llm.rb 13.1.0 → 15.0.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +716 -1826
- data/README.md +668 -253
- data/bin/llm.rb +156 -56
- data/data/alibaba.json +1999 -0
- data/data/anthropic.json +195 -252
- data/data/bedrock.json +2189 -1854
- data/data/deepinfra.json +1312 -728
- data/data/deepseek.json +42 -39
- data/data/google.json +1024 -404
- data/data/mistral.json +481 -401
- data/data/moonshot.json +384 -0
- data/data/openai.json +985 -1354
- data/data/xai.json +220 -115
- data/data/zai.json +166 -166
- data/docs/deepdive/advanced/cancellation.md +74 -0
- data/docs/deepdive/advanced/compaction.md +85 -0
- data/docs/deepdive/advanced/context.md +280 -0
- data/docs/deepdive/advanced/guard.md +371 -0
- data/docs/deepdive/advanced/transformer.md +67 -0
- data/docs/deepdive/advanced/transports.md +45 -0
- data/docs/deepdive/features/builtin_tools.md +191 -0
- data/docs/deepdive/features/concurrency.md +110 -0
- data/docs/deepdive/features/database.md +449 -0
- data/docs/deepdive/features/embeddings.md +157 -0
- data/docs/deepdive/features/repl.md +141 -0
- data/docs/deepdive/fundamentals/agents.md +253 -0
- data/docs/deepdive/fundamentals/providers.md +159 -0
- data/docs/deepdive/fundamentals/schema.md +61 -0
- data/docs/deepdive/fundamentals/skills.md +111 -0
- data/docs/deepdive/fundamentals/stream.md +143 -0
- data/docs/deepdive/fundamentals/tools.md +329 -0
- data/docs/deepdive/media/audio.md +122 -0
- data/docs/deepdive/media/images.md +89 -0
- data/docs/deepdive/media/ocr.md +48 -0
- data/docs/deepdive/protocols/a2a.md +106 -0
- data/docs/deepdive/protocols/mcp.md +111 -0
- data/docs/deepdive/reference/cost.md +109 -0
- data/docs/deepdive/reference/model_registry.md +271 -0
- data/docs/deepdive/reference/object.md +108 -0
- data/docs/deepdive/reference/tracer.md +187 -0
- data/{resources → docs}/deepdive.md +37 -23
- data/lib/llm/a2a/transport/http.rb +1 -1
- data/lib/llm/active_record/acts_as_llm.rb +19 -5
- data/lib/llm/agent.rb +107 -17
- data/lib/llm/context.rb +164 -134
- data/lib/llm/cost.rb +114 -49
- data/lib/llm/error.rb +7 -8
- data/lib/llm/function/array.rb +1 -1
- data/lib/llm/function/async/task.rb +2 -0
- data/lib/llm/function/fiber/task.rb +2 -0
- data/lib/llm/function/fork/task.rb +16 -1
- data/lib/llm/function/ractor/task.rb +2 -0
- data/lib/llm/function/sequential/group.rb +20 -10
- data/lib/llm/function/sequential/task.rb +2 -9
- data/lib/llm/function/task.rb +4 -0
- data/lib/llm/function/thread/task.rb +2 -0
- data/lib/llm/function.rb +33 -4
- data/lib/llm/guard/loop.rb +89 -0
- data/lib/llm/guard/null.rb +19 -0
- data/lib/llm/guard.rb +61 -0
- data/lib/llm/message.rb +5 -4
- data/lib/llm/provider.rb +43 -0
- data/lib/llm/providers/alibaba/error_handler.rb +34 -0
- data/lib/llm/providers/alibaba/request_adapter.rb +13 -0
- data/lib/llm/providers/alibaba.rb +93 -0
- data/lib/llm/providers/anthropic/stream_parser.rb +1 -0
- data/lib/llm/providers/anthropic.rb +1 -9
- data/lib/llm/providers/bedrock/stream_parser.rb +1 -0
- data/lib/llm/providers/bedrock.rb +9 -9
- data/lib/llm/providers/deepseek/request_adapter.rb +2 -33
- data/lib/llm/providers/google/stream_parser.rb +1 -0
- data/lib/llm/providers/google.rb +1 -9
- data/lib/llm/providers/moonshot.rb +76 -0
- data/lib/llm/providers/ollama.rb +1 -9
