash"
|
|
337
295
|
},
|
|
338
|
-
"
|
|
339
|
-
"id": "
|
|
340
|
-
"name": "
|
|
341
|
-
"description": "
|
|
342
|
-
"family": "
|
|
343
|
-
"attachment":
|
|
296
|
+
"XiaomiMiMo/MiMo-V2.5": {
|
|
297
|
+
"id": "XiaomiMiMo/MiMo-V2.5",
|
|
298
|
+
"name": "MiMo-V2.5",
|
|
299
|
+
"description": "Open MiMo model for multimodal coding agents and long-context automation",
|
|
300
|
+
"family": "mimo",
|
|
301
|
+
"attachment": true,
|
|
344
302
|
"reasoning": true,
|
|
345
303
|
"reasoning_options": [
|
|
346
304
|
{
|
|
@@ -348,14 +306,20 @@
|
|
|
348
306
|
}
|
|
349
307
|
],
|
|
350
308
|
"tool_call": true,
|
|
309
|
+
"interleaved": {
|
|
310
|
+
"field": "reasoning_content"
|
|
311
|
+
},
|
|
351
312
|
"structured_output": true,
|
|
352
313
|
"temperature": true,
|
|
353
|
-
"knowledge": "
|
|
354
|
-
"release_date": "2026-
|
|
355
|
-
"last_updated": "2026-
|
|
314
|
+
"knowledge": "2024-12",
|
|
315
|
+
"release_date": "2026-04-22",
|
|
316
|
+
"last_updated": "2026-04-22",
|
|
356
317
|
"modalities": {
|
|
357
318
|
"input": [
|
|
358
|
-
"text"
|
|
319
|
+
"text",
|
|
320
|
+
"image",
|
|
321
|
+
"audio",
|
|
322
|
+
"video"
|
|
359
323
|
],
|
|
360
324
|
"output": [
|
|
361
325
|
"text"
|
|
@@ -363,35 +327,34 @@
|
|
|
363
327
|
},
|
|
364
328
|
"open_weights": true,
|
|
365
329
|
"limit": {
|
|
366
|
-
"context":
|
|
367
|
-
"output":
|
|
330
|
+
"context": 262144,
|
|
331
|
+
"output": 16384
|
|
368
332
|
},
|
|
369
333
|
"cost": {
|
|
370
|
-
"input": 0.
|
|
371
|
-
"output": 0.
|
|
372
|
-
"cache_read": 0.
|
|
373
|
-
}
|
|
334
|
+
"input": 0.14,
|
|
335
|
+
"output": 0.28,
|
|
336
|
+
"cache_read": 0.0028
|
|
337
|
+
},
|
|
338
|
+
"canonical_model_id": "xiaomi/mimo-v2.5"
|
|
374
339
|
},
|
|
375
|
-
"
|
|
376
|
-
"id": "
|
|
377
|
-
"name": "
|
|
378
|
-
"description": "
|
|
379
|
-
"family": "
|
|
380
|
-
"attachment":
|
|
340
|
+
"thinkingmachines/Inkling-Small": {
|
|
341
|
+
"id": "thinkingmachines/Inkling-Small",
|
|
342
|
+
"name": "Inkling Small",
|
|
343
|
+
"description": "Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio",
|
|
344
|
+
"family": "ling",
|
|
345
|
+
"attachment": true,
|
|
381
346
|
"reasoning": true,
|
|
382
|
-
"reasoning_options": [
|
|
383
|
-
{
|
|
384
|
-
"type": "toggle"
|
|
385
|
-
}
|
|
386
|
-
],
|
|
347
|
+
"reasoning_options": [],
|
|
387
348
|
"tool_call": true,
|
|
388
349
|
"structured_output": true,
|
|
389
350
|
"temperature": true,
|
|
390
|
-
"release_date": "
|
|
391
|
-
"last_updated": "
|
|
351
|
+
"release_date": "2026-07-30",
|
|
352
|
+
"last_updated": "2026-07-30",
|
|
392
353
|
"modalities": {
|
|
393
354
|
"input": [
|
|
394
|
-
"text"
|
|
355
|
+
"text",
|
|
356
|
+
"image",
|
|
357
|
+
"audio"
|
|
395
358
|
],
|
|
396
359
|
"output": [
|
|
397
360
|
"text"
|
|
@@ -399,75 +362,64 @@
|
|
|
399
362
|
},
|
|
400
363
|
"open_weights": true,
|
|
401
364
|
"limit": {
|
|
402
|
-
"context":
|
|
403
|
-
"output":
|
|
365
|
+
"context": 524288,
|
|
366
|
+
"output": 1048576
|
|
404
367
|
},
|
|
405
368
|
"cost": {
|
|
406
|
-
"input": 0.
|
|
407
|
-
"output":
|
|
408
|
-
"cache_read": 0.
|
|
409
|
-
}
|
|
369
|
+
"input": 0.45,
|
|
370
|
+
"output": 1.2,
|
|
371
|
+
"cache_read": 0.1
|
|
372
|
+
},
|
|
373
|
+
"canonical_model_id": "thinkingmachines/inkling-small"
|
|
410
374
|
},
|
|
411
|
-
"
|
|
412
|
-
"id": "
|
|
413
|
-
"name": "
|
|
414
|
-
"description": "
|
|
415
|
-
"
|
|
375
|
+
"thinkingmachines/Inkling": {
|
|
376
|
+
"id": "thinkingmachines/Inkling",
|
|
377
|
+
"name": "Inkling",
|
|
378
|
+
"description": "Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio",
|
|
379
|
+
"family": "ling",
|
|
380
|
+
"attachment": true,
|
|
416
381
|
"reasoning": true,
|
|
417
382
|
"reasoning_options": [],
|
|
418
383
|
"tool_call": true,
|
|
419
|
-
"interleaved": {
|
|
420
|
-
"field": "reasoning_content"
|
|
421
|
-
},
|
|
422
384
|
"structured_output": true,
|
|
423
385
|
"temperature": true,
|
|
424
|
-
"
|
|
425
|
-
"
|
|
426
|
-
"last_updated": "2025-05-28",
|
|
386
|
+
"release_date": "2026-07-15",
|
|
387
|
+
"last_updated": "2026-07-15",
|
|
427
388
|
"modalities": {
|
|
428
389
|
"input": [
|
|
429
|
-
"text"
|
|
390
|
+
"text",
|
|
391
|
+
"image",
|
|
392
|
+
"audio"
|
|
430
393
|
],
|
|
431
394
|
"output": [
|
|
432
395
|
"text"
|
|
433
396
|
]
|
|
434
397
|
},
|
|
435
|
-
"open_weights":
|
|
398
|
+
"open_weights": true,
|
|
436
399
|
"limit": {
|
|
437
|
-
"context":
|
|
438
|
-
"output":
|
|
400
|
+
"context": 524288,
|
|
401
|
+
"output": 1048576
|
|
439
402
|
},
|
|
440
403
|
"cost": {
|
|
441
|
-
"input": 0.
|
|
442
|
-
"output":
|
|
443
|
-
"cache_read": 0.
|
|
444
|
-
}
|
|
404
|
+
"input": 0.95,
|
|
405
|
+
"output": 4.05,
|
|
406
|
+
"cache_read": 0.16
|
|
407
|
+
},
|
|
408
|
+
"canonical_model_id": "thinkingmachines/inkling"
|
|
445
409
|
},
|
|
446
|
-
"
|
|
447
|
-
"id": "
|
|
448
|
-
"name": "
|
|
449
|
-
"description": "
|
|
450
|
-
"family": "
|
|
410
|
+
"google/gemma-3-27b-it": {
|
|
411
|
+
"id": "google/gemma-3-27b-it",
|
|
412
|
+
"name": "Gemma 3 27B IT",
|
|
413
|
+
"description": "Largest open Gemma 3 instruction model for multilingual text generation and visual understanding",
|
|
414
|
+
"family": "gemma",
|
|
451
415
|
"attachment": true,
|
|
452
|
-
"reasoning":
|
|
453
|
-
"reasoning_options": [
|
|
454
|
-
{
|
|
455
|
-
"type": "effort",
|
|
456
|
-
"values": [
|
|
457
|
-
"none",
|
|
458
|
-
"low",
|
|
459
|
-
"high",
|
|
460
|
-
"xhigh",
|
|
461
|
-
"max"
|
|
462
|
-
]
|
|
463
|
-
}
|
|
464
|
-
],
|
|
416
|
+
"reasoning": false,
|
|
465
417
|
"tool_call": true,
|
|
466
418
|
"structured_output": true,
|
|
467
419
|
"temperature": true,
|
|
468
|
-
"knowledge": "
|
|
469
|
-
"release_date": "
|
|
470
|
-
"last_updated": "
|
|
420
|
+
"knowledge": "2024-08",
|
|
421
|
+
"release_date": "2025-03-12",
|
|
422
|
+
"last_updated": "2025-03-12",
|
|
471
423
|
"modalities": {
|
|
472
424
|
"input": [
|
|
473
425
|
"text",
|
|
@@ -479,42 +431,32 @@
|
|
|
479
431
|
},
|
|
480
432
|
"open_weights": true,
|
|
481
433
|
"limit": {
|
|
482
|
-
"context":
|
|
483
|
-
"output":
|
|
434
|
+
"context": 131072,
|
|
435
|
+
"output": 131072
|
|
484
436
|
},
|
|
485
437
|
"cost": {
|
|
486
|
-
"input": 0.
|
|
487
|
-
"output": 0.
|
|
488
|
-
|
|
489
|
-
|
|
438
|
+
"input": 0.08,
|
|
439
|
+
"output": 0.16
|
|
440
|
+
},
|
|
441
|
+
"canonical_model_id": "google/gemma-3-27b-it"
|
|
490
442
|
},
|
|
491
|
-
"
|
|
492
|
-
"id": "
|
|
493
|
-
"name": "
|
|
494
|
-
"description": "
|
|
495
|
-
"family": "
|
|
496
|
-
"attachment":
|
|
497
|
-
"reasoning":
|
|
498
|
-
"reasoning_options": [
|
|
499
|
-
{
|
|
500
|
-
"type": "toggle"
|
|
501
|
-
},
|
|
502
|
-
{
|
|
503
|
-
"type": "effort",
|
|
504
|
-
"values": [
|
|
505
|
-
"high",
|
|
506
|
-
"max"
|
|
507
|
-
]
|
|
508
|
-
}
|
|
509
|
-
],
|
|
443
|
+
"google/gemma-3-12b-it": {
|
|
444
|
+
"id": "google/gemma-3-12b-it",
|
|
445
|
+
"name": "Gemma 3 12B IT",
|
|
446
|
+
"description": "Open multimodal Gemma instruction model for multilingual text generation and image understanding",
|
|
447
|
+
"family": "gemma",
|
|
448
|
+
"attachment": true,
|
|
449
|
+
"reasoning": false,
|
|
510
450
|
"tool_call": true,
|
|
511
451
|
"structured_output": true,
|
|
512
452
|
"temperature": true,
|
|
513
|
-
"
|
|
514
|
-
"
|
|
453
|
+
"knowledge": "2024-08",
|
|
454
|
+
"release_date": "2025-03-12",
|
|
455
|
+
"last_updated": "2025-03-12",
|
|
515
456
|
"modalities": {
|
|
516
457
|
"input": [
|
|
517
|
-
"text"
|
|
458
|
+
"text",
|
|
459
|
+
"image"
|
|
518
460
|
],
|
|
519
461
|
"output": [
|
|
520
462
|
"text"
|
|
@@ -522,154 +464,32 @@
|
|
|
522
464
|
},
|
|
523
465
|
"open_weights": true,
|
|
524
466
|
"limit": {
|
|
525
|
-
"context":
|
|
526
|
-
"output":
|
|
467
|
+
"context": 131072,
|
|
468
|
+
"output": 131072
|
|
527
469
|
},
|
|
528
470
|
"cost": {
|
|
529
|
-
"input":
|
|
530
|
-
"output":
|
|
531
|
-
"cache_read": 0.1
|
|
532
|
-
}
|
|
533
|
-
},
|
|
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",
|
|
538
|
-
"attachment": false,
|
|
539
|
-
"reasoning": true,
|
|
540
|
-
"reasoning_options": [
|
|
541
|
-
{
|
|
542
|
-
"type": "toggle"
|
|
543
|
-
}
|
|
544
|
-
],
|
|
545
|
-
"tool_call": true,
|
|
546
|
-
"interleaved": {
|
|
547
|
-
"field": "reasoning_content"
|
|
471
|
+
"input": 0.05,
|
|
472
|
+
"output": 0.15
|
|
548
473
|
},
|
|
549
|
-
"
|
|
550
|
-
"temperature": true,
|
|
551
|
-
"knowledge": "2024-12",
|
|
552
|
-
"release_date": "2025-12-02",
|
|
553
|
-
"last_updated": "2025-12-02",
|
|
554
|
-
"modalities": {
|
|
555
|
-
"input": [
|
|
556
|
-
"text"
|
|
557
|
-
],
|
|
558
|
-
"output": [
|
|
559
|
-
"text"
|
|
560
|
-
]
|
|
561
|
-
},
|
|
562
|
-
"open_weights": false,
|
|
563
|
-
"limit": {
|
|
564
|
-
"context": 163840,
|
|
565
|
-
"output": 64000
|
|
566
|
-
},
|
|
567
|
-
"cost": {
|
|
568
|
-
"input": 0.26,
|
|
569
|
-
"output": 0.38,
|
|
570
|
-
"cache_read": 0.13
|
|
571
|
-
}
|
|
572
|
-
},
|
|
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,
|
|
579
|
-
"reasoning": true,
|
|
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
|
-
],
|
|
594
|
-
"tool_call": true,
|
|
595
|
-
"interleaved": {
|
|
596
|
-
"field": "reasoning_content"
|
|
597
|
-
},
|
|
598
|
-
"structured_output": true,
|
|
599
|
-
"temperature": true,
|
|
600
|
-
"knowledge": "2025-05",
|
|
601
|
-
"release_date": "2026-04-24",
|
|
602
|
-
"last_updated": "2026-04-24",
|
|
603
|
-
"modalities": {
|
|
604
|
-
"input": [
|
|
605
|
-
"text"
|
|
606
|
-
],
|
|
607
|
-
"output": [
|
|
608
|
-
"text"
|
|
609
|
-
]
|
|
610
|
-
},
|
|
611
|
-
"open_weights": true,
|
|
612
|
-
"limit": {
|
|
613
|
-
"context": 1048576,
|
|
614
|
-
"output": 16384
|
|
615
|
-
},
|
|
616
|
-
"cost": {
|
|
617
|
-
"input": 1.3,
|
|
618
|
-
"output": 2.6,
|
|
619
|
-
"cache_read": 0.1
|
|
620
|
-
}
|
|
474
|
+
"canonical_model_id": "google/gemma-3-12b-it"
|
|
621
475
|
},
|
|
622
|
-
"
|
|
623
|
-
"id": "
|
|
624
|
-
"name": "
|
|
625
|
-
"description": "Open
|
|
626
|
-
"family": "
|
|
476
|
+
"google/gemma-3-4b-it": {
|
|
477
|
+
"id": "google/gemma-3-4b-it",
|
|
478
|
+
"name": "Gemma 3 4B IT",
|
|
479
|
+
"description": "Open multimodal Gemma instruction model for efficient text generation and image understanding",
|
|
480
|
+
"family": "gemma",
|
|
627
481
|
"attachment": true,
|
|
628
|
-
"reasoning":
|
|
629
|
-
"reasoning_options": [],
|
|
482
|
+
"reasoning": false,
|
|
630
483
|
"tool_call": true,
|
|
631
484
|
"structured_output": true,
|
|
632
485
|
"temperature": true,
|
|
633
|
-
"
|
|
634
|
-
"
|
|
486
|
+
"knowledge": "2024-08",
|
|
487
|
+
"release_date": "2025-03-12",
|
|
488
|
+
"last_updated": "2025-03-12",
|
|
635
489
|
"modalities": {
|
|
636
490
|
"input": [
|
|
637
491
|
"text",
|
|
638
|
-
"image"
|
|
639
|
-
"video",
|
|
640
|
-
"audio"
|
|
641
|
-
],
|
|
642
|
-
"output": [
|
|
643
|
-
"text"
|
|
644
|
-
]
|
|
645
|
-
},
|
|
646
|
-
"open_weights": true,
|
|
647
|
-
"limit": {
|
|
648
|
-
"context": 262144,
|
|
649
|
-
"output": 65536
|
|
650
|
-
},
|
|
651
|
-
"status": "deprecated",
|
|
652
|
-
"cost": {
|
|
653
|
-
"input": 0.2,
|
|
654
|
-
"output": 0.8
|
|
655
|
-
}
|
|
656
|
-
},
|
|
657
|
-
"nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": {
|
|
658
|
-
"id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5",
|
|
659
|
-
"name": "Llama 3.3 Nemotron Super 49B v1.5",
|
|
660
|
-
"description": "Nemotron model for efficient reasoning, coding, and specialized AI agents",
|
|
661
|
-
"family": "nemotron",
|
|
662
|
-
"attachment": false,
|
|
663
|
-
"reasoning": true,
|
|
664
|
-
"reasoning_options": [],
|
|
665
|
-
"tool_call": true,
|
|
666
|
-
"structured_output": true,
|
|
667
|
-
"temperature": true,
|
|
668
|
-
"release_date": "2025-07-25",
|
|
669
|
-
"last_updated": "2025-07-25",
|
|
670
|
-
"modalities": {
|
|
671
|
-
"input": [
|
|
672
|
-
"text"
|
|
492
|
+
"image"
|
|
673
493
|
],
|
|
674
494
|
"output": [
|
|
675
495
|
"text"
|
|
@@ -680,18 +500,18 @@
|
|
|
680
500
|
"context": 131072,
|
|
681
501
|
"output": 131072
|
|
682
502
|
},
|
|
683
|
-
"status": "deprecated",
|
|
684
503
|
"cost": {
|
|
685
|
-
"input": 0.
|
|
686
|
-
"output": 0.