- data/lib/llm/providers/openai/responses/stream_parser.rb +1 -0
- data/lib/llm/providers/openai/responses.rb +6 -9
- data/lib/llm/providers/openai/schema.rb +37 -0
- data/lib/llm/providers/openai/stream_parser.rb +1 -0
- data/lib/llm/providers/openai.rb +4 -12
- data/lib/llm/registry/model.rb +186 -0
- data/lib/llm/registry.rb +45 -14
- data/lib/llm/repl/bar.rb +11 -12
- data/lib/llm/repl/buffer.rb +43 -16
- data/lib/llm/repl/color.rb +85 -0
- data/lib/llm/repl/command.rb +12 -0
- data/lib/llm/repl/commands/model.rb +39 -0
- data/lib/llm/repl/input/cache.rb +45 -0
- data/lib/llm/repl/input/char.rb +46 -0
- data/lib/llm/repl/input/row.rb +39 -0
- data/lib/llm/repl/input.rb +327 -79
- data/lib/llm/repl/markdown/table.rb +6 -2
- data/lib/llm/repl/markdown.rb +56 -6
- data/lib/llm/repl/node.rb +7 -0
- data/lib/llm/repl/status.rb +54 -5
- data/lib/llm/repl/stream.rb +16 -4
- data/lib/llm/repl/walker.rb +3 -2
- data/lib/llm/repl/window.rb +111 -11
- data/lib/llm/repl.rb +47 -21
- data/lib/llm/sequel/plugin.rb +19 -5
- data/lib/llm/skill.rb +21 -8
- data/lib/llm/stream.rb +35 -7
- data/lib/llm/tool.rb +27 -0
- data/lib/llm/tools/rg.rb +2 -1
- data/lib/llm/transformer/null.rb +21 -0
- data/lib/llm/transformer.rb +55 -0
- data/lib/llm/transport/curb.rb +23 -3
- data/lib/llm/usage.rb +155 -9
- data/lib/llm/version.rb +1 -1
- data/lib/llm.rb +121 -29
- data/llm.gemspec +16 -10
- metadata +100 -13
- data/lib/llm/loop_guard.rb +0 -107
data/data/deepinfra.json
CHANGED
|
@@ -7,24 +7,22 @@
|
|
|
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",
|
|
27
|
-
"video"
|
|
25
|
+
"text"
|
|
28
26
|
],
|
|
29
27
|
"output": [
|
|
30
28
|
"text"
|
|
@@ -33,31 +31,40 @@
|
|
|
33
31
|
"open_weights": true,
|
|
34
32
|
"limit": {
|
|
35
33
|
"context": 262144,
|
|
36
|
-
"output":
|
|
34
|
+
"output": 64000
|
|
37
35
|
},
|
|
38
36
|
"cost": {
|
|
39
|
-
"input": 0.
|
|
40
|
-
"output": 0.
|
|
37
|
+
"input": 0.14,
|
|
38
|
+
"output": 0.58,
|
|
39
|
+
"cache_read": 0.035
|
|
41
40
|
}
|
|
42
41
|
},
|
|
43
|
-
"
|
|
44
|
-
"id": "
|
|
45
|
-
"name": "
|
|
46
|
-
"description": "
|
|
47
|
-
"family": "
|
|
42
|
+
"XiaomiMiMo/MiMo-V2.5": {
|
|
43
|
+
"id": "XiaomiMiMo/MiMo-V2.5",
|
|
44
|
+
"name": "MiMo-V2.5",
|
|
45
|
+
"description": "Open MiMo model for multimodal coding agents and long-context automation",
|
|
46
|
+
"family": "mimo",
|
|
48
47
|
"attachment": true,
|
|
49
48
|
"reasoning": true,
|
|
50
|
-
"reasoning_options": [
|
|
49
|
+
"reasoning_options": [
|
|
50
|
+
{
|
|
51
|
+
"type": "toggle"
|
|
52
|
+
}
|
|
53
|
+
],
|
|
51
54
|
"tool_call": true,
|
|
55
|
+
"interleaved": {
|
|
56
|
+
"field": "reasoning_content"
|
|
57
|
+
},
|
|
52
58
|
"structured_output": true,
|
|
53
59
|
"temperature": true,
|
|
54
|
-
"knowledge": "
|
|
55
|
-
"release_date": "2026-
|
|
56
|
-
"last_updated": "2026-04-
|
|
60
|
+
"knowledge": "2024-12",
|
|
61
|
+
"release_date": "2026-04-22",
|
|
62
|
+
"last_updated": "2026-04-22",
|
|
57
63
|
"modalities": {
|
|
58
64
|
"input": [
|
|
59
65
|
"text",
|
|
60
66
|
"image",
|
|
67
|
+
"audio",
|
|
61
68
|
"video"
|
|
62
69
|
],
|
|
63
70
|
"output": [
|
|
@@ -67,30 +74,39 @@
|
|
|
67
74
|
"open_weights": true,
|
|
68
75
|
"limit": {
|
|
69
76
|
"context": 262144,
|
|
70
|
-
"output":
|
|
77
|
+
"output": 16384
|
|
71
78
|
},
|
|
72
79
|
"cost": {
|
|
73
|
-
"input": 0.
|
|
74
|
-
"output":
|
|
75
|
-
"cache_read": 0.
|
|
80
|
+
"input": 0.4,
|
|
81
|
+
"output": 2,
|
|
82
|
+
"cache_read": 0.08
|
|
76
83
|
}
|
|
77
84
|
},
|
|
78
|
-
"
|
|
79
|
-
"id": "
|
|
80
|
-
"name": "
|
|
81
|
-
"description": "
|
|
82
|
-
"family": "
|
|
83
|
-
"attachment":
|
|
84
|
-
"reasoning":
|
|
85
|
+
"XiaomiMiMo/MiMo-V2.5-Pro": {
|
|
86
|
+
"id": "XiaomiMiMo/MiMo-V2.5-Pro",
|
|
87
|
+
"name": "MiMo-V2.5-Pro",
|
|
88
|
+
"description": "Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution",
|
|
89
|
+
"family": "mimo",
|
|
90
|
+
"attachment": true,
|
|
91
|
+
"reasoning": true,
|
|
92
|
+
"reasoning_options": [
|
|
93
|
+
{
|
|
94
|
+
"type": "toggle"
|
|
95
|
+
}
|
|
96
|
+
],
|
|
85
97
|
"tool_call": true,
|
|
98
|
+
"interleaved": {
|
|
99
|
+
"field": "reasoning_content"
|
|
100
|
+
},
|
|
86
101
|
"structured_output": true,
|
|
87
102
|
"temperature": true,
|
|
88
|
-
"knowledge": "
|
|
89
|
-
"release_date": "
|
|
90
|
-
"last_updated": "
|
|
103
|
+
"knowledge": "2024-12",
|
|
104
|
+
"release_date": "2026-04-22",
|
|
105
|
+
"last_updated": "2026-04-22",
|
|
91
106
|
"modalities": {
|
|
92
107
|
"input": [
|
|
93
|
-
"text"
|
|
108
|
+
"text",
|
|
109
|
+
"audio"
|
|
94
110
|
],
|
|
95
111
|
"output": [
|
|
96
112
|
"text"
|
|
@@ -98,29 +114,27 @@
|
|
|
98
114
|
},
|
|
99
115
|
"open_weights": true,
|
|
100
116
|
"limit": {
|
|
101
|
-
"context":
|
|
102
|
-
"output":
|
|
117
|
+
"context": 1048576,
|
|
118
|
+
"output": 16384
|
|
103
119
|
},
|
|
104
120
|
"cost": {
|
|
105
|
-
"input":
|
|
106
|
-
"output":
|
|
107
|
-
"cache_read": 0.
|
|
121
|
+
"input": 1,
|
|
122
|
+
"output": 3,
|
|
123
|
+
"cache_read": 0.2
|
|
108
124
|
}
|
|
109
125
|
},
|
|
110
|
-
"
|
|
111
|
-
"id": "
|
|
112
|
-
"name": "
|
|
113
|
-
"description": "
|
|
114
|
-
"family": "
|
|
126
|
+
"MiniMaxAI/MiniMax-M2.7": {
|
|
127
|
+
"id": "MiniMaxAI/MiniMax-M2.7",
|
|
128
|
+
"name": "MiniMax-M2.7",
|
|
129
|
+
"description": "Open MiniMax flagship for coding agents, office automation, and complex environments",
|
|
130
|
+
"family": "minimax",
|
|
115
131
|
"attachment": false,
|
|
116
132
|
"reasoning": true,
|
|
117
133
|
"reasoning_options": [],
|
|
118
134
|
"tool_call": true,
|
|
119
|
-
"structured_output": true,
|
|
120
135
|
"temperature": true,
|
|
121
|
-
"
|
|
122
|
-
"
|
|
123
|
-
"last_updated": "2025-04",
|
|
136
|
+
"release_date": "2026-03-18",
|
|
137
|
+
"last_updated": "2026-03-18",
|
|
124
138
|
"modalities": {
|
|
125
139
|
"input": [
|
|
126
140
|
"text"
|
|
@@ -131,28 +145,28 @@
|
|
|
131
145
|
},
|
|
132
146
|
"open_weights": true,
|
|
133
147
|
"limit": {
|
|
134
|
-
"context":
|
|
135
|
-
"output":
|
|
148
|
+
"context": 196608,
|
|
149
|
+
"output": 131072
|
|
136
150
|
},
|
|
137
151
|
"cost": {
|
|
138
|
-
"input": 0.