|
|
687
|
-
}
|
|
504
|
+
"input": 0.05,
|
|
505
|
+
"output": 0.1
|
|
506
|
+
},
|
|
507
|
+
"canonical_model_id": "google/gemma-3-4b-it"
|
|
688
508
|
},
|
|
689
|
-
"
|
|
690
|
-
"id": "
|
|
691
|
-
"name": "
|
|
692
|
-
"description": "
|
|
693
|
-
"family": "
|
|
694
|
-
"attachment":
|
|
509
|
+
"google/gemma-4-31B-it": {
|
|
510
|
+
"id": "google/gemma-4-31B-it",
|
|
511
|
+
"name": "Gemma 4 31B IT",
|
|
512
|
+
"description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
|
|
513
|
+
"family": "gemma",
|
|
514
|
+
"attachment": true,
|
|
695
515
|
"reasoning": true,
|
|
696
516
|
"reasoning_options": [
|
|
697
517
|
{
|
|
@@ -699,45 +519,15 @@
|
|
|
699
519
|
}
|
|
700
520
|
],
|
|
701
521
|
"tool_call": true,
|
|
702
|
-
"temperature": true,
|
|
703
|
-
"release_date": "2025-12-15",
|
|
704
|
-
"last_updated": "2025-12-15",
|
|
705
|
-
"modalities": {
|
|
706
|
-
"input": [
|
|
707
|
-
"text"
|
|
708
|
-
],
|
|
709
|
-
"output": [
|
|
710
|
-
"text"
|
|
711
|
-
]
|
|
712
|
-
},
|
|
713
|
-
"open_weights": true,
|
|
714
|
-
"limit": {
|
|
715
|
-
"context": 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-3-4b-it": {
|
|
725
|
-
"id": "google/gemma-3-4b-it",
|
|
726
|
-
"name": "Gemma 3 4B IT",
|
|
727
|
-
"description": "Open multimodal Gemma instruction model for efficient text generation and image understanding",
|
|
728
|
-
"family": "gemma",
|
|
729
|
-
"attachment": true,
|
|
730
|
-
"reasoning": false,
|
|
731
|
-
"tool_call": true,
|
|
732
522
|
"structured_output": true,
|
|
733
523
|
"temperature": true,
|
|
734
|
-
"
|
|
735
|
-
"
|
|
736
|
-
"last_updated": "2025-03-12",
|
|
524
|
+
"release_date": "2026-04-02",
|
|
525
|
+
"last_updated": "2026-04-02",
|
|
737
526
|
"modalities": {
|
|
738
527
|
"input": [
|
|
739
528
|
"text",
|
|
740
|
-
"image"
|
|
529
|
+
"image",
|
|
530
|
+
"video"
|
|
741
531
|
],
|
|
742
532
|
"output": [
|
|
743
533
|
"text"
|
|
@@ -745,18 +535,19 @@
|
|
|
745
535
|
},
|
|
746
536
|
"open_weights": true,
|
|
747
537
|
"limit": {
|
|
748
|
-
"context":
|
|
749
|
-
"output":
|
|
538
|
+
"context": 262144,
|
|
539
|
+
"output": 32768
|
|
750
540
|
},
|
|
751
541
|
"cost": {
|
|
752
|
-
"input": 0.
|
|
753
|
-
"output": 0.
|
|
754
|
-
}
|
|
542
|
+
"input": 0.13,
|
|
543
|
+
"output": 0.38
|
|
544
|
+
},
|
|
545
|
+
"canonical_model_id": "google/gemma-4-31b-it"
|
|
755
546
|
},
|
|
756
|
-
"google/gemma-4-
|
|
757
|
-
"id": "google/gemma-4-
|
|
758
|
-
"name": "Gemma 4
|
|
759
|
-
"description": "
|
|
547
|
+
"google/gemma-4-26B-A4B-it": {
|
|
548
|
+
"id": "google/gemma-4-26B-A4B-it",
|
|
549
|
+
"name": "Gemma 4 26B A4B IT",
|
|
550
|
+
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
|
|
760
551
|
"family": "gemma",
|
|
761
552
|
"attachment": true,
|
|
762
553
|
"reasoning": true,
|
|
@@ -773,8 +564,7 @@
|
|
|
773
564
|
"modalities": {
|
|
774
565
|
"input": [
|
|
775
566
|
"text",
|
|
776
|
-
"image"
|
|
777
|
-
"video"
|
|
567
|
+
"image"
|
|
778
568
|
],
|
|
779
569
|
"output": [
|
|
780
570
|
"text"
|
|
@@ -786,9 +576,10 @@
|
|
|
786
576
|
"output": 32768
|
|
787
577
|
},
|
|
788
578
|
"cost": {
|
|
789
|
-
"input": 0.
|
|
790
|
-
"output": 0.
|
|
791
|
-
}
|
|
579
|
+
"input": 0.07,
|
|
580
|
+
"output": 0.34
|
|
581
|
+
},
|
|
582
|
+
"canonical_model_id": "google/gemma-4-26b-a4b-it"
|
|
792
583
|
},
|
|
793
584
|
"google/gemma-4-E4B-it": {
|
|
794
585
|
"id": "google/gemma-4-E4B-it",
|
|
@@ -825,21 +616,22 @@
|
|
|
825
616
|
"cost": {
|
|
826
617
|
"input": 0.02,
|
|
827
618
|
"output": 0.1
|
|
828
|
-
}
|
|
619
|
+
},
|
|
620
|
+
"canonical_model_id": "google/gemma-4-E4B-it"
|
|
829
621
|
},
|
|
830
|
-
"
|
|
831
|
-
"id": "
|
|
832
|
-
"name": "
|
|
833
|
-
"description": "
|
|
834
|
-
"family": "
|
|
622
|
+
"Qwen/Qwen3-VL-235B-A22B-Instruct": {
|
|
623
|
+
"id": "Qwen/Qwen3-VL-235B-A22B-Instruct",
|
|
624
|
+
"name": "Qwen3 VL 235B A22B Instruct",
|
|
625
|
+
"description": "Qwen vision-language instruct model for visual reasoning, documents, and agent tasks",
|
|
626
|
+
"family": "qwen",
|
|
835
627
|
"attachment": true,
|
|
836
628
|
"reasoning": false,
|
|
837
629
|
"tool_call": true,
|
|
838
630
|
"structured_output": true,
|
|
839
631
|
"temperature": true,
|
|
840
|
-
"knowledge": "
|
|
841
|
-
"release_date": "2025-
|
|
842
|
-
"last_updated": "2025-
|
|
632
|
+
"knowledge": "2025-03-31",
|
|
633
|
+
"release_date": "2025-09-23",
|
|
634
|
+
"last_updated": "2025-09-23",
|
|
843
635
|
"modalities": {
|
|
844
636
|
"input": [
|
|
845
637
|
"text",
|
|
@@ -851,35 +643,46 @@
|
|
|
851
643
|
},
|
|
852
644
|
"open_weights": true,
|
|
853
645
|
"limit": {
|
|
854
|
-
"context":
|
|
855
|
-
"output":
|
|
646
|
+
"context": 262144,
|
|
647
|
+
"output": 32768
|
|
856
648
|
},
|
|
857
649
|
"cost": {
|
|
858
|
-
"input": 0.
|
|
859
|
-
"output": 0.
|
|
860
|
-
|
|
650
|
+
"input": 0.2,
|
|
651
|
+
"output": 0.88,
|
|
652
|
+
"cache_read": 0.11
|
|
653
|
+
},
|
|
654
|
+
"canonical_model_id": "alibaba/qwen3-vl-235b-a22b-instruct"
|
|
861
655
|
},
|
|
862
|
-
"
|
|
863
|
-
"id": "
|
|
864
|
-
"name": "
|
|
865
|
-
"description": "
|
|
866
|
-
"family": "
|
|
656
|
+
"Qwen/Qwen3.8-27B": {
|
|
657
|
+
"id": "Qwen/Qwen3.8-27B",
|
|
658
|
+
"name": "Qwen3.8 27B",
|
|
659
|
+
"description": "Dense 27B vision-language model for coding, agent tasks, and image and video understanding",
|
|
660
|
+
"family": "qwen",
|
|
867
661
|
"attachment": true,
|
|
868
662
|
"reasoning": true,
|
|
869
663
|
"reasoning_options": [
|
|
870
664
|
{
|
|
871
665
|
"type": "toggle"
|
|
666
|
+
},
|
|
667
|
+
{
|
|
668
|
+
"type": "effort",
|
|
669
|
+
"values": [
|
|
670
|
+
"low",
|
|
671
|
+
"medium",
|
|
672
|
+
"xhigh"
|
|
673
|
+
]
|
|
872
674
|
}
|
|
873
675
|
],
|
|
874
676
|
"tool_call": true,
|
|
875
677
|
"structured_output": true,
|
|
876
678
|
"temperature": true,
|
|
877
|
-
"release_date": "2026-
|
|
878
|
-
"last_updated": "2026-
|
|
679
|
+
"release_date": "2026-08-14",
|
|
680
|
+
"last_updated": "2026-08-14",
|
|
879
681
|
"modalities": {
|
|
880
682
|
"input": [
|
|
881
683
|
"text",
|
|
882
|
-
"image"
|
|
684
|
+
"image",
|
|
685
|
+
"video"
|
|
883
686
|
],
|
|
884
687
|
"output": [
|
|
885
688
|
"text"
|
|
@@ -891,27 +694,28 @@
|
|
|
891
694
|
"output": 32768
|
|
892
695
|
},
|
|
893
696
|
"cost": {
|
|
894
|
-
"input": 0.
|
|
895
|
-
"output":
|
|
896
|
-
|
|
697
|
+
"input": 0.2,
|
|
698
|
+
"output": 2.5,
|
|
699
|
+
"cache_read": 0.05
|
|
700
|
+
},
|
|
701
|
+
"canonical_model_id": "alibaba/qwen3.8-27b"
|
|
897
702
|
},
|
|
898
|
-
"
|
|
899
|
-
"id": "
|
|
900
|
-
"name": "
|
|
901
|
-
"description": "
|
|
902
|
-
"family": "
|
|
903
|
-
"attachment":
|
|
703
|
+
"Qwen/Qwen3-Next-80B-A3B-Instruct": {
|
|
704
|
+
"id": "Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
705
|
+
"name": "Qwen3-Next 80B-A3B Instruct",
|
|
706
|
+
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
707
|
+
"family": "qwen",
|
|
708
|
+
"attachment": false,
|
|
904
709
|
"reasoning": false,
|
|
905
710
|
"tool_call": true,
|
|
906
711
|
"structured_output": true,
|
|
907
712
|
"temperature": true,
|
|
908
|
-
"knowledge": "
|
|
909
|
-
"release_date": "2025-
|
|
910
|
-
"last_updated": "2025-
|
|
713
|
+
"knowledge": "2025-04",
|
|
714
|
+
"release_date": "2025-09",
|
|
715
|
+
"last_updated": "2025-09",
|
|
911
716
|
"modalities": {
|
|
912
717
|
"input": [
|
|
913
|
-
"text"
|
|
914
|
-
"image"
|
|
718
|
+
"text"
|
|
915
719
|
],
|
|
916
720
|
"output": [
|
|
917
721
|
"text"
|
|
@@ -919,35 +723,27 @@
|
|
|
919
723
|
},
|
|
920
724
|
"open_weights": true,
|
|
921
725
|
"limit": {
|
|
922
|
-
"context":
|
|
923
|
-
"output":
|
|
726
|
+
"context": 262144,
|
|
727
|
+
"output": 32768
|
|
924
728
|
},
|
|
925
729
|
"cost": {
|
|
926
|
-
"input": 0.
|
|
927
|
-
"output":
|
|
928
|
-
}
|
|
730
|
+
"input": 0.09,
|
|
731
|
+
"output": 1.1
|
|
732
|
+
},
|
|
733
|
+
"canonical_model_id": "alibaba/qwen3-next-80b-a3b-instruct"
|
|
929
734
|
},
|
|
930
|
-
"
|
|
931
|
-
"id": "
|
|
932
|
-
"name": "
|
|
933
|
-
"description": "
|
|
934
|
-
"family": "
|
|
735
|
+
"Qwen/Qwen3.7-Max": {
|
|
736
|
+
"id": "Qwen/Qwen3.7-Max",
|
|
737
|
+
"name": "Qwen3.7 Max",
|
|
738
|
+
"description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
|
|
739
|
+
"family": "qwen",
|
|
935
740
|
"attachment": false,
|
|
936
|
-
"reasoning":
|
|
937
|
-
"reasoning_options": [
|
|
938
|
-
{
|
|
939
|
-
"type": "toggle"
|
|
940
|
-
}
|
|
941
|
-
],
|
|
741
|
+
"reasoning": false,
|
|
942
742
|
"tool_call": true,
|
|
943
|
-
"interleaved": {
|
|
944
|
-
"field": "reasoning_content"
|
|
945
|
-
},
|
|
946
743
|
"structured_output": true,
|
|
947
744
|
"temperature": true,
|
|
948
|
-
"
|
|
949
|
-
"
|
|
950
|
-
"last_updated": "2026-04-07",
|
|
745
|
+
"release_date": "2026-05-21",
|
|
746
|
+
"last_updated": "2026-05-21",
|
|
951
747
|
"modalities": {
|
|
952
748
|
"input": [
|
|
953
749
|
"text"
|
|
@@ -956,90 +752,92 @@
|
|
|
956
752
|
"text"
|
|
957
753
|
]
|
|
958
754
|
},
|
|
959
|
-
"open_weights":
|
|
755
|
+
"open_weights": false,
|
|
960
756
|
"limit": {
|
|
961
|
-
"context":
|
|
962
|
-
"output":
|
|
757
|
+
"context": 256000,
|
|
758
|
+
"output": 65536
|
|
963
759
|
},
|
|
964
760
|
"cost": {
|
|
965
|
-
"input":
|
|
966
|
-
"output":
|
|
967
|
-
"cache_read": 0.
|
|
968
|
-
|
|
761
|
+
"input": 2.5,
|
|
762
|
+
"output": 7.5,
|
|
763
|
+
"cache_read": 0.5,
|
|
764
|
+
"tiers": [
|
|
765
|
+
{
|
|
766
|
+
"input": 5,
|
|
767
|
+
"output": 15,
|
|
768
|
+
"cache_read": 1,
|
|
769
|
+
"tier": {
|
|
770
|
+
"type": "context",
|
|
771
|
+
"size": 32000
|
|
772
|
+
}
|
|
773
|
+
},
|
|
774
|
+
{
|
|
775
|
+
"input": 6.25,
|
|
776
|
+
"output": 18.5,
|
|
777
|
+
"cache_read": 1.25,
|
|
778
|
+
"tier": {
|
|
779
|
+
"type": "context",
|
|
780
|
+
"size": 128000
|
|
781
|
+
}
|
|
782
|
+
}
|
|
783
|
+
]
|
|
784
|
+
},
|
|
785
|
+
"canonical_model_id": "alibaba/qwen3.7-max"
|
|
969
786
|
},
|
|
970
|
-
"
|
|
971
|
-
"id": "
|
|
972
|
-
"name": "
|
|
973
|
-
"description": "
|
|
974
|
-
"family": "
|
|
975
|
-
"attachment":
|
|
976
|
-
"reasoning":
|
|
977
|
-
"reasoning_options": [
|
|
978
|
-
{
|
|
979
|
-
"type": "effort",
|
|
980
|
-
"values": [
|
|
981
|
-
"low",
|
|
982
|
-
"high",
|
|
983
|
-
"max"
|
|
984
|
-
]
|
|
985
|
-
}
|
|
986
|
-
],
|
|
787
|
+
"Qwen/Qwen3.8-Max": {
|
|
788
|
+
"id": "Qwen/Qwen3.8-Max",
|
|
789
|
+
"name": "Qwen3.8 Max",
|
|
790
|
+
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
|
|
791
|
+
"family": "qwen",
|
|
792
|
+
"attachment": true,
|
|
793
|
+
"reasoning": false,
|
|
987
794
|
"tool_call": true,
|
|
988
795
|
"structured_output": true,
|
|
989
796
|
"temperature": true,
|
|
990
|
-
"release_date": "2026-08-
|
|
991
|
-
"last_updated": "2026-08-
|
|
797
|
+
"release_date": "2026-08-03",
|
|
798
|
+
"last_updated": "2026-08-03",
|
|
992
799
|
"modalities": {
|
|
993
800
|
"input": [
|
|
994
|
-
"text"
|
|
801
|
+
"text",
|
|
802
|
+
"image",
|
|
803
|
+
"video",
|
|
804
|
+
"pdf"
|
|
995
805
|
],
|
|
996
806
|
"output": [
|
|
997
807
|
"text"
|
|
998
808
|
]
|
|
999
809
|
},
|
|
1000
|
-
"open_weights":
|
|
810
|
+
"open_weights": false,
|
|
1001
811
|
"limit": {
|
|
1002
|
-
"context":
|
|
812
|
+
"context": 256000,
|
|
1003
813
|
"output": 131072
|
|
1004
814
|
},
|
|
1005
815
|
"cost": {
|
|
1006
|
-
"input": 1.
|
|
1007
|
-
"output": 4,
|
|
1008
|
-
"cache_read": 0.
|
|
1009
|
-
}
|
|
816
|
+
"input": 1.65,
|
|
817
|
+
"output": 4.951,
|
|
818
|
+
"cache_read": 0.206
|
|
819
|
+
},
|
|
820
|
+
"canonical_model_id": "alibaba/qwen3.8-max"
|
|
1010
821
|
},
|
|
1011
|
-
"
|
|
1012
|
-
"id": "
|
|
1013
|
-
"name": "
|
|
1014
|
-
"description": "
|
|
1015
|
-
"family": "
|
|
1016
|
-
"attachment":
|
|
822
|
+
"Qwen/Qwen3.5-27B": {
|
|
823
|
+
"id": "Qwen/Qwen3.5-27B",
|
|
824
|
+
"name": "Qwen3.5 27B",
|
|
825
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
826
|
+
"family": "qwen",
|
|
827
|
+
"attachment": true,
|
|
1017
828
|
"reasoning": true,
|
|
1018
|
-
"reasoning_options": [
|
|
1019
|
-
{
|
|
1020
|
-
"type": "toggle"
|
|
1021
|
-
},
|
|
1022
|
-
{
|
|
1023
|
-
"type": "effort",
|
|
1024
|
-
"values": [
|
|
1025
|
-
"low",
|
|
1026
|
-
"medium",
|
|
1027
|
-
"high",
|
|
1028
|
-
"xhigh"
|
|
1029
|
-
]
|
|
1030
|
-
}
|
|
1031
|
-
],
|
|
829
|
+
"reasoning_options": [],
|
|
1032
830
|
"tool_call": true,
|
|
1033
|
-
"interleaved": {
|
|
1034
|
-
"field": "reasoning_content"
|
|
1035
|
-
},
|
|
1036
831
|
"structured_output": true,
|
|
1037
832
|
"temperature": true,
|
|
1038
|
-
"release_date": "2026-
|
|
1039
|
-
"last_updated": "2026-
|
|
833
|
+
"release_date": "2026-02-23",
|
|
834
|
+
"last_updated": "2026-02-23",
|
|
1040
835
|
"modalities": {
|
|
1041
836
|
"input": [
|
|
1042
|
-
"text"
|
|
837
|
+
"text",
|
|
838
|
+
"image",
|
|
839
|
+
"video",
|
|
840
|
+
"audio"
|
|
1043
841
|
],
|
|
1044
842
|
"output": [
|
|
1045
843
|
"text"
|
|
@@ -1047,35 +845,33 @@
|
|
|
1047
845
|
},
|
|
1048
846
|
"open_weights": true,
|
|
1049
847
|
"limit": {
|
|
1050
|
-
"context":
|
|
1051
|
-
"output":
|
|
848
|
+
"context": 262144,
|
|
849
|
+
"output": 65536
|
|
1052
850
|
},
|
|
1053
851
|
"cost": {
|
|
1054
|
-
"input": 0.
|
|
1055
|
-
"output": 2.