|
|
139
|
-
"output":
|
|
152
|
+
"input": 0.25,
|
|
153
|
+
"output": 1,
|
|
154
|
+
"cache_read": 0.05
|
|
140
155
|
}
|
|
141
156
|
},
|
|
142
|
-
"
|
|
143
|
-
"id": "
|
|
144
|
-
"name": "
|
|
145
|
-
"description": "
|
|
146
|
-
"family": "
|
|
157
|
+
"MiniMaxAI/MiniMax-M3": {
|
|
158
|
+
"id": "MiniMaxAI/MiniMax-M3",
|
|
159
|
+
"name": "MiniMax-M3",
|
|
160
|
+
"description": "MiniMax multimodal model for long-context coding, perception, and agent planning",
|
|
161
|
+
"family": "minimax",
|
|
147
162
|
"attachment": true,
|
|
148
163
|
"reasoning": true,
|
|
149
164
|
"reasoning_options": [],
|
|
150
165
|
"tool_call": true,
|
|
151
166
|
"structured_output": true,
|
|
152
167
|
"temperature": true,
|
|
153
|
-
"
|
|
154
|
-
"
|
|
155
|
-
"last_updated": "2026-04-20",
|
|
168
|
+
"release_date": "2026-06-01",
|
|
169
|
+
"last_updated": "2026-06-01",
|
|
156
170
|
"modalities": {
|
|
157
171
|
"input": [
|
|
158
172
|
"text",
|
|
@@ -165,34 +179,34 @@
|
|
|
165
179
|
},
|
|
166
180
|
"open_weights": true,
|
|
167
181
|
"limit": {
|
|
168
|
-
"context":
|
|
169
|
-
"output":
|
|
182
|
+
"context": 524288,
|
|
183
|
+
"output": 128000
|
|
170
184
|
},
|
|
171
185
|
"cost": {
|
|
172
|
-
"input": 0.
|
|
173
|
-
"output":
|
|
174
|
-
"cache_read": 0.
|
|
186
|
+
"input": 0.28,
|
|
187
|
+
"output": 1.1,
|
|
188
|
+
"cache_read": 0.056
|
|
175
189
|
}
|
|
176
190
|
},
|
|
177
|
-
"
|
|
178
|
-
"id": "
|
|
179
|
-
"name": "
|
|
180
|
-
"description": "
|
|
181
|
-
"family": "
|
|
182
|
-
"attachment":
|
|
191
|
+
"MiniMaxAI/MiniMax-M2.5": {
|
|
192
|
+
"id": "MiniMaxAI/MiniMax-M2.5",
|
|
193
|
+
"name": "MiniMax M2.5",
|
|
194
|
+
"description": "MiniMax model for chat, coding, office work, and agentic tasks",
|
|
195
|
+
"family": "minimax",
|
|
196
|
+
"attachment": false,
|
|
183
197
|
"reasoning": true,
|
|
184
198
|
"reasoning_options": [],
|
|
185
199
|
"tool_call": true,
|
|
186
|
-
"
|
|
200
|
+
"interleaved": {
|
|
201
|
+
"field": "reasoning_content"
|
|
202
|
+
},
|
|
187
203
|
"temperature": true,
|
|
188
|
-
"
|
|
189
|
-
"
|
|
204
|
+
"knowledge": "2025-06",
|
|
205
|
+
"release_date": "2026-02-12",
|
|
206
|
+
"last_updated": "2026-02-12",
|
|
190
207
|
"modalities": {
|
|
191
208
|
"input": [
|
|
192
|
-
"text"
|
|
193
|
-
"image",
|
|
194
|
-
"video",
|
|
195
|
-
"audio"
|
|
209
|
+
"text"
|
|
196
210
|
],
|
|
197
211
|
"output": [
|
|
198
212
|
"text"
|
|
@@ -200,33 +214,67 @@
|
|
|
200
214
|
},
|
|
201
215
|
"open_weights": true,
|
|
202
216
|
"limit": {
|
|
203
|
-
"context":
|
|
204
|
-
"output":
|
|
217
|
+
"context": 196608,
|
|
218
|
+
"output": 131072
|
|
205
219
|
},
|
|
220
|
+
"status": "deprecated",
|
|
206
221
|
"cost": {
|
|
207
|
-
"input": 0.