|
|
1056
|
-
|
|
1057
|
-
|
|
852
|
+
"input": 0.26,
|
|
853
|
+
"output": 2.6
|
|
854
|
+
},
|
|
855
|
+
"canonical_model_id": "alibaba/qwen3.5-27b"
|
|
1058
856
|
},
|
|
1059
|
-
"
|
|
1060
|
-
"id": "
|
|
1061
|
-
"name": "
|
|
1062
|
-
"description": "
|
|
1063
|
-
"family": "
|
|
1064
|
-
"attachment":
|
|
857
|
+
"Qwen/Qwen3.6-35B-A3B": {
|
|
858
|
+
"id": "Qwen/Qwen3.6-35B-A3B",
|
|
859
|
+
"name": "Qwen3.6 35B A3B",
|
|
860
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
861
|
+
"family": "qwen",
|
|
862
|
+
"attachment": true,
|
|
1065
863
|
"reasoning": true,
|
|
1066
864
|
"reasoning_options": [],
|
|
1067
865
|
"tool_call": true,
|
|
1068
|
-
"interleaved": {
|
|
1069
|
-
"field": "reasoning_content"
|
|
1070
|
-
},
|
|
1071
866
|
"structured_output": true,
|
|
1072
867
|
"temperature": true,
|
|
1073
|
-
"
|
|
1074
|
-
"
|
|
1075
|
-
"last_updated": "2026-01-19",
|
|
868
|
+
"release_date": "2026-04-01",
|
|
869
|
+
"last_updated": "2026-04-01",
|
|
1076
870
|
"modalities": {
|
|
1077
871
|
"input": [
|
|
1078
|
-
"text"
|
|
872
|
+
"text",
|
|
873
|
+
"image",
|
|
874
|
+
"video"
|
|
1079
875
|
],
|
|
1080
876
|
"output": [
|
|
1081
877
|
"text"
|
|
@@ -1083,37 +879,27 @@
|
|
|
1083
879
|
},
|
|
1084
880
|
"open_weights": true,
|
|
1085
881
|
"limit": {
|
|
1086
|
-
"context":
|
|
1087
|
-
"output":
|
|
882
|
+
"context": 262144,
|
|
883
|
+
"output": 81920
|
|
1088
884
|
},
|
|
1089
|
-
"status": "deprecated",
|
|
1090
885
|
"cost": {
|
|
1091
|
-
"input": 0.
|
|
1092
|
-
"output": 0.
|
|
1093
|
-
"cache_read": 0.01
|
|
886
|
+
"input": 0.1,
|
|
887
|
+
"output": 0.95
|
|
1094
888
|
}
|
|
1095
889
|
},
|
|
1096
|
-
"
|
|
1097
|
-
"id": "
|
|
1098
|
-
"name": "
|
|
1099
|
-
"description": "
|
|
1100
|
-
"family": "
|
|
890
|
+
"Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": {
|
|
891
|
+
"id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo",
|
|
892
|
+
"name": "Qwen3 Coder 480B A35B Instruct Turbo",
|
|
893
|
+
"description": "Qwen coding model for software agents, repository edits, and code reasoning",
|
|
894
|
+
"family": "qwen",
|
|
1101
895
|
"attachment": false,
|
|
1102
|
-
"reasoning":
|
|
1103
|
-
"reasoning_options": [
|
|
1104
|
-
{
|
|
1105
|
-
"type": "toggle"
|
|
1106
|
-
}
|
|
1107
|
-
],
|
|
896
|
+
"reasoning": false,
|
|
1108
897
|
"tool_call": true,
|
|
1109
|
-
"interleaved": {
|
|
1110
|
-
"field": "reasoning_content"
|
|
1111
|
-
},
|
|
1112
898
|
"structured_output": true,
|
|
1113
899
|
"temperature": true,
|
|
1114
900
|
"knowledge": "2025-04",
|
|
1115
|
-
"release_date": "2025-
|
|
1116
|
-
"last_updated": "2025-
|
|
901
|
+
"release_date": "2025-07-23",
|
|
902
|
+
"last_updated": "2025-07-23",
|
|
1117
903
|
"modalities": {
|
|
1118
904
|
"input": [
|
|
1119
905
|
"text"
|
|
@@ -1124,36 +910,28 @@
|
|
|
1124
910
|
},
|
|
1125
911
|
"open_weights": true,
|
|
1126
912
|
"limit": {
|
|
1127
|
-
"context":
|
|
1128
|
-
"output":
|
|
913
|
+
"context": 262144,
|
|
914
|
+
"output": 66536
|
|
1129
915
|
},
|
|
1130
916
|
"cost": {
|
|
1131
|
-
"input": 0.
|
|
1132
|
-
"output": 1
|
|
1133
|
-
"cache_read": 0.
|
|
917
|
+
"input": 0.3,
|
|
918
|
+
"output": 1,
|
|
919
|
+
"cache_read": 0.1
|
|
1134
920
|
}
|
|
1135
921
|
},
|
|
1136
|
-
"
|
|
1137
|
-
"id": "
|
|
1138
|
-
"name": "
|
|
1139
|
-
"description": "Flagship
|
|
1140
|
-
"family": "
|
|
922
|
+
"Qwen/Qwen3-Max": {
|
|
923
|
+
"id": "Qwen/Qwen3-Max",
|
|
924
|
+
"name": "Qwen3 Max",
|
|
925
|
+
"description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use",
|
|
926
|
+
"family": "qwen",
|
|
1141
927
|
"attachment": false,
|
|
1142
|
-
"reasoning":
|
|
1143
|
-
"reasoning_options": [
|
|
1144
|
-
{
|
|
1145
|
-
"type": "toggle"
|
|
1146
|
-
}
|
|
1147
|
-
],
|
|
928
|
+
"reasoning": false,
|
|
1148
929
|
"tool_call": true,
|
|
1149
|
-
"interleaved": {
|
|
1150
|
-
"field": "reasoning_content"
|
|
1151
|
-
},
|
|
1152
930
|
"structured_output": true,
|
|
1153
931
|
"temperature": true,
|
|
1154
|
-
"knowledge": "2025-
|
|
1155
|
-
"release_date": "
|
|
1156
|
-
"last_updated": "
|
|
932
|
+
"knowledge": "2025-04",
|
|
933
|
+
"release_date": "2025-09-23",
|
|
934
|
+
"last_updated": "2025-09-23",
|
|
1157
935
|
"modalities": {
|
|
1158
936
|
"input": [
|
|
1159
937
|
"text"
|
|
@@ -1162,39 +940,50 @@
|
|
|
1162
940
|
"text"
|
|
1163
941
|
]
|
|
1164
942
|
},
|
|
1165
|
-
"open_weights":
|
|
943
|
+
"open_weights": false,
|
|
1166
944
|
"limit": {
|
|
1167
|
-
"context":
|
|
1168
|
-
"output":
|
|
945
|
+
"context": 256000,
|
|
946
|
+
"output": 65536
|
|
1169
947
|
},
|
|
1170
|
-
"status": "deprecated",
|
|
1171
948
|
"cost": {
|
|
1172
|
-
"input":
|
|
1173
|
-
"output":
|
|
1174
|
-
"cache_read": 0.
|
|
1175
|
-
|
|
949
|
+
"input": 1.2,
|
|
950
|
+
"output": 6,
|
|
951
|
+
"cache_read": 0.24,
|
|
952
|
+
"tiers": [
|
|
953
|
+
{
|
|
954
|
+
"input": 2.4,
|
|
955
|
+
"output": 12,
|
|
956
|
+
"cache_read": 0.48,
|
|
957
|
+
"tier": {
|
|
958
|
+
"type": "context",
|
|
959
|
+
"size": 32000
|
|
960
|
+
}
|
|
961
|
+
},
|
|
962
|
+
{
|
|
963
|
+
"input": 3,
|
|
964
|
+
"output": 15,
|
|
965
|
+
"cache_read": 0.6,
|
|
966
|
+
"tier": {
|
|
967
|
+
"type": "context",
|
|
968
|
+
"size": 128000
|
|
969
|
+
}
|
|
970
|
+
}
|
|
971
|
+
]
|
|
972
|
+
},
|
|
973
|
+
"canonical_model_id": "alibaba/qwen3-max"
|
|
1176
974
|
},
|
|
1177
|
-
"
|
|
1178
|
-
"id": "
|
|
1179
|
-
"name": "
|
|
1180
|
-
"description": "
|
|
1181
|
-
"family": "
|
|
975
|
+
"Qwen/Qwen3-235B-A22B-Instruct-2507": {
|
|
976
|
+
"id": "Qwen/Qwen3-235B-A22B-Instruct-2507",
|
|
977
|
+
"name": "Qwen3 235B-A22B Instruct 2507",
|
|
978
|
+
"description": "Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use",
|
|
979
|
+
"family": "qwen",
|
|
1182
980
|
"attachment": false,
|
|
1183
|
-
"reasoning":
|
|
1184
|
-
"reasoning_options": [
|
|
1185
|
-
{
|
|
1186
|
-
"type": "toggle"
|
|
1187
|
-
}
|
|
1188
|
-
],
|
|
981
|
+
"reasoning": false,
|
|
1189
982
|
"tool_call": true,
|
|
1190
|
-
"interleaved": {
|
|
1191
|
-
"field": "reasoning_content"
|
|
1192
|
-
},
|
|
1193
983
|
"structured_output": true,
|
|
1194
984
|
"temperature": true,
|
|
1195
|
-
"
|
|
1196
|
-
"
|
|
1197
|
-
"last_updated": "2025-09-30",
|
|
985
|
+
"release_date": "2025-07-21",
|
|
986
|
+
"last_updated": "2025-07-21",
|
|
1198
987
|
"modalities": {
|
|
1199
988
|
"input": [
|
|
1200
989
|
"text"
|
|
@@ -1205,77 +994,100 @@
|
|
|
1205
994
|
},
|
|
1206
995
|
"open_weights": true,
|
|
1207
996
|
"limit": {
|
|
1208
|
-
"context":
|
|
1209
|
-
"output":
|
|
997
|
+
"context": 262144,
|
|
998
|
+
"output": 16384
|
|
1210
999
|
},
|
|
1211
1000
|
"cost": {
|
|
1212
|
-
"input": 0.
|
|
1213
|
-
"output":
|
|
1214
|
-
|
|
1215
|
-
|
|
1001
|
+
"input": 0.09,
|
|
1002
|
+
"output": 0.55
|
|
1003
|
+
},
|
|
1004
|
+
"canonical_model_id": "alibaba/qwen3-235b-a22b-instruct-2507"
|
|
1216
1005
|
},
|
|
1217
|
-
"
|
|
1218
|
-
"id": "
|
|
1219
|
-
"name": "
|
|
1220
|
-
"description": "
|
|
1221
|
-
"family": "
|
|
1006
|
+
"Qwen/Qwen3.8-Flash": {
|
|
1007
|
+
"id": "Qwen/Qwen3.8-Flash",
|
|
1008
|
+
"name": "Qwen3.8 Flash",
|
|
1009
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1010
|
+
"family": "qwen",
|
|
1222
1011
|
"attachment": true,
|
|
1223
|
-
"reasoning":
|
|
1224
|
-
"reasoning_options": [
|
|
1225
|
-
{
|
|
1226
|
-
"type": "effort",
|
|
1227
|
-
"values": [
|
|
1228
|
-
"low",
|
|
1229
|
-
"high",
|
|
1230
|
-
"max"
|
|
1231
|
-
]
|
|
1232
|
-
}
|
|
1233
|
-
],
|
|
1012
|
+
"reasoning": false,
|
|
1234
1013
|
"tool_call": true,
|
|
1235
1014
|
"structured_output": true,
|
|
1236
|
-
"temperature": true,
|
|
1237
1015
|
"release_date": "2026-08-26",
|
|
1238
1016
|
"last_updated": "2026-08-26",
|
|
1239
1017
|
"modalities": {
|
|
1240
1018
|
"input": [
|
|
1241
1019
|
"text",
|
|
1242
1020
|
"image",
|
|
1243
|
-
"video"
|
|
1244
|
-
"pdf"
|
|
1021
|
+
"video"
|
|
1245
1022
|
],
|
|
1246
1023
|
"output": [
|
|
1247
1024
|
"text"
|
|
1248
1025
|
]
|
|
1249
1026
|
},
|
|
1250
|
-
"open_weights":
|
|
1027
|
+
"open_weights": false,
|
|
1251
1028
|
"limit": {
|
|
1252
|
-
"context":
|
|
1029
|
+
"context": 1000000,
|
|
1253
1030
|
"output": 131072
|
|
1254
1031
|
},
|
|
1255
1032
|
"cost": {
|
|
1256
|
-
"input": 0.
|
|
1257
|
-
"output": 0.
|
|
1258
|
-
"cache_read": 0.
|
|
1259
|
-
}
|
|
1033
|
+
"input": 0.113,
|
|
1034
|
+
"output": 0.382,
|
|
1035
|
+
"cache_read": 0.0141
|
|
1036
|
+
},
|
|
1037
|
+
"canonical_model_id": "alibaba/qwen3.8-flash"
|
|
1260
1038
|
},
|
|
1261
|
-
"
|
|
1262
|
-
"id": "
|
|
1263
|
-
"name": "
|
|
1264
|
-
"description": "
|
|
1265
|
-
"family": "
|
|
1039
|
+
"Qwen/Qwen3-32B": {
|
|
1040
|
+
"id": "Qwen/Qwen3-32B",
|
|
1041
|
+
"name": "Qwen3 32B",
|
|
1042
|
+
"description": "Dense open Qwen model for self-hosted chat, reasoning, and coding",
|
|
1043
|
+
"family": "qwen",
|
|
1044
|
+
"attachment": false,
|
|
1045
|
+
"reasoning": true,
|
|
1046
|
+
"reasoning_options": [],
|
|
1047
|
+
"tool_call": true,
|
|
1048
|
+
"structured_output": true,
|
|
1049
|
+
"temperature": true,
|
|
1050
|
+
"knowledge": "2025-04",
|
|
1051
|
+
"release_date": "2025-04",
|
|
1052
|
+
"last_updated": "2025-04",
|
|
1053
|
+
"modalities": {
|
|
1054
|
+
"input": [
|
|
1055
|
+
"text"
|
|
1056
|
+
],
|
|
1057
|
+
"output": [
|
|
1058
|
+
"text"
|
|
1059
|
+
]
|
|
1060
|
+
},
|
|
1061
|
+
"open_weights": true,
|
|
1062
|
+
"limit": {
|
|
1063
|
+
"context": 40960,
|
|
1064
|
+
"output": 16384
|
|
1065
|
+
},
|
|
1066
|
+
"cost": {
|
|
1067
|
+
"input": 0.08,
|
|
1068
|
+
"output": 0.28
|
|
1069
|
+
},
|
|
1070
|
+
"canonical_model_id": "alibaba/qwen3-32b"
|
|
1071
|
+
},
|
|
1072
|
+
"Qwen/Qwen3.5-35B-A3B": {
|
|
1073
|
+
"id": "Qwen/Qwen3.5-35B-A3B",
|
|
1074
|
+
"name": "Qwen 3.5 35B A3B",
|
|
1075
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1076
|
+
"family": "qwen",
|
|
1266
1077
|
"attachment": true,
|
|
1267
1078
|
"reasoning": true,
|
|
1268
1079
|
"reasoning_options": [],
|
|
1269
1080
|
"tool_call": true,
|
|
1270
1081
|
"structured_output": true,
|
|
1271
1082
|
"temperature": true,
|
|
1272
|
-
"
|
|
1273
|
-
"
|
|
1083
|
+
"knowledge": "2025-01",
|
|
1084
|
+
"release_date": "2026-02-01",
|
|
1085
|
+
"last_updated": "2026-04-20",
|
|
1274
1086
|
"modalities": {
|
|
1275
1087
|
"input": [
|
|
1276
1088
|
"text",
|
|
1277
1089
|
"image",
|
|
1278
|
-
"
|
|
1090
|
+
"video"
|
|
1279
1091
|
],
|
|
1280
1092
|
"output": [
|
|
1281
1093
|
"text"
|
|
@@ -1283,31 +1095,33 @@
|
|
|
1283
1095
|
},
|
|
1284
1096
|
"open_weights": true,
|
|
1285
1097
|
"limit": {
|
|
1286
|
-
"context":
|
|
1287
|
-
"output":
|
|
1098
|
+
"context": 262144,
|
|
1099
|
+
"output": 81920
|
|
1288
1100
|
},
|
|
1289
1101
|
"cost": {
|
|
1290
|
-
"input": 0.
|
|
1291
|
-
"output": 1
|
|
1292
|
-
"cache_read": 0.
|
|
1102
|
+
"input": 0.14,
|
|
1103
|
+
"output": 1,
|
|
1104
|
+
"cache_read": 0.05
|
|
1293
1105
|
}
|
|
1294
1106
|
},
|
|
1295
|
-
"
|
|
1296
|
-
"id": "
|
|
1297
|
-
"name": "
|
|
1298
|
-
"description": "
|
|
1299
|
-
"family": "
|
|
1107
|
+
"Qwen/Qwen3.5-122B-A10B": {
|
|
1108
|
+
"id": "Qwen/Qwen3.5-122B-A10B",
|
|
1109
|
+
"name": "Qwen3.5 122B-A10B",
|
|
1110
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1111
|
+
"family": "qwen",
|
|
1300
1112
|
"attachment": true,
|
|
1301
1113
|
"reasoning": true,
|
|
1302
1114
|
"reasoning_options": [],
|
|
1303
1115
|
"tool_call": true,
|
|
1116
|
+
"structured_output": true,
|
|
1304
1117
|
"temperature": true,
|
|
1305
|
-
"release_date": "2026-
|
|
1306
|
-
"last_updated": "2026-
|
|
1118
|
+
"release_date": "2026-02-23",
|
|
1119
|
+
"last_updated": "2026-02-23",
|
|
1307
1120
|
"modalities": {
|
|
1308
1121
|
"input": [
|
|
1309
1122
|
"text",
|
|
1310
1123
|
"image",
|
|
1124
|
+
"video",
|
|
1311
1125
|
"audio"
|
|
1312
1126
|
],
|
|
1313
1127
|
"output": [
|
|
@@ -1316,115 +1130,52 @@
|
|
|
1316
1130
|
},
|
|
1317
1131
|
"open_weights": true,
|
|
1318
1132
|
"limit": {
|
|
1319
|
-
"context":
|
|
1320
|
-
"output":
|
|
1133
|
+
"context": 262144,
|
|
1134
|
+
"output": 65536
|
|
1321
1135
|
},
|
|
1322
1136
|
"cost": {
|
|
1323
|
-
"input": 0.
|
|
1324
|
-
"output": 4
|
|
1325
|
-
|
|
1326
|
-
|
|
1137
|
+
"input": 0.29,
|
|
1138
|
+
"output": 2.4
|
|
1139
|
+
},
|
|
1140
|
+
"canonical_model_id": "alibaba/qwen3.5-122b-a10b"
|
|
1327
1141
|
},
|
|
1328
|
-
"Qwen/Qwen3.
|
|
1329
|
-
"id": "Qwen/Qwen3.
|
|
1330
|
-
"name": "Qwen3.
|
|
1331
|
-
"description": "Qwen
|
|
1142
|
+
"Qwen/Qwen3.5-9B": {
|
|
1143
|
+
"id": "Qwen/Qwen3.5-9B",
|
|
1144
|
+
"name": "Qwen3.5 9B",
|
|
1145
|
+
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1332
1146
|
"family": "qwen",
|
|
1333
|
-
"attachment":
|
|
1334
|
-
"reasoning":
|
|
1147
|
+
"attachment": true,
|
|
1148
|
+
"reasoning": true,
|
|
1149
|
+
"reasoning_options": [],
|
|
1335
1150
|
"tool_call": true,
|
|
1336
1151
|
"structured_output": true,
|
|
1337
1152
|
"temperature": true,
|
|
1338
|
-
"release_date": "2026-
|
|
1339
|
-
"last_updated": "2026-
|
|
1153
|
+
"release_date": "2026-02-23",
|
|
1154
|
+
"last_updated": "2026-02-23",
|
|
1340
1155
|
"modalities": {
|
|
1341
1156
|
"input": [
|
|
1342
|
-
"text"
|
|
1157
|
+
"text",
|
|
1158
|
+
"image",
|
|
1159
|
+
"video"
|
|
1343
1160
|
],
|
|
1344
1161
|
"output": [
|
|
1345
1162
|
"text"
|
|
1346
1163
|
]
|
|
1347
1164
|
},
|
|
1348
|
-
"open_weights":
|
|
1165
|
+
"open_weights": true,
|
|
1349
1166
|
"limit": {
|
|
1350
|
-
"context":
|
|
1167
|
+
"context": 262144,
|
|
1351
1168
|
"output": 65536
|
|
1352
1169
|
},
|
|
1353
1170
|
"cost": {
|
|
1354
|
-
"input":
|
|
1355
|
-
"output":
|
|
1356
|
-
"cache_read": 0.5,
|
|
1357
|
-
"tiers": [
|
|
1358
|
-
{
|
|
1359
|
-
"input": 5,
|
|
1360
|
-
"output": 15,
|
|
1361
|
-
"cache_read": 1,
|
|
1362
|
-
"tier": {
|
|
1363
|
-
"type": "context",
|
|
1364
|
-
"size": 32000
|
|
1365
|
-
}
|
|
1366
|
-
},
|
|
1367
|
-
{
|
|
1368
|
-
"input": 6.25,
|
|
1369
|
-
"output": 18.5,
|
|
1370
|
-
"cache_read": 1.25,
|
|
1371
|
-
"tier": {
|
|
1372
|
-
"type": "context",
|
|
1373
|
-
"size": 128000
|
|
1374
|
-
}
|
|
1375
|
-
}
|
|
1376
|
-
]
|
|
1377
|
-
}
|
|
1378
|
-
},
|
|
1379
|
-
"Qwen/Qwen3.8-27B": {
|
|
1380
|
-
"id": "Qwen/Qwen3.8-27B",
|
|
1381
|
-
"name": "Qwen3.8 27B",
|
|
1382
|
-
"description": "Dense 27B vision-language model for coding, agent tasks, and image and video understanding",
|
|
1383
|
-
"family": "qwen",
|
|
1384
|
-
"attachment": true,
|
|
1385
|
-
"reasoning": true,
|
|
1386
|
-
"reasoning_options": [
|
|
1387
|
-
{
|
|
1388
|
-
"type": "toggle"
|
|
1389
|
-
},
|
|
1390
|
-
{
|
|
1391
|
-
"type": "effort",
|
|
1392
|
-
"values": [
|
|
1393
|
-
"low",
|
|
1394
|
-
"medium",
|
|
1395
|
-
"xhigh"
|
|
1396
|
-
]
|
|
1397
|
-
}
|
|
1398
|
-
],
|
|
1399
|
-
"tool_call": true,
|
|
1400
|
-
"structured_output": true,
|
|
1401
|
-
"temperature": true,
|
|
1402
|
-
"release_date": "2026-08-14",
|
|
1403
|
-
"last_updated": "2026-08-14",
|
|
1404
|
-
"modalities": {
|
|
1405
|
-
"input": [
|
|
1406
|
-
"text",
|
|
1407
|
-
"image",
|
|
1408
|
-
"video"
|
|
1409
|
-
],
|
|
1410
|
-
"output": [
|
|
1411
|
-
"text"
|
|
1412
|
-
]
|
|
1413
|
-
},
|
|
1414
|
-
"open_weights": true,
|
|
1415
|
-
"limit": {
|
|
1416
|
-
"context": 262144,
|
|
1417
|
-
"output": 32768
|
|
1171
|
+
"input": 0.1,
|
|
1172
|
+
"output": 0.15
|
|
1418
1173
|
},
|
|
1419
|
-
"
|
|
1420
|
-
"input": 0.2,
|
|
1421
|
-
"output": 2.5,
|
|
1422
|
-
"cache_read": 0.05
|
|
1423
|
-
}
|
|
1174
|
+
"canonical_model_id": "alibaba/qwen3.5-9b"
|
|
1424
1175
|
},
|
|
1425
|
-
"Qwen/Qwen3.5-
|
|
1426
|
-
"id": "Qwen/Qwen3.5-
|
|
1427
|
-
"name": "
|
|
1176
|
+
"Qwen/Qwen3.5-397B-A17B": {
|
|
1177
|
+
"id": "Qwen/Qwen3.5-397B-A17B",
|
|
1178
|
+
"name": "Qwen 3.5 397B A17B",
|
|
1428
1179
|
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1429
1180
|
"family": "qwen",
|
|
1430
1181
|
"attachment": true,
|
|
@@ -1433,14 +1184,14 @@
|
|
|
1433
1184
|
"tool_call": true,
|
|
1434
1185
|
"structured_output": true,
|
|
1435
1186
|
"temperature": true,
|
|
1436
|
-
"
|
|
1437
|
-
"
|
|
1187
|
+
"knowledge": "2025-01",
|
|
1188
|
+
"release_date": "2026-02-01",
|
|
1189
|
+
"last_updated": "2026-04-20",
|
|
1438
1190
|
"modalities": {
|
|
1439
1191
|
"input": [
|
|
1440
1192
|
"text",
|
|
1441
1193
|
"image",
|
|
1442
|
-
"video"
|
|
1443
|
-
"audio"
|
|
1194
|
+
"video"
|
|
1444
1195
|
],
|
|
1445
1196
|
"output": [
|
|
1446
1197
|
"text"
|
|
@@ -1449,26 +1200,36 @@
|
|
|
1449
1200
|
"open_weights": true,
|
|
1450
1201
|
"limit": {
|
|
1451
1202
|
"context": 262144,
|
|
1452
|
-
"output":
|
|
1203
|
+
"output": 81920
|
|
1453
1204
|
},
|
|
1454
1205
|
"cost": {
|
|
1455
|
-
"input": 0.
|
|
1456
|
-
"output":
|
|
1206
|
+
"input": 0.45,
|
|
1207
|
+
"output": 3,
|
|
1208
|
+
"cache_read": 0.22
|
|
1457
1209
|
}
|
|
1458
1210
|
},
|
|
1459
|
-
"Qwen/Qwen3-
|
|
1460
|
-
"id": "Qwen/Qwen3-
|
|
1461
|
-
"name": "Qwen3
|
|
1462
|
-
"description": "
|
|
1211
|
+
"Qwen/Qwen3.8-2.4T-A95B": {
|
|
1212
|
+
"id": "Qwen/Qwen3.8-2.4T-A95B",
|
|
1213
|
+
"name": "Qwen3.8 2.4T A95B",
|
|
1214
|
+
"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",
|
|
1463
1215
|
"family": "qwen",
|
|
1464
1216
|
"attachment": false,
|
|
1465
|
-
"reasoning":
|
|
1217
|
+
"reasoning": true,
|
|
1218
|
+
"reasoning_options": [
|
|
1219
|
+
{
|
|
1220
|
+
"type": "effort",
|
|
1221
|
+
"values": [
|
|
1222
|
+
"low",
|
|
1223
|
+
"medium",
|
|
1224
|
+
"xhigh"
|
|
1225
|
+
]
|
|
1226
|
+
}
|
|
1227
|
+
],
|
|
1466
1228
|
"tool_call": true,
|
|
1467
1229
|
"structured_output": true,
|
|
1468
1230
|
"temperature": true,
|
|
1469
|
-
"
|
|
1470
|
-
"
|
|
1471
|
-
"last_updated": "2025-07-23",
|
|
1231
|
+
"release_date": "2026-08-12",
|
|
1232
|
+
"last_updated": "2026-08-12",
|
|
1472
1233
|
"modalities": {
|
|
1473
1234
|
"input": [
|
|
1474
1235
|
"text"
|
|
@@ -1480,80 +1241,14 @@
|
|
|
1480
1241
|
"open_weights": true,
|
|
1481
1242
|
"limit": {
|
|
1482
1243
|
"context": 262144,
|
|
1483
|
-
"output": 66536
|
|
1484
|
-
},
|
|
1485
|
-
"cost": {
|
|
1486
|
-
"input": 0.3,
|
|
1487
|
-
"output": 1,
|
|
1488
|
-
"cache_read": 0.1
|
|
1489
|
-
}
|
|
1490
|
-
},
|
|
1491
|
-
"Qwen/Qwen3.8-Max": {
|
|
1492
|
-
"id": "Qwen/Qwen3.8-Max",
|
|
1493
|
-
"name": "Qwen3.8 Max",
|
|
1494
|
-
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
|
|
1495
|
-
"family": "qwen",
|
|
1496
|
-
"attachment": true,
|
|
1497
|
-
"reasoning": false,
|
|
1498
|
-
"tool_call": true,
|
|
1499
|
-
"structured_output": true,
|
|
1500
|
-
"temperature": true,
|
|
1501
|
-
"release_date": "2026-08-03",
|
|
1502
|
-
"last_updated": "2026-08-03",
|
|
1503
|
-
"modalities": {
|
|
1504
|
-
"input": [
|
|
1505
|
-
"text",
|
|
1506
|
-
"image",
|
|
1507
|
-
"video",
|
|
1508
|
-
"pdf"
|
|
1509
|
-
],
|
|
1510
|
-
"output": [
|
|
1511
|
-
"text"
|
|
1512
|
-
]
|
|
1513
|
-
},
|
|
1514
|
-
"open_weights": false,
|
|
1515
|
-
"limit": {
|
|
1516
|
-
"context": 256000,
|
|
1517
1244
|
"output": 131072
|
|
1518
1245
|
},
|
|
1519
1246
|
"cost": {
|
|
1520
|
-
"input":
|
|
1521
|
-
"output":
|
|
1522
|
-
"cache_read": 0.
|
|
1523
|
-
}
|
|
1524
|
-
},
|
|
1525
|
-
"Qwen/Qwen3.5-9B": {
|
|
1526
|
-
"id": "Qwen/Qwen3.5-9B",
|
|
1527
|
-
"name": "Qwen3.5 9B",
|
|
1528
|
-
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1529
|
-
"family": "qwen",
|
|
1530
|
-
"attachment": true,
|
|
1531
|
-
"reasoning": true,
|
|
1532
|
-
"reasoning_options": [],
|
|
1533
|
-
"tool_call": true,
|
|
1534
|
-
"structured_output": true,
|
|
1535
|
-
"temperature": true,
|
|
1536
|
-
"release_date": "2026-02-23",
|
|
1537
|
-
"last_updated": "2026-02-23",
|
|
1538
|
-
"modalities": {
|
|
1539
|
-
"input": [
|
|
1540
|
-
"text",
|
|
1541
|
-
"image",
|
|
1542
|
-
"video"
|
|
1543
|
-
],
|
|
1544
|
-
"output": [
|
|
1545
|
-
"text"
|
|
1546
|
-
]
|
|
1547
|
-
},
|
|
1548
|
-
"open_weights": true,
|
|
1549
|
-
"limit": {
|
|
1550
|
-
"context": 262144,
|
|
1551
|
-
"output": 65536
|
|
1247
|
+
"input": 2,
|
|
1248
|
+
"output": 6,
|
|
1249
|
+
"cache_read": 0.2
|
|
1552
1250
|
},
|
|
1553
|
-
"
|
|
1554
|
-
"input": 0.1,
|
|
1555
|
-
"output": 0.15
|
|
1556
|
-
}
|
|
1251
|
+
"canonical_model_id": "alibaba/qwen3.8-2.4t-a95b"
|
|
1557
1252
|
},
|
|
1558
1253
|
"Qwen/Qwen3-30B-A3B": {
|
|
1559
1254
|
"id": "Qwen/Qwen3-30B-A3B",
|
|
@@ -1584,11 +1279,12 @@
|
|
|
1584
1279
|
"cost": {
|
|
1585
1280
|
"input": 0.12,
|
|
1586
1281
|
"output": 0.5
|
|
1587
|
-
}
|
|
1282
|
+
},
|
|
1283
|
+
"canonical_model_id": "alibaba/qwen3-30b-a3b"
|
|
1588
1284
|
},
|
|
1589
|
-
"Qwen/Qwen3.
|
|
1590
|
-
"id": "Qwen/Qwen3.
|
|
1591
|
-
"name": "Qwen3.
|
|
1285
|
+
"Qwen/Qwen3.6-27B": {
|
|
1286
|
+
"id": "Qwen/Qwen3.6-27B",
|
|
1287
|
+
"name": "Qwen3.6 27B",
|
|
1592
1288
|
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1593
1289
|
"family": "qwen",
|
|
1594
1290
|
"attachment": true,
|
|
@@ -1597,8 +1293,8 @@
|
|
|
1597
1293
|
"tool_call": true,
|
|
1598
1294
|
"structured_output": true,
|
|
1599
1295
|
"temperature": true,
|
|
1600
|
-
"release_date": "2026-
|
|
1601
|
-
"last_updated": "2026-
|
|
1296
|
+
"release_date": "2026-04-22",
|
|
1297
|
+
"last_updated": "2026-04-22",
|
|
1602
1298
|
"modalities": {
|
|
1603
1299
|
"input": [
|
|
1604
1300
|
"text",
|
|
@@ -1616,32 +1312,28 @@
|
|
|
1616
1312
|
"output": 65536
|
|
1617
1313
|
},
|
|
1618
1314
|
"cost": {
|
|
1619
|
-
"input": 0.
|
|
1620
|
-
"output": 2
|
|
1621
|
-
}
|
|
1315
|
+
"input": 0.32,
|
|
1316
|
+
"output": 3.2
|
|
1317
|
+
},
|
|
1318
|
+
"canonical_model_id": "alibaba/qwen3.6-27b"
|
|
1622
1319
|
},
|
|
1623
|
-
"
|
|
1624
|
-
"id": "
|
|
1625
|
-
"name": "
|
|
1626
|
-
"description": "
|
|
1627
|
-
"family": "
|
|
1320
|
+
"deepseek-ai/DeepSeek-V3.1": {
|
|
1321
|
+
"id": "deepseek-ai/DeepSeek-V3.1",
|
|
1322
|
+
"name": "DeepSeek-V3.1",
|
|
1323
|
+
"description": "Hybrid-reasoning DeepSeek model with thinking and non-thinking modes",
|
|
1324
|
+
"family": "deepseek",
|
|
1628
1325
|
"attachment": false,
|
|
1629
1326
|
"reasoning": true,
|
|
1630
1327
|
"reasoning_options": [
|
|
1631
1328
|
{
|
|
1632
|
-
"type": "
|
|
1633
|
-
"values": [
|
|
1634
|
-
"low",
|
|
1635
|
-
"medium",
|
|
1636
|
-
"xhigh"
|
|
1637
|
-
]
|
|
1329
|
+
"type": "toggle"
|
|
1638
1330
|
}
|
|
1639
1331
|
],
|
|
1640
1332
|
"tool_call": true,
|
|
1641
1333
|
"structured_output": true,
|
|
1642
1334
|
"temperature": true,
|
|
1643
|
-
"release_date": "
|
|
1644
|
-
"last_updated": "
|
|
1335
|
+
"release_date": "2025-08-21",
|
|
1336
|
+
"last_updated": "2025-08-21",
|
|
1645
1337
|
"modalities": {
|
|
1646
1338
|
"input": [
|
|
1647
1339
|
"text"
|
|
@@ -1652,27 +1344,34 @@
|
|
|
1652
1344
|
},
|
|
1653
1345
|
"open_weights": true,
|
|
1654
1346
|
"limit": {
|
|
1655
|
-
"context":
|
|
1656
|
-
"output":
|
|
1347
|
+
"context": 163840,
|
|
1348
|
+
"output": 8192
|
|
1657
1349
|
},
|
|
1658
1350
|
"cost": {
|
|
1659
|
-
"input":
|
|
1660
|
-
"output":
|
|
1661
|
-
"cache_read": 0.
|
|
1662
|
-
}
|
|
1351
|
+
"input": 0.25,
|
|
1352
|
+
"output": 0.95,
|
|
1353
|
+
"cache_read": 0.13
|
|
1354
|
+
},
|
|
1355
|
+
"canonical_model_id": "deepseek/deepseek-v3.1"
|
|
1663
1356
|
},
|
|
1664
|
-
"
|
|
1665
|
-
"id": "
|
|
1666
|
-
"name": "
|
|
1667
|
-
"description": "
|
|
1668
|
-
"family": "
|
|
1357
|
+
"deepseek-ai/DeepSeek-V4-Flash-0731": {
|
|
1358
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash-0731",
|
|
1359
|
+
"name": "DeepSeek V4 Flash 0731",
|
|
1360
|
+
"description": "Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding",
|
|
1361
|
+
"family": "deepseek-flash",
|
|
1669
1362
|
"attachment": false,
|
|
1670
|
-
"reasoning":
|
|
1671
|
-
"
|
|
1672
|
-
|
|
1673
|
-
|
|
1674
|
-
|
|
1675
|
-
|
|
1363
|
+
"reasoning": true,
|
|
1364
|
+
"reasoning_options": [
|
|
1365
|
+
{
|
|
1366
|
+
"type": "toggle"
|
|
1367
|
+
}
|
|
1368
|
+
],
|
|
1369
|
+
"tool_call": true,
|
|
1370
|
+
"structured_output": true,
|
|
1371
|
+
"temperature": true,
|
|
1372
|
+
"knowledge": "2025-05",
|
|
1373
|
+
"release_date": "2026-07-31",
|
|
1374
|
+
"last_updated": "2026-07-31",
|
|
1676
1375
|
"modalities": {
|
|
1677
1376
|
"input": [
|
|
1678
1377
|
"text"
|
|
@@ -1683,59 +1382,85 @@
|
|
|
1683
1382
|
},
|
|
1684
1383
|
"open_weights": true,
|
|
1685
1384
|
"limit": {
|
|
1686
|
-
"context":
|
|
1687
|
-
"output":
|
|
1385
|
+
"context": 1048576,
|
|
1386
|
+
"output": 384000
|
|
1688
1387
|
},
|
|
1689
1388
|
"cost": {
|
|
1690
|
-
"input": 0.
|
|
1691
|
-
"output": 0.