|
|
208
|
-
"output":
|
|
222
|
+
"input": 0.15,
|
|
223
|
+
"output": 1.15,
|
|
224
|
+
"cache_read": 0.03
|
|
209
225
|
}
|
|
210
226
|
},
|
|
211
|
-
"
|
|
212
|
-
"id": "
|
|
213
|
-
"name": "
|
|
214
|
-
"description": "
|
|
215
|
-
"family": "
|
|
216
|
-
"attachment":
|
|
227
|
+
"nvidia/Llama-3.3-Nemotron-Super-49B-v1.5": {
|
|
228
|
+
"id": "nvidia/Llama-3.3-Nemotron-Super-49B-v1.5",
|
|
229
|
+
"name": "Llama 3.3 Nemotron Super 49B v1.5",
|
|
230
|
+
"description": "Nemotron model for efficient reasoning, coding, and specialized AI agents",
|
|
231
|
+
"family": "nemotron",
|
|
232
|
+
"attachment": false,
|
|
217
233
|
"reasoning": true,
|
|
218
234
|
"reasoning_options": [],
|
|
219
235
|
"tool_call": true,
|
|
220
236
|
"structured_output": true,
|
|
221
237
|
"temperature": true,
|
|
222
|
-
"release_date": "
|
|
223
|
-
"last_updated": "
|
|
238
|
+
"release_date": "2025-07-25",
|
|
239
|
+
"last_updated": "2025-07-25",
|
|
224
240
|
"modalities": {
|
|
225
241
|
"input": [
|
|
226
|
-
"text"
|
|
227
|
-
|
|
228
|
-
|
|
229
|
-
"
|
|
242
|
+
"text"
|
|
243
|
+
],
|
|
244
|
+
"output": [
|
|
245
|
+
"text"
|
|
246
|
+
]
|
|
247
|
+
},
|
|
248
|
+
"open_weights": true,
|
|
249
|
+
"limit": {
|
|
250
|
+
"context": 131072,
|
|
251
|
+
"output": 131072
|
|
252
|
+
},
|
|
253
|
+
"status": "deprecated",
|
|
254
|
+
"cost": {
|
|
255
|
+
"input": 0.4,
|
|
256
|
+
"output": 0.4
|
|
257
|
+
}
|
|
258
|
+
},
|
|
259
|
+
"nvidia/Nemotron-3-Nano-30B-A3B": {
|
|
260
|
+
"id": "nvidia/Nemotron-3-Nano-30B-A3B",
|
|
261
|
+
"name": "Nemotron 3 Nano 30B A3B",
|
|
262
|
+
"description": "Small Nemotron 3 MoE for efficient coding, math, and long-context agents",
|
|
263
|
+
"family": "nemotron",
|
|
264
|
+
"attachment": false,
|
|
265
|
+
"reasoning": true,
|
|
266
|
+
"reasoning_options": [
|
|
267
|
+
{
|
|
268
|
+
"type": "toggle"
|
|
269
|
+
}
|
|
270
|
+
],
|
|
271
|
+
"tool_call": true,
|
|
272
|
+
"temperature": true,
|
|
273
|
+
"release_date": "2025-12-15",
|
|
274
|
+
"last_updated": "2025-12-15",
|
|
275
|
+
"modalities": {
|
|
276
|
+
"input": [
|
|
277
|
+
"text"
|
|
230
278
|
],
|
|
231
279
|
"output": [
|
|
232
280
|
"text"
|
|
@@ -235,26 +283,27 @@
|
|
|
235
283
|
"open_weights": true,
|
|
236
284
|
"limit": {
|
|
237
285
|
"context": 262144,
|
|
238
|
-
"output":
|
|
286
|
+
"output": 262144
|
|
239
287
|
},
|
|
240
288
|
"cost": {
|
|
241
|
-
"input": 0.