|
|
1692
|
-
|
|
1389
|
+
"input": 0.06,
|
|
1390
|
+
"output": 0.18,
|
|
1391
|
+
"cache_read": 0.015
|
|
1392
|
+
},
|
|
1393
|
+
"canonical_model_id": "deepseek/deepseek-v4-flash-0731"
|
|
1693
1394
|
},
|
|
1694
|
-
"
|
|
1695
|
-
"id": "
|
|
1696
|
-
"name": "
|
|
1697
|
-
"description": "
|
|
1698
|
-
"family": "
|
|
1699
|
-
"attachment":
|
|
1700
|
-
"reasoning":
|
|
1395
|
+
"deepseek-ai/DeepSeek-V4-Pro-0813": {
|
|
1396
|
+
"id": "deepseek-ai/DeepSeek-V4-Pro-0813",
|
|
1397
|
+
"name": "DeepSeek V4 Pro 0813",
|
|
1398
|
+
"description": "DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes",
|
|
1399
|
+
"family": "deepseek-thinking",
|
|
1400
|
+
"attachment": false,
|
|
1401
|
+
"reasoning": true,
|
|
1402
|
+
"reasoning_options": [
|
|
1403
|
+
{
|
|
1404
|
+
"type": "toggle"
|
|
1405
|
+
},
|
|
1406
|
+
{
|
|
1407
|
+
"type": "effort",
|
|
1408
|
+
"values": [
|
|
1409
|
+
"high",
|
|
1410
|
+
"max"
|
|
1411
|
+
]
|
|
1412
|
+
}
|
|
1413
|
+
],
|
|
1701
1414
|
"tool_call": true,
|
|
1702
1415
|
"structured_output": true,
|
|
1703
|
-
"
|
|
1704
|
-
"
|
|
1416
|
+
"temperature": true,
|
|
1417
|
+
"release_date": "2026-08-12",
|
|
1418
|
+
"last_updated": "2026-08-22",
|
|
1705
1419
|
"modalities": {
|
|
1706
1420
|
"input": [
|
|
1707
|
-
"text"
|
|
1708
|
-
"image",
|
|
1709
|
-
"video"
|
|
1421
|
+
"text"
|
|
1710
1422
|
],
|
|
1711
1423
|
"output": [
|
|
1712
1424
|
"text"
|
|
1713
1425
|
]
|
|
1714
1426
|
},
|
|
1715
|
-
"open_weights":
|
|
1427
|
+
"open_weights": true,
|
|
1716
1428
|
"limit": {
|
|
1717
|
-
"context":
|
|
1718
|
-
"output":
|
|
1429
|
+
"context": 1048576,
|
|
1430
|
+
"output": 384000
|
|
1719
1431
|
},
|
|
1720
1432
|
"cost": {
|
|
1721
|
-
"input":
|
|
1722
|
-
"output":
|
|
1723
|
-
"cache_read": 0.
|
|
1724
|
-
}
|
|
1433
|
+
"input": 1.3,
|
|
1434
|
+
"output": 2.6,
|
|
1435
|
+
"cache_read": 0.1
|
|
1436
|
+
},
|
|
1437
|
+
"canonical_model_id": "deepseek/deepseek-v4-pro-0813"
|
|
1725
1438
|
},
|
|
1726
|
-
"
|
|
1727
|
-
"id": "
|
|
1728
|
-
"name": "
|
|
1729
|
-
"description": "
|
|
1730
|
-
"family": "
|
|
1439
|
+
"deepseek-ai/DeepSeek-V4.1-Flash": {
|
|
1440
|
+
"id": "deepseek-ai/DeepSeek-V4.1-Flash",
|
|
1441
|
+
"name": "DeepSeek V4.1 Flash",
|
|
1442
|
+
"description": "DeepSeek V4.1 Flash model for reasoning and agentic coding",
|
|
1443
|
+
"family": "deepseek-flash",
|
|
1731
1444
|
"attachment": true,
|
|
1732
|
-
"reasoning":
|
|
1445
|
+
"reasoning": true,
|
|
1446
|
+
"reasoning_options": [
|
|
1447
|
+
{
|
|
1448
|
+
"type": "effort",
|
|
1449
|
+
"values": [
|
|
1450
|
+
"none",
|
|
1451
|
+
"low",
|
|
1452
|
+
"high",
|
|
1453
|
+
"xhigh",
|
|
1454
|
+
"max"
|
|
1455
|
+
]
|
|
1456
|
+
}
|
|
1457
|
+
],
|
|
1733
1458
|
"tool_call": true,
|
|
1734
1459
|
"structured_output": true,
|
|
1735
1460
|
"temperature": true,
|
|
1736
|
-
"knowledge": "2025-
|
|
1737
|
-
"release_date": "
|
|
1738
|
-
"last_updated": "
|
|
1461
|
+
"knowledge": "2025-05",
|
|
1462
|
+
"release_date": "2026-09-10",
|
|
1463
|
+
"last_updated": "2026-09-10",
|
|
1739
1464
|
"modalities": {
|
|
1740
1465
|
"input": [
|
|
1741
1466
|
"text",
|
|
@@ -1747,28 +1472,46 @@
|
|
|
1747
1472
|
},
|
|
1748
1473
|
"open_weights": true,
|
|
1749
1474
|
"limit": {
|
|
1750
|
-
"context":
|
|
1751
|
-
"output":
|
|
1475
|
+
"context": 1048576,
|
|
1476
|
+
"output": 384000
|
|
1752
1477
|
},
|
|
1753
1478
|
"cost": {
|
|
1754
1479
|
"input": 0.2,
|
|
1755
|
-
"output": 0.
|
|
1756
|
-
"cache_read": 0.
|
|
1757
|
-
}
|
|
1480
|
+
"output": 0.6,
|
|
1481
|
+
"cache_read": 0.006
|
|
1482
|
+
},
|
|
1483
|
+
"canonical_model_id": "deepseek/deepseek-v4.1-flash"
|
|
1758
1484
|
},
|
|
1759
|
-
"
|
|
1760
|
-
"id": "
|
|
1761
|
-
"name": "
|
|
1762
|
-
"description": "
|
|
1763
|
-
"family": "
|
|
1485
|
+
"deepseek-ai/DeepSeek-V4-Flash": {
|
|
1486
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash",
|
|
1487
|
+
"name": "DeepSeek V4 Flash",
|
|
1488
|
+
"description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work",
|
|
1489
|
+
"family": "deepseek-flash",
|
|
1764
1490
|
"attachment": false,
|
|
1765
|
-
"reasoning":
|
|
1491
|
+
"reasoning": true,
|
|
1492
|
+
"reasoning_options": [
|
|
1493
|
+
{
|
|
1494
|
+
"type": "toggle"
|
|
1495
|
+
},
|
|
1496
|
+
{
|
|
1497
|
+
"type": "effort",
|
|
1498
|
+
"values": [
|
|
1499
|
+
"low",
|
|
1500
|
+
"medium",
|
|
1501
|
+
"high",
|
|
1502
|
+
"xhigh"
|
|
1503
|
+
]
|
|
1504
|
+
}
|
|
1505
|
+
],
|
|
1766
1506
|
"tool_call": true,
|
|
1507
|
+
"interleaved": {
|
|
1508
|
+
"field": "reasoning_content"
|
|
1509
|
+
},
|
|
1767
1510
|
"structured_output": true,
|
|
1768
1511
|
"temperature": true,
|
|
1769
|
-
"knowledge": "2025-
|
|
1770
|
-
"release_date": "
|
|
1771
|
-
"last_updated": "
|
|
1512
|
+
"knowledge": "2025-05",
|
|
1513
|
+
"release_date": "2026-04-24",
|
|
1514
|
+
"last_updated": "2026-04-24",
|
|
1772
1515
|
"modalities": {
|
|
1773
1516
|
"input": [
|
|
1774
1517
|
"text"
|
|
@@ -1779,33 +1522,49 @@
|
|
|
1779
1522
|
},
|
|
1780
1523
|
"open_weights": true,
|
|
1781
1524
|
"limit": {
|
|
1782
|
-
"context":
|
|
1783
|
-
"output":
|
|
1525
|
+
"context": 1048576,
|
|
1526
|
+
"output": 16384
|
|
1784
1527
|
},
|
|
1785
1528
|
"cost": {
|
|
1786
1529
|
"input": 0.09,
|
|
1787
|
-
"output":
|
|
1788
|
-
|
|
1530
|
+
"output": 0.18,
|
|
1531
|
+
"cache_read": 0.018
|
|
1532
|
+
},
|
|
1533
|
+
"canonical_model_id": "deepseek/deepseek-v4-flash"
|
|
1789
1534
|
},
|
|
1790
|
-
"
|
|
1791
|
-
"id": "
|
|
1792
|
-
"name": "
|
|
1793
|
-
"description": "
|
|
1794
|
-
"family": "
|
|
1795
|
-
"attachment":
|
|
1535
|
+
"deepseek-ai/DeepSeek-V4-Pro": {
|
|
1536
|
+
"id": "deepseek-ai/DeepSeek-V4-Pro",
|
|
1537
|
+
"name": "DeepSeek V4 Pro",
|
|
1538
|
+
"description": "Open MoE flagship with million-token context for coding and long agent runs",
|
|
1539
|
+
"family": "deepseek-thinking",
|
|
1540
|
+
"attachment": false,
|
|
1796
1541
|
"reasoning": true,
|
|
1797
|
-
"reasoning_options": [
|
|
1542
|
+
"reasoning_options": [
|
|
1543
|
+
{
|
|
1544
|
+
"type": "toggle"
|
|
1545
|
+
},
|
|
1546
|
+
{
|
|
1547
|
+
"type": "effort",
|
|
1548
|
+
"values": [
|
|
1549
|
+
"low",
|
|
1550
|
+
"medium",
|
|
1551
|
+
"high",
|
|
1552
|
+
"xhigh"
|
|
1553
|
+
]
|
|
1554
|
+
}
|
|
1555
|
+
],
|
|
1798
1556
|
"tool_call": true,
|
|
1557
|
+
"interleaved": {
|
|
1558
|
+
"field": "reasoning_content"
|
|
1559
|
+
},
|
|
1799
1560
|
"structured_output": true,
|
|
1800
1561
|
"temperature": true,
|
|
1801
|
-
"knowledge": "2025-
|
|
1802
|
-
"release_date": "2026-
|
|
1803
|
-
"last_updated": "2026-04-
|
|
1562
|
+
"knowledge": "2025-05",
|
|
1563
|
+
"release_date": "2026-04-24",
|
|
1564
|
+
"last_updated": "2026-04-24",
|
|
1804
1565
|
"modalities": {
|
|
1805
1566
|
"input": [
|
|
1806
|
-
"text"
|
|
1807
|
-
"image",
|
|
1808
|
-
"video"
|
|
1567
|
+
"text"
|
|
1809
1568
|
],
|
|
1810
1569
|
"output": [
|
|
1811
1570
|
"text"
|
|
@@ -1813,64 +1572,71 @@
|
|
|
1813
1572
|
},
|
|
1814
1573
|
"open_weights": true,
|
|
1815
1574
|
"limit": {
|
|
1816
|
-
"context":
|
|
1817
|
-
"output":
|
|
1575
|
+
"context": 1048576,
|
|
1576
|
+
"output": 16384
|
|
1818
1577
|
},
|
|
1819
1578
|
"cost": {
|
|
1820
|
-
"input":
|
|
1821
|
-
"output":
|
|
1822
|
-
"cache_read": 0.
|
|
1823
|
-
}
|
|
1579
|
+
"input": 1.3,
|
|
1580
|
+
"output": 2.6,
|
|
1581
|
+
"cache_read": 0.1
|
|
1582
|
+
},
|
|
1583
|
+
"canonical_model_id": "deepseek/deepseek-v4-pro"
|
|
1824
1584
|
},
|
|
1825
|
-
"
|
|
1826
|
-
"id": "
|
|
1827
|
-
"name": "
|
|
1828
|
-
"description": "
|
|
1829
|
-
"
|
|
1830
|
-
"attachment": true,
|
|
1585
|
+
"deepseek-ai/DeepSeek-V3.2": {
|
|
1586
|
+
"id": "deepseek-ai/DeepSeek-V3.2",
|
|
1587
|
+
"name": "DeepSeek-V3.2",
|
|
1588
|
+
"description": "DeepSeek chat model for instruction following, coding, and analysis",
|
|
1589
|
+
"attachment": false,
|
|
1831
1590
|
"reasoning": true,
|
|
1832
|
-
"reasoning_options": [
|
|
1591
|
+
"reasoning_options": [
|
|
1592
|
+
{
|
|
1593
|
+
"type": "toggle"
|
|
1594
|
+
}
|
|
1595
|
+
],
|
|
1833
1596
|
"tool_call": true,
|
|
1597
|
+
"interleaved": {
|
|
1598
|
+
"field": "reasoning_content"
|
|
1599
|
+
},
|
|
1834
1600
|
"structured_output": true,
|
|
1835
1601
|
"temperature": true,
|
|
1836
|
-
"knowledge": "
|
|
1837
|
-
"release_date": "
|
|
1838
|
-
"last_updated": "
|
|
1602
|
+
"knowledge": "2024-12",
|
|
1603
|
+
"release_date": "2025-12-02",
|
|
1604
|
+
"last_updated": "2025-12-02",
|
|
1839
1605
|
"modalities": {
|
|
1840
1606
|
"input": [
|
|
1841
|
-
"text"
|
|
1842
|
-
"image",
|
|
1843
|
-
"video"
|
|
1607
|
+
"text"
|
|
1844
1608
|
],
|
|
1845
1609
|
"output": [
|
|
1846
1610
|
"text"
|
|
1847
1611
|
]
|
|
1848
1612
|
},
|
|
1849
|
-
"open_weights":
|
|
1613
|
+
"open_weights": false,
|
|
1850
1614
|
"limit": {
|
|
1851
|
-
"context":
|
|
1852
|
-
"output":
|
|
1615
|
+
"context": 163840,
|
|
1616
|
+
"output": 64000
|
|
1853
1617
|
},
|
|
1854
1618
|
"cost": {
|
|
1855
|
-
"input": 0.
|
|
1856
|
-
"output":
|
|
1857
|
-
"cache_read": 0.
|
|
1619
|
+
"input": 0.26,
|
|
1620
|
+
"output": 0.38,
|
|
1621
|
+
"cache_read": 0.13
|
|
1858
1622
|
}
|
|
1859
1623
|
},
|
|
1860
|
-
"
|
|
1861
|
-
"id": "
|
|
1862
|
-
"name": "
|
|
1863
|
-
"description": "
|
|
1864
|
-
"
|
|
1865
|
-
"attachment": false,
|
|
1624
|
+
"deepseek-ai/DeepSeek-R1-0528": {
|
|
1625
|
+
"id": "deepseek-ai/DeepSeek-R1-0528",
|
|
1626
|
+
"name": "DeepSeek-R1-0528",
|
|
1627
|
+
"description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools",
|
|
1628
|
+
"attachment": false,
|
|
1866
1629
|
"reasoning": true,
|
|
1867
1630
|
"reasoning_options": [],
|
|
1868
1631
|
"tool_call": true,
|
|
1632
|
+
"interleaved": {
|
|
1633
|
+
"field": "reasoning_content"
|
|
1634
|
+
},
|
|
1869
1635
|
"structured_output": true,
|
|
1870
1636
|
"temperature": true,
|
|
1871
|
-
"knowledge": "
|
|
1872
|
-
"release_date": "2025-
|
|
1873
|
-
"last_updated": "2025-
|
|
1637
|
+
"knowledge": "2024-07",
|
|
1638
|
+
"release_date": "2025-05-28",
|
|
1639
|
+
"last_updated": "2025-05-28",
|
|
1874
1640
|
"modalities": {
|
|
1875
1641
|
"input": [
|
|
1876
1642
|
"text"
|
|
@@ -1879,35 +1645,32 @@
|
|
|
1879
1645
|
"text"
|
|
1880
1646
|
]
|
|
1881
1647
|
},
|
|
1882
|
-
"open_weights":
|
|
1648
|
+
"open_weights": false,
|
|
1883
1649
|
"limit": {
|
|
1884
|
-
"context":
|
|
1885
|
-
"output":
|
|
1650
|
+
"context": 163840,
|
|
1651
|
+
"output": 64000
|
|
1886
1652
|
},
|
|
1887
1653
|
"cost": {
|
|
1888
|
-
"input": 0.
|
|
1889
|
-
"output":
|
|
1654
|
+
"input": 0.5,
|
|
1655
|
+
"output": 2.15,
|
|
1656
|
+
"cache_read": 0.35
|
|
1890
1657
|
}
|
|
1891
1658
|
},
|
|
1892
|
-
"
|
|
1893
|
-
"id": "
|
|
1894
|
-
"name": "
|
|
1895
|
-
"description": "
|
|
1896
|
-
"family": "
|
|
1897
|
-
"attachment":
|
|
1898
|
-
"reasoning":
|
|
1899
|
-
"reasoning_options": [],
|
|
1659
|
+
"deepseek-ai/DeepSeek-V3": {
|
|
1660
|
+
"id": "deepseek-ai/DeepSeek-V3",
|
|
1661
|
+
"name": "DeepSeek-V3",
|
|
1662
|
+
"description": "Open DeepSeek MoE chat model for coding, math, and general reasoning",
|
|
1663
|
+
"family": "deepseek",
|
|
1664
|
+
"attachment": false,
|
|
1665
|
+
"reasoning": false,
|
|
1900
1666
|
"tool_call": true,
|
|
1901
1667
|
"structured_output": true,
|
|
1902
1668
|
"temperature": true,
|
|
1903
|
-
"release_date": "
|
|
1904
|
-
"last_updated": "
|
|
1669
|
+
"release_date": "2024-12-26",
|
|
1670
|
+
"last_updated": "2024-12-26",
|
|
1905
1671
|
"modalities": {
|
|
1906
1672
|
"input": [
|
|
1907
|
-
"text"
|
|
1908
|
-
"image",
|
|
1909
|
-
"video",
|
|
1910
|
-
"audio"
|
|
1673
|
+
"text"
|
|
1911
1674
|
],
|
|
1912
1675
|
"output": [
|
|
1913
1676
|
"text"
|
|
@@ -1915,32 +1678,30 @@
|
|
|
1915
1678
|
},
|
|
1916
1679
|
"open_weights": true,
|
|
1917
1680
|
"limit": {
|
|
1918
|
-
"context":
|
|
1919
|
-
"output":
|
|
1681
|
+
"context": 163840,
|
|
1682
|
+
"output": 8192
|
|
1920
1683
|
},
|
|
1921
1684
|
"cost": {
|
|
1922
1685
|
"input": 0.32,
|
|
1923
|
-
"output":
|
|
1924
|
-
}
|
|
1686
|
+
"output": 0.89
|
|
1687
|
+
},
|
|
1688
|
+
"canonical_model_id": "deepseek/deepseek-v3"
|
|
1925
1689
|
},
|
|
1926
|
-
"
|
|
1927
|
-
"id": "
|
|
1928
|
-
"name": "
|
|
1929
|
-
"description": "
|
|
1930
|
-
"family": "
|
|
1931
|
-
"attachment":
|
|
1932
|
-
"reasoning":
|
|
1933
|
-
"reasoning_options": [],
|
|
1690
|
+
"deepseek-ai/DeepSeek-V3-0324": {
|
|
1691
|
+
"id": "deepseek-ai/DeepSeek-V3-0324",
|
|
1692
|
+
"name": "DeepSeek V3 0324",
|
|
1693
|
+
"description": "March 2025 checkpoint of DeepSeek-V3 with improved reasoning and coding",
|
|
1694
|
+
"family": "deepseek",
|
|
1695
|
+
"attachment": false,
|
|
1696
|
+
"reasoning": false,
|
|
1934
1697
|
"tool_call": true,
|
|
1935
1698
|
"structured_output": true,
|
|
1936
1699
|
"temperature": true,
|
|
1937
|
-
"release_date": "
|
|
1938
|
-
"last_updated": "
|
|
1700
|
+
"release_date": "2025-03-24",
|
|
1701
|
+
"last_updated": "2025-03-24",
|
|
1939
1702
|
"modalities": {
|
|
1940
1703
|
"input": [
|
|
1941
|
-
"text"
|
|
1942
|
-
"image",
|
|
1943
|
-
"video"
|
|
1704
|
+
"text"
|
|
1944
1705
|
],
|
|
1945
1706
|
"output": [
|
|
1946
1707
|
"text"
|
|
@@ -1948,85 +1709,77 @@
|
|
|
1948
1709
|
},
|
|
1949
1710
|
"open_weights": true,
|
|
1950
1711
|
"limit": {
|
|
1951
|
-
"context":
|
|
1952
|
-
"output":
|
|
1712
|
+
"context": 163840,
|
|
1713
|
+
"output": 163840
|
|
1953
1714
|
},
|
|
1954
1715
|
"cost": {
|
|
1955
|
-
"input": 0.
|
|
1956
|
-
"output": 0.
|
|
1957
|
-
|
|
1716
|
+
"input": 0.24,
|
|
1717
|
+
"output": 0.9,
|
|
1718
|
+
"cache_read": 0.135
|
|
1719
|
+
},
|
|
1720
|
+
"canonical_model_id": "deepseek/deepseek-v3-0324"
|
|
1958
1721
|
},
|
|
1959
|
-
"
|
|
1960
|
-
"id": "
|
|
1961
|
-
"name": "
|
|
1962
|
-
"description": "
|
|
1963
|
-
"family": "
|
|
1964
|
-
"attachment":
|
|
1965
|
-
"reasoning":
|
|
1722
|
+
"deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {
|
|
1723
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
|
|
1724
|
+
"name": "DeepSeek V4 Flash Vision Exp",
|
|
1725
|
+
"description": "Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work",
|
|
1726
|
+
"family": "deepseek-flash",
|
|
1727
|
+
"attachment": true,
|
|
1728
|
+
"reasoning": true,
|
|
1729
|
+
"reasoning_options": [
|
|
1730
|
+
{
|
|
1731
|
+
"type": "effort",
|
|
1732
|
+
"values": [
|
|
1733
|
+
"none",
|
|
1734
|
+
"low",
|
|
1735
|
+
"high",
|
|
1736
|
+
"max"
|
|
1737
|
+
]
|
|
1738
|
+
}
|
|
1739
|
+
],
|
|
1966
1740
|
"tool_call": true,
|
|
1967
1741
|
"structured_output": true,
|
|
1968
1742
|
"temperature": true,
|
|
1969
|
-
"
|
|
1970
|
-
"
|
|
1971
|
-
"last_updated": "2025-09-23",
|
|
1743
|
+
"release_date": "2026-08-21",
|
|
1744
|
+
"last_updated": "2026-09-01",
|
|
1972
1745
|
"modalities": {
|
|
1973
1746
|
"input": [
|
|
1974
|
-
"text"
|
|
1747
|
+
"text",
|
|
1748
|
+
"image"
|
|
1975
1749
|
],
|
|
1976
1750
|
"output": [
|
|
1977
1751
|
"text"
|
|
1978
1752
|
]
|
|
1979
1753
|
},
|
|
1980
|
-
"open_weights":
|
|
1754
|
+
"open_weights": true,
|
|
1981
1755
|
"limit": {
|
|
1982
|
-
"context":
|
|
1983
|
-
"output":
|
|
1756
|
+
"context": 1048576,
|
|
1757
|
+
"output": 384000
|
|
1984
1758
|
},
|
|
1985
1759
|
"cost": {
|
|
1986
|
-
"input":
|
|
1987
|
-
"output":
|
|
1988
|
-
"cache_read": 0.
|
|
1989
|
-
|
|
1990
|
-
|
|
1991
|
-
"input": 2.4,
|
|
1992
|
-
"output": 12,
|
|
1993
|
-
"cache_read": 0.48,
|
|
1994
|
-
"tier": {
|
|
1995
|
-
"type": "context",
|
|
1996
|
-
"size": 32000
|
|
1997
|
-
}
|
|
1998
|
-
},
|
|
1999
|
-
{
|
|
2000
|
-
"input": 3,
|
|
2001
|
-
"output": 15,
|
|
2002
|
-
"cache_read": 0.6,
|
|
2003
|
-
"tier": {
|
|
2004
|
-
"type": "context",
|
|
2005
|
-
"size": 128000
|
|
2006
|
-
}
|
|
2007
|
-
}
|
|
2008
|
-
]
|
|
2009
|
-
}
|
|
1760
|
+
"input": 0.44,
|
|
1761
|
+
"output": 1.32,
|
|
1762
|
+
"cache_read": 0.014
|
|
1763
|
+
},
|
|
1764
|
+
"canonical_model_id": "deepseek/deepseek-v4-flash-vision-exp"
|
|
2010
1765
|
},
|
|
2011
|
-
"
|
|
2012
|
-
"id": "
|
|
2013
|
-
"name": "
|
|
2014
|
-
"description": "
|
|
2015
|
-
"
|
|
2016
|
-
"attachment": false,
|
|
1766
|
+
"stepfun-ai/Step-3.7-Flash": {
|
|
1767
|
+
"id": "stepfun-ai/Step-3.7-Flash",
|
|
1768
|
+
"name": "Step 3.7 Flash",
|
|
1769
|
+
"description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts",
|
|
1770
|
+
"attachment": true,
|
|
2017
1771
|
"reasoning": true,
|
|
2018
1772
|
"reasoning_options": [],
|
|
2019
1773
|
"tool_call": true,
|
|
2020
|
-
"interleaved": {
|
|
2021
|
-
"field": "reasoning_content"
|
|
2022
|
-
},
|
|
2023
1774
|
"temperature": true,
|
|
2024
|
-
"knowledge": "
|
|
2025
|
-
"release_date": "2026-
|
|
2026
|
-
"last_updated": "2026-
|
|
1775
|
+
"knowledge": "2026-03-01",
|
|
1776
|
+
"release_date": "2026-05-29",
|
|
1777
|
+
"last_updated": "2026-05-29",
|
|
2027
1778
|
"modalities": {
|
|
2028
1779
|
"input": [
|
|
2029
|
-
"text"
|
|
1780
|
+
"text",
|
|
1781
|
+
"image",
|
|
1782
|
+
"video"
|
|
2030
1783
|
],
|
|
2031
1784
|
"output": [
|
|
2032
1785
|
"text"
|
|
@@ -2034,15 +1787,15 @@
|
|
|
2034
1787
|
},
|
|
2035
1788
|
"open_weights": true,
|
|
2036
1789
|
"limit": {
|
|
2037
|
-
"context":
|
|
2038
|
-
"output":
|
|
1790
|
+
"context": 262144,
|
|
1791
|
+
"output": 256000
|
|
2039
1792
|
},
|
|
2040
|
-
"status": "deprecated",
|
|
2041
1793
|
"cost": {
|
|
2042
|
-
"input": 0.
|
|
1794
|
+
"input": 0.2,
|
|
2043
1795
|
"output": 1.15,
|
|
2044
|
-
"cache_read": 0.
|
|
2045
|
-
}
|
|
1796
|
+
"cache_read": 0.04
|
|
1797
|
+
},
|
|
1798
|
+
"canonical_model_id": "stepfun/step-3.7-flash"
|
|
2046
1799
|
},
|
|
2047
1800
|
"MiniMaxAI/MiniMax-M3": {
|
|
2048
1801
|
"id": "MiniMaxAI/MiniMax-M3",
|
|
@@ -2076,6 +1829,43 @@
|
|
|
2076
1829
|
"input": 0.28,
|
|
2077
1830
|
"output": 1.1,
|
|
2078
1831
|
"cache_read": 0.056
|
|
1832
|
+
},
|
|
1833
|
+
"canonical_model_id": "minimax/MiniMax-M3"
|
|
1834
|
+
},
|
|
1835
|
+
"MiniMaxAI/MiniMax-M2.5": {
|
|
1836
|
+
"id": "MiniMaxAI/MiniMax-M2.5",
|
|
1837
|
+
"name": "MiniMax M2.5",
|
|
1838
|
+
"description": "MiniMax model for chat, coding, office work, and agentic tasks",
|
|
1839
|
+
"family": "minimax",
|
|
1840
|
+
"attachment": false,
|
|
1841
|
+
"reasoning": true,
|
|
1842
|
+
"reasoning_options": [],
|
|
1843
|
+
"tool_call": true,
|
|
1844
|
+
"interleaved": {
|
|
1845
|
+
"field": "reasoning_content"
|
|
1846
|
+
},
|
|
1847
|
+
"temperature": true,
|
|
1848
|
+
"knowledge": "2025-06",
|
|
1849
|
+
"release_date": "2026-02-12",
|
|
1850
|
+
"last_updated": "2026-02-12",
|
|
1851
|
+
"modalities": {
|
|
1852
|
+
"input": [
|
|
1853
|
+
"text"
|
|
1854
|
+
],
|
|
1855
|
+
"output": [
|
|
1856
|
+
"text"
|
|
1857
|
+
]
|
|
1858
|
+
},
|
|
1859
|
+
"open_weights": true,
|
|
1860
|
+
"limit": {
|
|
1861
|
+
"context": 196608,
|
|
1862
|
+
"output": 131072
|
|
1863
|
+
},
|
|
1864
|
+
"status": "deprecated",
|
|
1865
|
+
"cost": {
|
|
1866
|
+
"input": 0.15,
|
|
1867
|
+
"output": 1.15,
|
|
1868
|
+
"cache_read": 0.03
|
|
2079
1869
|
}
|
|
2080
1870
|
},
|
|
2081
1871
|
"MiniMaxAI/MiniMax-M2.7": {
|
|
@@ -2108,23 +1898,293 @@
|
|
|
2108
1898
|
"input": 0.25,
|
|
2109
1899
|
"output": 1,
|
|
2110
1900
|
"cache_read": 0.05
|
|
1901
|
+
},
|
|
1902
|
+
"canonical_model_id": "minimax/MiniMax-M2.7"
|
|
1903
|
+
},
|
|
1904
|
+
"moonshotai/Kimi-K2.6": {
|
|
1905
|
+
"id": "moonshotai/Kimi-K2.6",
|
|
1906
|
+
"name": "Kimi K2.6",
|
|
1907
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
1908
|
+
"family": "kimi-k2",
|
|
1909
|
+
"attachment": true,
|
|
1910
|
+
"reasoning": true,
|
|
1911
|
+
"reasoning_options": [
|
|
1912
|
+
{
|
|
1913
|
+
"type": "toggle"
|
|
1914
|
+
}
|
|
1915
|
+
],
|
|
1916
|
+
"tool_call": true,
|
|
1917
|
+
"interleaved": {
|
|
1918
|
+
"field": "reasoning_content"
|
|
1919
|
+
},
|
|
1920
|
+
"structured_output": true,
|
|
1921
|
+
"temperature": true,
|
|
1922
|
+
"knowledge": "2024-04",
|
|
1923
|
+
"release_date": "2026-04-21",
|
|
1924
|
+
"last_updated": "2026-04-21",
|
|
1925
|
+
"modalities": {
|
|
1926
|
+
"input": [
|
|
1927
|
+
"text",
|
|
1928
|
+
"image",
|
|
1929
|
+
"video"
|
|
1930
|
+
],
|
|
1931
|
+
"output": [
|
|
1932
|
+
"text"
|
|
1933
|
+
]
|
|
1934
|
+
},
|
|
1935
|
+
"open_weights": true,
|
|
1936
|
+
"limit": {
|
|
1937
|
+
"context": 262144,
|
|
1938
|
+
"output": 16384
|
|
1939
|
+
},
|
|
1940
|
+
"cost": {
|
|
1941
|
+
"input": 0.75,
|
|
1942
|
+
"output": 3.5,
|
|
1943
|
+
"cache_read": 0.15
|
|
2111
1944
|
}
|
|
2112
1945
|
},
|
|
2113
|
-
"
|
|
2114
|
-
"id": "
|
|
2115
|
-
"name": "
|
|
2116
|
-
"description": "
|
|
2117
|
-
"family": "
|
|
1946
|
+
"moonshotai/Kimi-K3": {
|
|
1947
|
+
"id": "moonshotai/Kimi-K3",
|
|
1948
|
+
"name": "Kimi K3",
|
|
1949
|
+
"description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
|
|
1950
|
+
"family": "kimi-k3",
|
|
2118
1951
|
"attachment": true,
|
|
2119
|
-
"reasoning":
|
|
1952
|
+
"reasoning": true,
|
|
1953
|
+
"reasoning_options": [
|
|
1954
|
+
{
|
|
1955
|
+
"type": "effort",
|
|
1956
|
+
"values": [
|
|
1957
|
+
"low",
|
|
1958
|
+
"high",
|
|
1959
|
+
"max"
|
|
1960
|
+
]
|
|
1961
|
+
}
|
|
1962
|
+
],
|
|
2120
1963
|
"tool_call": true,
|
|
2121
1964
|
"structured_output": true,
|
|
2122
|
-
"
|
|
2123
|
-
"
|
|
1965
|
+
"temperature": false,
|
|
1966
|
+
"release_date": "2026-07-16",
|
|
1967
|
+
"last_updated": "2026-07-16",
|
|
1968
|
+
"modalities": {
|
|
1969
|
+
"input": [
|
|
1970
|
+
"text",
|
|
1971
|
+
"image"
|
|
1972
|
+
],
|
|
1973
|
+
"output": [
|
|
1974
|
+
"text"
|
|
1975
|
+
]
|
|
1976
|
+
},
|
|
1977
|
+
"open_weights": true,
|
|
1978
|
+
"limit": {
|
|
1979
|
+
"context": 1048576,
|
|
1980
|
+
"output": 131072
|
|
1981
|
+
},
|
|
1982
|
+
"cost": {
|
|
1983
|
+
"input": 2.85,
|
|
1984
|
+
"output": 14.25,
|
|
1985
|
+
"cache_read": 0.285
|
|
1986
|
+
},
|
|
1987
|
+
"canonical_model_id": "moonshotai/kimi-k3"
|
|
1988
|
+
},
|
|
1989
|
+
"moonshotai/Kimi-K2.7-Code": {
|
|
1990
|
+
"id": "moonshotai/Kimi-K2.7-Code",
|
|
1991
|
+
"name": "Kimi K2.7 Code",
|
|
1992
|
+
"description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
|
|
1993
|
+
"family": "kimi-k2",
|
|
1994
|
+
"attachment": true,
|
|
1995
|
+
"reasoning": true,
|
|
1996
|
+
"reasoning_options": [
|
|
1997
|
+
{
|
|
1998
|
+
"type": "toggle"
|
|
1999
|
+
}
|
|
2000
|
+
],
|
|
2001
|
+
"tool_call": true,
|
|
2002
|
+
"interleaved": {
|
|
2003
|
+
"field": "reasoning_content"
|
|
2004
|
+
},
|
|
2005
|
+
"structured_output": true,
|
|
2006
|
+
"temperature": false,
|
|
2007
|
+
"knowledge": "2025-01",
|
|
2008
|
+
"release_date": "2026-06-12",
|
|
2009
|
+
"last_updated": "2026-06-12",
|
|
2010
|
+
"modalities": {
|
|
2011
|
+
"input": [
|
|
2012
|
+
"text",
|
|
2013
|
+
"image",
|
|
2014
|
+
"video"
|
|
2015
|
+
],
|
|
2016
|
+
"output": [
|
|
2017
|
+
"text"
|
|
2018
|
+
]
|
|
2019
|
+
},
|
|
2020
|
+
"open_weights": true,
|
|
2021
|
+
"limit": {
|
|
2022
|
+
"context": 262144,
|
|
2023
|
+
"output": 262144
|
|
2024
|
+
},
|
|
2025
|
+
"cost": {
|
|
2026
|
+
"input": 0.68,
|
|
2027
|
+
"output": 3.4,
|
|
2028
|
+
"cache_read": 0.136
|
|
2029
|
+
},
|
|
2030
|
+
"canonical_model_id": "moonshotai/kimi-k2.7-code"
|
|
2031
|
+
},
|
|
2032
|
+
"moonshotai/Kimi-K2.5": {
|
|
2033
|
+
"id": "moonshotai/Kimi-K2.5",
|
|
2034
|
+
"name": "Kimi K2.5",
|
|
2035
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
2036
|
+
"family": "kimi-k2",
|
|
2037
|
+
"attachment": true,
|
|
2038
|
+
"reasoning": true,
|
|
2039
|
+
"reasoning_options": [
|
|
2040
|
+
{
|
|
2041
|
+
"type": "toggle"
|
|
2042
|
+
}
|
|
2043
|
+
],
|
|
2044
|
+
"tool_call": true,
|
|
2045
|
+
"interleaved": {
|
|
2046
|
+
"field": "reasoning_content"
|
|
2047
|
+
},
|
|
2048
|
+
"structured_output": true,
|
|
2049
|
+
"temperature": true,
|
|
2050
|
+
"knowledge": "2025-01",
|
|
2051
|
+
"release_date": "2026-01-27",
|
|
2052
|
+
"last_updated": "2026-01-27",
|
|
2053
|
+
"modalities": {
|
|
2054
|
+
"input": [
|
|
2055
|
+
"text",
|
|
2056
|
+
"image",
|
|
2057
|
+
"video"
|
|
2058
|
+
],
|
|
2059
|
+
"output": [
|
|
2060
|
+
"text"
|
|
2061
|
+
]
|
|
2062
|
+
},
|
|
2063
|
+
"open_weights": true,
|
|
2064
|
+
"limit": {
|
|
2065
|
+
"context": 262144,
|
|
2066
|
+
"output": 32768
|
|
2067
|
+
},
|
|
2068
|
+
"status": "deprecated",
|
|
2069
|
+
"cost": {
|
|
2070
|
+
"input": 0.45,
|
|
2071
|
+
"output": 2.25,
|
|
2072
|
+
"cache_read": 0.07
|
|
2073
|
+
}
|
|
2074
|
+
},
|
|
2075
|
+
"zai-org/GLM-5.1": {
|
|
2076
|
+
"id": "zai-org/GLM-5.1",
|
|
2077
|
+
"name": "GLM-5.1",
|
|
2078
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
2079
|
+
"family": "glm",
|
|
2080
|
+
"attachment": false,
|
|
2081
|
+
"reasoning": true,
|
|
2082
|
+
"reasoning_options": [
|
|
2083
|
+
{
|
|
2084
|
+
"type": "toggle"
|
|
2085
|
+
}
|
|
2086
|
+
],
|
|
2087
|
+
"tool_call": true,
|
|
2088
|
+
"interleaved": {
|
|
2089
|
+
"field": "reasoning_content"
|
|
2090
|
+
},
|
|
2091
|
+
"structured_output": true,
|
|
2092
|
+
"temperature": true,
|
|
2093
|
+
"knowledge": "2025-04",
|
|
2094
|
+
"release_date": "2026-04-07",
|
|
2095
|
+
"last_updated": "2026-04-07",
|
|
2096
|
+
"modalities": {
|
|
2097
|
+
"input": [
|
|
2098
|
+
"text"
|
|
2099
|
+
],
|
|
2100
|
+
"output": [
|
|
2101
|
+
"text"
|
|
2102
|
+
]
|
|
2103
|
+
},
|
|
2104
|
+
"open_weights": true,
|
|
2105
|
+
"limit": {
|
|
2106
|
+
"context": 202752,
|
|
2107
|
+
"output": 16384
|
|
2108
|
+
},
|
|
2109
|
+
"cost": {
|
|
2110
|
+
"input": 1.05,
|
|
2111
|
+
"output": 3.5,
|
|
2112
|
+
"cache_read": 0.205
|
|
2113
|
+
}
|
|
2114
|
+
},
|
|
2115
|
+
"zai-org/GLM-5.2": {
|
|
2116
|
+
"id": "zai-org/GLM-5.2",
|
|
2117
|
+
"name": "GLM-5.2",
|
|
2118
|
+
"description": "Open flagship GLM for long-horizon coding agents and million-token context work",
|
|
2119
|
+
"family": "glm",
|
|
2120
|
+
"attachment": false,
|
|
2121
|
+
"reasoning": true,
|
|
2122
|
+
"reasoning_options": [
|
|
2123
|
+
{
|
|
2124
|
+
"type": "toggle"
|
|
2125
|
+
},
|
|
2126
|
+
{
|
|
2127
|
+
"type": "effort",
|
|
2128
|
+
"values": [
|
|
2129
|
+
"low",
|
|
2130
|
+
"medium",
|
|
2131
|
+
"high",
|
|
2132
|
+
"xhigh"
|
|
2133
|
+
]
|
|
2134
|
+
}
|
|
2135
|
+
],
|
|
2136
|
+
"tool_call": true,
|
|
2137
|
+
"interleaved": {
|
|
2138
|
+
"field": "reasoning_content"
|
|
2139
|
+
},
|
|
2140
|
+
"structured_output": true,
|
|
2141
|
+
"temperature": true,
|
|
2142
|
+
"release_date": "2026-06-13",
|
|
2143
|
+
"last_updated": "2026-06-13",
|
|
2144
|
+
"modalities": {
|
|
2145
|
+
"input": [
|
|
2146
|
+
"text"
|
|
2147
|
+
],
|
|
2148
|
+
"output": [
|
|
2149
|
+
"text"
|
|
2150
|
+
]
|
|
2151
|
+
},
|
|
2152
|
+
"open_weights": true,
|
|
2153
|
+
"limit": {
|
|
2154
|
+
"context": 1048576,
|
|
2155
|
+
"output": 32768
|
|
2156
|
+
},
|
|
2157
|
+
"cost": {
|
|
2158
|
+
"input": 0.75,
|
|
2159
|
+
"output": 2.4,
|
|
2160
|
+
"cache_read": 0.14
|
|
2161
|
+
},
|
|
2162
|
+
"canonical_model_id": "zhipuai/glm-5.2"
|
|
2163
|
+
},
|
|
2164
|
+
"zai-org/GLM-4.7": {
|
|
2165
|
+
"id": "zai-org/GLM-4.7",
|
|
2166
|
+
"name": "GLM-4.7",
|
|
2167
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
2168
|
+
"family": "glm",
|
|
2169
|
+
"attachment": false,
|
|
2170
|
+
"reasoning": true,
|
|
2171
|
+
"reasoning_options": [
|
|
2172
|
+
{
|
|
2173
|
+
"type": "toggle"
|
|
2174
|
+
}
|
|
2175
|
+
],
|
|
2176
|
+
"tool_call": true,
|
|
2177
|
+
"interleaved": {
|
|
2178
|
+
"field": "reasoning_content"
|
|
2179
|
+
},
|
|
2180
|
+
"structured_output": true,
|
|
2181
|
+
"temperature": true,
|
|
2182
|
+
"knowledge": "2025-04",
|
|
2183
|
+
"release_date": "2025-12-22",
|
|
2184
|
+
"last_updated": "2025-12-22",
|
|
2124
2185
|
"modalities": {
|
|
2125
2186
|
"input": [
|
|
2126
|
-
"text"
|
|
2127
|
-
"image"
|
|
2187
|
+
"text"
|
|
2128
2188
|
],
|
|
2129
2189
|
"output": [
|
|
2130
2190
|
"text"
|
|
@@ -2132,25 +2192,36 @@
|
|
|
2132
2192
|
},
|
|
2133
2193
|
"open_weights": true,
|
|
2134
2194
|
"limit": {
|
|
2135
|
-
"context":
|
|
2195
|
+
"context": 202752,
|
|
2136
2196
|
"output": 16384
|
|
2137
2197
|
},
|
|
2138
2198
|
"cost": {
|
|
2139
|
-
"input": 0.
|
|
2140
|
-
"output":
|
|
2199
|
+
"input": 0.4,
|
|
2200
|
+
"output": 1.75,
|
|
2201
|
+
"cache_read": 0.08
|
|
2141
2202
|
}
|
|
2142
2203
|
},
|
|
2143
|
-
"
|
|
2144
|
-
"id": "
|
|
2145
|
-
"name": "
|
|
2146
|
-
"description": "
|
|
2147
|
-
"family": "
|
|
2204
|
+
"zai-org/GLM-5": {
|
|
2205
|
+
"id": "zai-org/GLM-5",
|
|
2206
|
+
"name": "GLM-5",
|
|
2207
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
2208
|
+
"family": "glm",
|
|
2148
2209
|
"attachment": false,
|
|
2149
|
-
"reasoning":
|
|
2210
|
+
"reasoning": true,
|
|
2211
|
+
"reasoning_options": [
|
|
2212
|
+
{
|
|
2213
|
+
"type": "toggle"
|
|
2214
|
+
}
|
|
2215
|
+
],
|
|
2150
2216
|
"tool_call": true,
|
|
2217
|
+
"interleaved": {
|
|
2218
|
+
"field": "reasoning_content"
|
|
2219
|
+
},
|
|
2151
2220
|
"structured_output": true,
|
|
2152
|
-
"
|
|
2153
|
-
"
|
|
2221
|
+
"temperature": true,
|
|
2222
|
+
"knowledge": "2025-12",
|
|
2223
|
+
"release_date": "2026-02-12",
|
|
2224
|
+
"last_updated": "2026-02-12",
|
|
2154
2225
|
"modalities": {
|
|
2155
2226
|
"input": [
|
|
2156
2227
|
"text"
|
|
@@ -2161,29 +2232,44 @@
|
|
|
2161
2232
|
},
|
|
2162
2233
|
"open_weights": true,
|
|
2163
2234
|
"limit": {
|
|
2164
|
-
"context":
|
|
2235
|
+
"context": 202752,
|
|
2165
2236
|
"output": 16384
|
|
2166
2237
|
},
|
|
2238
|
+
"status": "deprecated",
|
|
2167
2239
|
"cost": {
|
|
2168
|
-
"input": 0.
|
|
2169
|
-
"output":
|
|
2240
|
+
"input": 0.6,
|
|
2241
|
+
"output": 2.08,
|
|
2242
|
+
"cache_read": 0.12
|
|
2170
2243
|
}
|
|
2171
2244
|
},
|
|
2172
|
-
"
|
|
2173
|
-
"id": "
|
|
2174
|
-
"name": "
|
|
2175
|
-
"description": "
|
|
2176
|
-
"family": "
|
|
2245
|
+
"zai-org/GLM-5.3-Flash": {
|
|
2246
|
+
"id": "zai-org/GLM-5.3-Flash",
|
|
2247
|
+
"name": "GLM-5.3-Flash",
|
|
2248
|
+
"description": "Native multimodal GLM model for efficient coding and long-horizon agent tasks",
|
|
2249
|
+
"family": "glm-flash",
|
|
2177
2250
|
"attachment": true,
|
|
2178
|
-
"reasoning":
|
|
2179
|
-
"
|
|
2251
|
+
"reasoning": true,
|
|
2252
|
+
"reasoning_options": [
|
|
2253
|
+
{
|
|
2254
|
+
"type": "effort",
|
|
2255
|
+
"values": [
|
|
2256
|
+
"low",
|
|
2257
|
+
"high",
|
|
2258
|
+
"max"
|
|
2259
|
+
]
|
|
2260
|
+
}
|
|
2261
|
+
],
|
|
2262
|
+
"tool_call": true,
|
|
2180
2263
|
"structured_output": true,
|
|
2181
|
-
"
|
|
2182
|
-
"
|
|
2264
|
+
"temperature": true,
|
|
2265
|
+
"release_date": "2026-08-26",
|
|
2266
|
+
"last_updated": "2026-08-26",
|
|
2183
2267
|
"modalities": {
|
|
2184
2268
|
"input": [
|
|
2185
2269
|
"text",
|
|
2186
|
-
"image"
|
|
2270
|
+
"image",
|
|
2271
|
+
"video",
|
|
2272
|
+
"pdf"
|
|
2187
2273
|
],
|
|
2188
2274
|
"output": [
|
|
2189
2275
|
"text"
|
|
@@ -2192,35 +2278,36 @@
|
|
|
2192
2278
|
"open_weights": true,
|
|
2193
2279
|
"limit": {
|
|
2194
2280
|
"context": 1048576,
|
|
2195
|
-
"output":
|
|
2281
|
+
"output": 131072
|
|
2196
2282
|
},
|
|
2197
2283
|
"cost": {
|
|
2198
|
-
"input": 0.
|
|
2199
|
-
"output": 0.
|
|
2200
|
-
|
|
2284
|
+
"input": 0.15,
|
|
2285
|
+
"output": 0.5,
|
|
2286
|
+
"cache_read": 0.03
|
|
2287
|
+
},
|
|
2288
|
+
"canonical_model_id": "zhipuai/glm-5.3-flash"
|
|
2201
2289
|
},
|
|
2202
|
-
"
|
|
2203
|
-
"id": "
|
|
2204
|
-
"name": "
|
|
2205
|
-
"description": "
|
|
2206
|
-
"family": "
|
|
2290
|
+
"zai-org/GLM-4.6": {
|
|
2291
|
+
"id": "zai-org/GLM-4.6",
|
|
2292
|
+
"name": "GLM-4.6",
|
|
2293
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
2294
|
+
"family": "glm",
|
|
2207
2295
|
"attachment": false,
|
|
2208
2296
|
"reasoning": true,
|
|
2209
2297
|
"reasoning_options": [
|
|
2210
2298
|
{
|
|
2211
|
-
"type": "
|
|
2212
|
-
"values": [
|
|
2213
|
-
"low",
|
|
2214
|
-
"medium",
|
|
2215
|
-
"high"
|
|
2216
|
-
]
|
|
2299
|
+
"type": "toggle"
|
|
2217
2300
|
}
|
|
2218
2301
|
],
|
|
2219
2302
|
"tool_call": true,
|
|
2303
|
+
"interleaved": {
|
|
2304
|
+
"field": "reasoning_content"
|
|
2305
|
+
},
|
|
2220
2306
|
"structured_output": true,
|
|
2221
2307
|
"temperature": true,
|
|
2222
|
-
"
|
|
2223
|
-
"
|
|
2308
|
+
"knowledge": "2025-04",
|
|
2309
|
+
"release_date": "2025-09-30",
|
|
2310
|
+
"last_updated": "2025-09-30",
|
|
2224
2311
|
"modalities": {
|
|
2225
2312
|
"input": [
|
|
2226
2313
|
"text"
|
|
@@ -2231,19 +2318,57 @@
|
|
|
2231
2318
|
},
|
|
2232
2319
|
"open_weights": true,
|
|
2233
2320
|
"limit": {
|
|
2234
|
-
"context":
|
|
2321
|
+
"context": 202752,
|
|
2322
|
+
"output": 131072
|
|
2323
|
+
},
|
|
2324
|
+
"cost": {
|
|
2325
|
+
"input": 0.5,
|
|
2326
|
+
"output": 2,
|
|
2327
|
+
"cache_read": 0.1
|
|
2328
|
+
}
|
|
2329
|
+
},
|
|
2330
|
+
"zai-org/GLM-4.7-Flash": {
|
|
2331
|
+
"id": "zai-org/GLM-4.7-Flash",
|
|
2332
|
+
"name": "GLM-4.7-Flash",
|
|
2333
|
+
"description": "Efficient GLM model for fast reasoning, coding, and agent workflows",
|
|
2334
|
+
"family": "glm-flash",
|
|
2335
|
+
"attachment": false,
|
|
2336
|
+
"reasoning": true,
|
|
2337
|
+
"reasoning_options": [],
|
|
2338
|
+
"tool_call": true,
|
|
2339
|
+
"interleaved": {
|
|
2340
|
+
"field": "reasoning_content"
|
|
2341
|
+
},
|
|
2342
|
+
"structured_output": true,
|
|
2343
|
+
"temperature": true,
|
|
2344
|
+
"knowledge": "2025-04",
|
|
2345
|
+
"release_date": "2026-01-19",
|
|
2346
|
+
"last_updated": "2026-01-19",
|
|
2347
|
+
"modalities": {
|
|
2348
|
+
"input": [
|
|
2349
|
+
"text"
|
|
2350
|
+
],
|
|
2351
|
+
"output": [
|
|
2352
|
+
"text"
|
|
2353
|
+
]
|
|
2354
|
+
},
|
|
2355
|
+
"open_weights": true,
|
|
2356
|
+
"limit": {
|
|
2357
|
+
"context": 202752,
|
|
2235
2358
|
"output": 16384
|
|
2236
2359
|
},
|
|
2360
|
+
"status": "deprecated",
|
|
2237
2361
|
"cost": {
|
|
2238
|
-
"input": 0.
|
|
2239
|
-
"output": 0.
|
|
2362
|
+
"input": 0.06,
|
|
2363
|
+
"output": 0.4,
|
|
2364
|
+
"cache_read": 0.01
|
|
2240
2365
|
}
|
|
2241
2366
|
},
|
|
2242
|
-
"
|
|
2243
|
-
"id": "
|
|
2244
|
-
"name": "
|
|
2245
|
-
"description": "
|
|
2246
|
-
"family": "
|
|
2367
|
+
"zai-org/GLM-5.3": {
|
|
2368
|
+
"id": "zai-org/GLM-5.3",
|
|
2369
|
+
"name": "GLM-5.3",
|
|
2370
|
+
"description": "Flagship GLM model for long-horizon coding, agents, and complex project delivery",
|
|
2371
|
+
"family": "glm",
|
|
2247
2372
|
"attachment": false,
|
|
2248
2373
|
"reasoning": true,
|
|
2249
2374
|
"reasoning_options": [
|
|
@@ -2251,16 +2376,16 @@
|
|
|
2251
2376
|
"type": "effort",
|
|
2252
2377
|
"values": [
|
|
2253
2378
|
"low",
|
|
2254
|
-
"
|
|
2255
|
-
"
|
|
2379
|
+
"high",
|
|
2380
|
+
"max"
|
|
2256
2381
|
]
|
|
2257
2382
|
}
|
|
2258
2383
|
],
|
|
2259
2384
|
"tool_call": true,
|
|
2260
2385
|
"structured_output": true,
|
|
2261
2386
|
"temperature": true,
|
|
2262
|
-
"release_date": "
|
|
2263
|
-
"last_updated": "
|
|
2387
|
+
"release_date": "2026-08-14",
|
|
2388
|
+
"last_updated": "2026-08-14",
|
|
2264
2389
|
"modalities": {
|
|
2265
2390
|
"input": [
|
|
2266
2391
|
"text"
|
|
@@ -2271,20 +2396,22 @@
|
|
|
2271
2396
|
},
|
|
2272
2397
|
"open_weights": true,
|
|
2273
2398
|
"limit": {
|
|
2274
|
-
"context":
|
|
2275
|
-
"output":
|
|
2399
|
+
"context": 1048576,
|
|
2400
|
+
"output": 131072
|
|
2276
2401
|
},
|
|
2277
2402
|
"cost": {
|
|
2278
|
-
"input": 0.
|
|
2279
|
-
"output":
|
|
2280
|
-
|
|
2403
|
+
"input": 0.9,
|
|
2404
|
+
"output": 4,
|
|
2405
|
+
"cache_read": 0.2
|
|
2406
|
+
},
|
|
2407
|
+
"canonical_model_id": "zhipuai/glm-5.3"
|
|
2281
2408
|
},
|
|
2282
|
-
"
|
|
2283
|
-
"id": "
|
|
2284
|
-
"name": "
|
|
2285
|
-
"description": "
|
|
2286
|
-
"family": "
|
|
2287
|
-
"attachment":
|
|
2409
|
+
"nvidia/Nemotron-3-Nano-30B-A3B": {
|
|
2410
|
+
"id": "nvidia/Nemotron-3-Nano-30B-A3B",
|
|
2411
|
+
"name": "Nemotron 3 Nano 30B A3B",
|
|
2412
|
+
"description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents",
|
|
2413
|
+
"family": "nemotron",
|
|
2414
|
+
"attachment": false,
|
|
2288
2415
|
"reasoning": true,
|
|
2289
2416
|
"reasoning_options": [
|
|
2290
2417
|
{
|
|
@@ -2292,19 +2419,12 @@
|
|
|
2292
2419
|
}
|
|
2293
2420
|
],
|
|
2294
2421
|
"tool_call": true,
|
|
2295
|
-
"interleaved": {
|
|
2296
|
-
"field": "reasoning_content"
|
|
2297
|
-
},
|
|
2298
|
-
"structured_output": true,
|
|
2299
2422
|
"temperature": true,
|
|
2300
|
-
"
|
|
2301
|
-
"
|
|
2302
|
-
"last_updated": "2026-01-27",
|
|
2423
|
+
"release_date": "2025-12-15",
|
|
2424
|
+
"last_updated": "2025-12-15",
|
|
2303
2425
|
"modalities": {
|
|
2304
2426
|
"input": [
|
|
2305
|
-
"text"
|
|
2306
|
-
"image",
|
|
2307
|
-
"video"
|
|
2427
|
+
"text"
|
|
2308
2428
|
],
|
|
2309
2429
|
"output": [
|
|
2310
2430
|
"text"
|
|
@@ -2313,41 +2433,31 @@
|
|
|
2313
2433
|
"open_weights": true,
|
|
2314
2434
|
"limit": {
|
|
2315
2435
|
"context": 262144,
|
|
2316
|
-
"output":
|
|
2436
|
+
"output": 262144
|
|
2317
2437
|
},
|
|
2318
|
-
"status": "deprecated",
|
|
2319
2438
|
"cost": {
|
|
2320
|
-
"input": 0.
|
|
2321
|
-
"output": 2
|
|
2322
|
-
"cache_read": 0.
|
|
2323
|
-
}
|
|
2439
|
+
"input": 0.05,
|
|
2440
|
+
"output": 0.2,
|
|
2441
|
+
"cache_read": 0.025
|
|
2442
|
+
},
|
|
2443
|
+
"canonical_model_id": "nvidia/nemotron-3-nano-30b-a3b"
|
|
2324
2444
|
},
|
|
2325
|
-
"
|
|
2326
|
-
"id": "
|
|
2327
|
-
"name": "
|
|
2328
|
-
"description": "
|
|
2329
|
-
"family": "
|
|
2330
|
-
"attachment":
|
|
2445
|
+
"nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": {
|
|
2446
|
+
"id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5",
|
|
2447
|
+
"name": "Llama 3.3 Nemotron Super 49B v1.5",
|
|
2448
|
+
"description": "Nemotron model for efficient reasoning, coding, and specialized AI agents",
|
|
2449
|
+
"family": "nemotron",
|
|
2450
|
+
"attachment": false,
|
|
2331
2451
|
"reasoning": true,
|
|
2332
|
-
"reasoning_options": [
|
|
2333
|
-
{
|
|
2334
|
-
"type": "toggle"
|
|
2335
|
-
}
|
|
2336
|
-
],
|
|
2452
|
+
"reasoning_options": [],
|
|
2337
2453
|
"tool_call": true,
|
|
2338
|
-
"interleaved": {
|
|
2339
|
-
"field": "reasoning_content"
|
|
2340
|
-
},
|
|
2341
2454
|
"structured_output": true,
|
|
2342
|
-
"temperature":
|
|
2343
|
-
"
|
|
2344
|
-
"
|
|
2345
|
-
"last_updated": "2026-06-12",
|
|
2455
|
+
"temperature": true,
|
|
2456
|
+
"release_date": "2025-07-25",
|
|
2457
|
+
"last_updated": "2025-07-25",
|
|
2346
2458
|
"modalities": {
|
|
2347
|
-
"input": [
|
|
2348
|
-
"text"
|
|
2349
|
-
"image",
|
|
2350
|
-
"video"
|
|
2459
|
+
"input": [
|
|
2460
|
+
"text"
|
|
2351
2461
|
],
|
|
2352
2462
|
"output": [
|
|
2353
2463
|
"text"
|
|
@@ -2355,41 +2465,35 @@
|
|
|
2355
2465
|
},
|
|
2356
2466
|
"open_weights": true,
|
|
2357
2467
|
"limit": {
|
|
2358
|
-
"context":
|
|
2359
|
-
"output":
|
|
2468
|
+
"context": 131072,
|
|
2469
|
+
"output": 131072
|
|
2360
2470
|
},
|
|
2471
|
+
"status": "deprecated",
|
|
2361
2472
|
"cost": {
|
|
2362
|
-
"input": 0.
|
|
2363
|
-
"output":
|
|
2364
|
-
|
|
2365
|
-
|
|
2473
|
+
"input": 0.4,
|
|
2474
|
+
"output": 0.4
|
|
2475
|
+
},
|
|
2476
|
+
"canonical_model_id": "nvidia/llama-3.3-nemotron-super-49b-v1.5"
|
|
2366
2477
|
},
|
|
2367
|
-
"
|
|
2368
|
-
"id": "
|
|
2369
|
-
"name": "
|
|
2370
|
-
"description": "
|
|
2371
|
-
"family": "
|
|
2478
|
+
"nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning": {
|
|
2479
|
+
"id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning",
|
|
2480
|
+
"name": "Nemotron 3 Nano Omni 30B A3B Reasoning",
|
|
2481
|
+
"description": "Open Nemotron omni model combining reasoning with text, vision, and audio",
|
|
2482
|
+
"family": "nemotron",
|
|
2372
2483
|
"attachment": true,
|
|
2373
2484
|
"reasoning": true,
|
|
2374
|
-
"reasoning_options": [
|
|
2375
|
-
{
|
|
2376
|
-
"type": "toggle"
|
|
2377
|
-
}
|
|
2378
|
-
],
|
|
2485
|
+
"reasoning_options": [],
|
|
2379
2486
|
"tool_call": true,
|
|
2380
|
-
"interleaved": {
|
|
2381
|
-
"field": "reasoning_content"
|
|
2382
|
-
},
|
|
2383
2487
|
"structured_output": true,
|
|
2384
2488
|
"temperature": true,
|
|
2385
|
-
"
|
|
2386
|
-
"
|
|
2387
|
-
"last_updated": "2026-04-21",
|
|
2489
|
+
"release_date": "2026-04-28",
|
|
2490
|
+
"last_updated": "2026-04-28",
|
|
2388
2491
|
"modalities": {
|
|
2389
2492
|
"input": [
|
|
2390
2493
|
"text",
|
|
2391
2494
|
"image",
|
|
2392
|
-
"video"
|
|
2495
|
+
"video",
|
|
2496
|
+
"audio"
|
|
2393
2497
|
],
|
|
2394
2498
|
"output": [
|
|
2395
2499
|
"text"
|
|
@@ -2398,36 +2502,38 @@
|
|
|
2398
2502
|
"open_weights": true,
|
|
2399
2503
|
"limit": {
|
|
2400
2504
|
"context": 262144,
|
|
2401
|
-
"output":
|
|
2505
|
+
"output": 65536
|
|
2402
2506
|
},
|
|
2507
|
+
"status": "deprecated",
|
|
2403
2508
|
"cost": {
|
|
2404
|
-
"input": 0.
|
|
2405
|
-
"output":
|
|
2406
|
-
|
|
2407
|
-
|
|
2509
|
+
"input": 0.2,
|
|
2510
|
+
"output": 0.8
|
|
2511
|
+
},
|
|
2512
|
+
"canonical_model_id": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning"
|
|
2408
2513
|
},
|
|
2409
|
-
"
|
|
2410
|
-
"id": "
|
|
2411
|
-
"name": "
|
|
2412
|
-
"description": "
|
|
2413
|
-
"family": "
|
|
2514
|
+
"ByteDance/Seed-2.0-code": {
|
|
2515
|
+
"id": "ByteDance/Seed-2.0-code",
|
|
2516
|
+
"name": "Seed 2.0 Code",
|
|
2517
|
+
"description": "ByteDance Seed coding model for multimodal software engineering and long-running agents",
|
|
2518
|
+
"family": "seed",
|
|
2414
2519
|
"attachment": true,
|
|
2415
2520
|
"reasoning": true,
|
|
2416
2521
|
"reasoning_options": [
|
|
2417
2522
|
{
|
|
2418
2523
|
"type": "effort",
|
|
2419
2524
|
"values": [
|
|
2525
|
+
"none",
|
|
2420
2526
|
"low",
|
|
2421
|
-
"
|
|
2422
|
-
"
|
|
2527
|
+
"medium",
|
|
2528
|
+
"high"
|
|
2423
2529
|
]
|
|
2424
2530
|
}
|
|
2425
2531
|
],
|
|
2426
2532
|
"tool_call": true,
|
|
2427
2533
|
"structured_output": true,
|
|
2428
|
-
"temperature":
|
|
2429
|
-
"release_date": "2026-
|
|
2430
|
-
"last_updated": "2026-
|
|
2534
|
+
"temperature": true,
|
|
2535
|
+
"release_date": "2026-02-14",
|
|
2536
|
+
"last_updated": "2026-02-14",
|
|
2431
2537
|
"modalities": {
|
|
2432
2538
|
"input": [
|
|
2433
2539
|
"text",
|
|
@@ -2437,75 +2543,144 @@
|
|
|
2437
2543
|
"text"
|
|
2438
2544
|
]
|
|
2439
2545
|
},
|
|
2440
|
-
"open_weights":
|
|
2546
|
+
"open_weights": false,
|
|
2441
2547
|
"limit": {
|
|
2442
|
-
"context":
|
|
2548
|
+
"context": 256000,
|
|
2443
2549
|
"output": 131072
|
|
2444
2550
|
},
|
|
2445
2551
|
"cost": {
|
|
2446
|
-
"input":
|
|
2447
|
-
"output":
|
|
2448
|
-
"cache_read": 0.
|
|
2449
|
-
|
|
2552
|
+
"input": 0.5,
|
|
2553
|
+
"output": 3,
|
|
2554
|
+
"cache_read": 0.1,
|
|
2555
|
+
"tiers": [
|
|
2556
|
+
{
|
|
2557
|
+
"input": 1,
|
|
2558
|
+
"output": 6,
|
|
2559
|
+
"cache_read": 0.2,
|
|
2560
|
+
"tier": {
|
|
2561
|
+
"type": "context",
|
|
2562
|
+
"size": 128000
|
|
2563
|
+
}
|
|
2564
|
+
}
|
|
2565
|
+
]
|
|
2566
|
+
},
|
|
2567
|
+
"canonical_model_id": "bytedance-seed/seed-2.0-code"
|
|
2450
2568
|
},
|
|
2451
|
-
"
|
|
2452
|
-
"id": "
|
|
2453
|
-
"name": "
|
|
2454
|
-
"description": "
|
|
2455
|
-
"family": "
|
|
2456
|
-
"attachment":
|
|
2569
|
+
"ByteDance/Seed-2.0-pro": {
|
|
2570
|
+
"id": "ByteDance/Seed-2.0-pro",
|
|
2571
|
+
"name": "Seed 2.0 Pro",
|
|
2572
|
+
"description": "Flagship ByteDance Seed 2.0 model for complex multimodal reasoning and long-horizon agent workflows",
|
|
2573
|
+
"family": "seed",
|
|
2574
|
+
"attachment": true,
|
|
2457
2575
|
"reasoning": true,
|
|
2458
2576
|
"reasoning_options": [],
|
|
2459
2577
|
"tool_call": true,
|
|
2460
2578
|
"structured_output": true,
|
|
2461
2579
|
"temperature": true,
|
|
2462
|
-
"release_date": "2026-
|
|
2463
|
-
"last_updated": "2026-
|
|
2580
|
+
"release_date": "2026-02-14",
|
|
2581
|
+
"last_updated": "2026-02-14",
|
|
2464
2582
|
"modalities": {
|
|
2465
2583
|
"input": [
|
|
2466
|
-
"text"
|
|
2584
|
+
"text",
|
|
2585
|
+
"image"
|
|
2467
2586
|
],
|
|
2468
2587
|
"output": [
|
|
2469
2588
|
"text"
|
|
2470
2589
|
]
|
|
2471
2590
|
},
|
|
2472
|
-
"open_weights":
|
|
2591
|
+
"open_weights": false,
|
|
2473
2592
|
"limit": {
|
|
2474
|
-
"context":
|
|
2475
|
-
"input": 192000,
|
|
2593
|
+
"context": 256000,
|
|
2476
2594
|
"output": 128000
|
|
2477
2595
|
},
|
|
2478
2596
|
"cost": {
|
|
2479
|
-
"input": 0.
|
|
2480
|
-
"output":
|
|
2481
|
-
"cache_read": 0.
|
|
2482
|
-
|
|
2597
|
+
"input": 0.5,
|
|
2598
|
+
"output": 3,
|
|
2599
|
+
"cache_read": 0.1,
|
|
2600
|
+
"tiers": [
|
|
2601
|
+
{
|
|
2602
|
+
"input": 1,
|
|
2603
|
+
"output": 6,
|
|
2604
|
+
"cache_read": 0.2,
|
|
2605
|
+
"tier": {
|
|
2606
|
+
"type": "context",
|
|
2607
|
+
"size": 128000
|
|
2608
|
+
}
|
|
2609
|
+
}
|
|
2610
|
+
]
|
|
2611
|
+
},
|
|
2612
|
+
"canonical_model_id": "bytedance-seed/seed-2.0-pro"
|
|
2483
2613
|
},
|
|
2484
|
-
"
|
|
2485
|
-
"id": "
|
|
2486
|
-
"name": "
|
|
2487
|
-
"description": "
|
|
2488
|
-
"family": "
|
|
2614
|
+
"ByteDance/Seed-2.0-mini": {
|
|
2615
|
+
"id": "ByteDance/Seed-2.0-mini",
|
|
2616
|
+
"name": "Seed 2.0 Mini",
|
|
2617
|
+
"description": "Lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks",
|
|
2618
|
+
"family": "seed",
|
|
2489
2619
|
"attachment": true,
|
|
2490
2620
|
"reasoning": true,
|
|
2621
|
+
"reasoning_options": [],
|
|
2622
|
+
"tool_call": true,
|
|
2623
|
+
"structured_output": true,
|
|
2624
|
+
"temperature": true,
|
|
2625
|
+
"release_date": "2026-02-14",
|
|
2626
|
+
"last_updated": "2026-02-14",
|
|
2627
|
+
"modalities": {
|
|
2628
|
+
"input": [
|
|
2629
|
+
"text",
|
|
2630
|
+
"image"
|
|
2631
|
+
],
|
|
2632
|
+
"output": [
|
|
2633
|
+
"text"
|
|
2634
|
+
]
|
|
2635
|
+
},
|
|
2636
|
+
"open_weights": false,
|
|
2637
|
+
"limit": {
|
|
2638
|
+
"context": 256000,
|
|
2639
|
+
"output": 32000
|
|
2640
|
+
},
|
|
2641
|
+
"cost": {
|
|
2642
|
+
"input": 0.1,
|
|
2643
|
+
"output": 0.4,
|
|
2644
|
+
"cache_read": 0.02,
|
|
2645
|
+
"tiers": [
|
|
2646
|
+
{
|
|
2647
|
+
"input": 0.2,
|
|
2648
|
+
"output": 0.8,
|
|
2649
|
+
"cache_read": 0.2,
|
|
2650
|
+
"tier": {
|
|
2651
|
+
"type": "context",
|
|
2652
|
+
"size": 128000
|
|
2653
|
+
}
|
|
2654
|
+
}
|
|
2655
|
+
]
|
|
2656
|
+
},
|
|
2657
|
+
"canonical_model_id": "bytedance-seed/seed-2.0-mini"
|
|
2658
|
+
},
|
|
2659
|
+
"openai/gpt-oss-20b": {
|
|
2660
|
+
"id": "openai/gpt-oss-20b",
|
|
2661
|
+
"name": "GPT OSS 20B",
|
|
2662
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
2663
|
+
"family": "gpt-oss",
|
|
2664
|
+
"attachment": false,
|
|
2665
|
+
"reasoning": true,
|
|
2491
2666
|
"reasoning_options": [
|
|
2492
2667
|
{
|
|
2493
|
-
"type": "
|
|
2668
|
+
"type": "effort",
|
|
2669
|
+
"values": [
|
|
2670
|
+
"low",
|
|
2671
|
+
"medium",
|
|
2672
|
+
"high"
|
|
2673
|
+
]
|
|
2494
2674
|
}
|
|
2495
2675
|
],
|
|
2496
2676
|
"tool_call": true,
|
|
2497
|
-
"interleaved": {
|
|
2498
|
-
"field": "reasoning_content"
|
|
2499
|
-
},
|
|
2500
2677
|
"structured_output": true,
|
|
2501
2678
|
"temperature": true,
|
|
2502
|
-
"
|
|
2503
|
-
"
|
|
2504
|
-
"last_updated": "2026-04-22",
|
|
2679
|
+
"release_date": "2025-08-05",
|
|
2680
|
+
"last_updated": "2025-08-05",
|
|
2505
2681
|
"modalities": {
|
|
2506
2682
|
"input": [
|
|
2507
|
-
"text"
|
|
2508
|
-
"audio"
|
|
2683
|
+
"text"
|
|
2509
2684
|
],
|
|
2510
2685
|
"output": [
|
|
2511
2686
|
"text"
|
|
@@ -2513,42 +2688,39 @@
|
|
|
2513
2688
|
},
|
|
2514
2689
|
"open_weights": true,
|
|
2515
2690
|
"limit": {
|
|
2516
|
-
"context":
|
|
2691
|
+
"context": 131072,
|
|
2517
2692
|
"output": 16384
|
|
2518
2693
|
},
|
|
2519
2694
|
"cost": {
|
|
2520
|
-
"input":
|
|
2521
|
-
"output":
|
|
2522
|
-
"cache_read": 0.2
|
|
2695
|
+
"input": 0.03,
|
|
2696
|
+
"output": 0.14
|
|
2523
2697
|
}
|
|
2524
2698
|
},
|
|
2525
|
-
"
|
|
2526
|
-
"id": "
|
|
2527
|
-
"name": "
|
|
2528
|
-
"description": "Open
|
|
2529
|
-
"family": "
|
|
2530
|
-
"attachment":
|
|
2699
|
+
"openai/gpt-oss-120b": {
|
|
2700
|
+
"id": "openai/gpt-oss-120b",
|
|
2701
|
+
"name": "GPT OSS 120B",
|
|
2702
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
2703
|
+
"family": "gpt-oss",
|
|
2704
|
+
"attachment": false,
|
|
2531
2705
|
"reasoning": true,
|
|
2532
2706
|
"reasoning_options": [
|
|
2533
2707
|
{
|
|
2534
|
-
"type": "
|
|
2708
|
+
"type": "effort",
|
|
2709
|
+
"values": [
|
|
2710
|
+
"low",
|
|
2711
|
+
"medium",
|
|
2712
|
+
"high"
|
|
2713
|
+
]
|
|
2535
2714
|
}
|
|
2536
2715
|
],
|
|
2537
2716
|
"tool_call": true,
|
|
2538
|
-
"interleaved": {
|
|
2539
|
-
"field": "reasoning_content"
|
|
2540
|
-
},
|
|
2541
2717
|
"structured_output": true,
|
|
2542
2718
|
"temperature": true,
|
|
2543
|
-
"
|
|
2544
|
-
"
|
|
2545
|
-
"last_updated": "2026-04-22",
|
|
2719
|
+
"release_date": "2025-08-05",
|
|
2720
|
+
"last_updated": "2025-08-05",
|
|
2546
2721
|
"modalities": {
|
|
2547
2722
|
"input": [
|
|
2548
|
-
"text"
|
|
2549
|
-
"image",
|
|
2550
|
-
"audio",
|
|
2551
|
-
"video"
|
|
2723
|
+
"text"
|
|
2552
2724
|
],
|
|
2553
2725
|
"output": [
|
|
2554
2726
|
"text"
|
|
@@ -2556,13 +2728,12 @@
|
|
|
2556
2728
|
},
|
|
2557
2729
|
"open_weights": true,
|
|
2558
2730
|
"limit": {
|
|
2559
|
-
"context":
|
|
2731
|
+
"context": 131072,
|
|
2560
2732
|
"output": 16384
|
|
2561
2733
|
},
|
|
2562
2734
|
"cost": {
|
|
2563
|
-
"input": 0.
|
|
2564
|
-
"output": 0.
|
|
2565
|
-
"cache_read": 0.0028
|
|
2735
|
+
"input": 0.037,
|
|
2736
|
+
"output": 0.17
|
|
2566
2737
|
}
|
|
2567
2738
|
}
|
|
2568
2739
|
}
|