|
|
242
|
-
"output":
|
|
289
|
+
"input": 0.05,
|
|
290
|
+
"output": 0.2,
|
|
291
|
+
"cache_read": 0.025
|
|
243
292
|
}
|
|
244
293
|
},
|
|
245
|
-
"
|
|
246
|
-
"id": "
|
|
247
|
-
"name": "
|
|
248
|
-
"description": "
|
|
249
|
-
"family": "
|
|
294
|
+
"nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning": {
|
|
295
|
+
"id": "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning",
|
|
296
|
+
"name": "Nemotron 3 Nano Omni 30B A3B Reasoning",
|
|
297
|
+
"description": "Open Nemotron omni model combining reasoning with text, vision, and audio",
|
|
298
|
+
"family": "nemotron",
|
|
250
299
|
"attachment": true,
|
|
251
300
|
"reasoning": true,
|
|
252
301
|
"reasoning_options": [],
|
|
253
302
|
"tool_call": true,
|
|
254
303
|
"structured_output": true,
|
|
255
304
|
"temperature": true,
|
|
256
|
-
"release_date": "2026-
|
|
257
|
-
"last_updated": "2026-
|
|
305
|
+
"release_date": "2026-04-28",
|
|
306
|
+
"last_updated": "2026-04-28",
|
|
258
307
|
"modalities": {
|
|
259
308
|
"input": [
|
|
260
309
|
"text",
|
|
@@ -271,27 +320,107 @@
|
|
|
271
320
|
"context": 262144,
|
|
272
321
|
"output": 65536
|
|
273
322
|
},
|
|
323
|
+
"status": "deprecated",
|
|
274
324
|
"cost": {
|
|
275
|
-
"input": 0.
|
|
276
|
-
"output":
|
|
325
|
+
"input": 0.2,
|
|
326
|
+
"output": 0.8
|
|
277
327
|
}
|
|
278
328
|
},
|
|
279
|
-
"
|
|
280
|
-
"id": "
|
|
281
|
-
"name": "
|
|
282
|
-
"description": "
|
|
283
|
-
"family": "
|
|
284
|
-
"attachment":
|
|
285
|
-
"reasoning":
|
|
329
|
+
"google/gemma-4-26B-A4B-it": {
|
|
330
|
+
"id": "google/gemma-4-26B-A4B-it",
|
|
331
|
+
"name": "Gemma 4 26B A4B IT",
|
|
332
|
+
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
|
|
333
|
+
"family": "gemma",
|
|
334
|
+
"attachment": true,
|
|
335
|
+
"reasoning": true,
|
|
336
|
+
"reasoning_options": [
|
|
337
|
+
{
|
|
338
|
+
"type": "toggle"
|
|
339
|
+
}
|
|
340
|
+
],
|
|
286
341
|
"tool_call": true,
|
|
287
342
|
"structured_output": true,
|
|
288
343
|
"temperature": true,
|
|
289
|
-
"
|
|
290
|
-
"
|
|
291
|
-
"
|
|
344
|
+
"release_date": "2026-04-02",
|
|
345
|
+
"last_updated": "2026-04-02",
|
|
346
|
+
"modalities": {
|
|
347
|
+
"input": [
|
|
348
|
+
"text",
|
|
349
|
+
"image"
|
|
350
|
+
],
|
|
351
|
+
"output": [
|
|
352
|
+
"text"
|
|
353
|
+
]
|
|
354
|
+
},
|
|
355
|
+
"open_weights": true,
|
|
356
|
+
"limit": {
|
|
357
|
+
"context": 262144,
|
|
358
|
+
"output": 32768
|
|
359
|
+
},
|
|
360
|
+
"cost": {
|
|
361
|
+
"input": 0.07,
|
|
362
|
+
"output": 0.34
|
|
363
|
+
}
|
|
364
|
+
},
|
|
365
|
+
"google/gemma-4-E4B-it": {
|
|
366
|
+
"id": "google/gemma-4-E4B-it",
|
|
367
|
+
"name": "Gemma 4 E4B IT",
|
|
368
|
+
"description": "Open Gemma instruction model for efficient chat and self-hosted deployments",
|
|
369
|
+
"family": "gemma",
|
|
370
|
+
"attachment": true,
|
|
371
|
+
"reasoning": true,
|
|
372
|
+
"reasoning_options": [
|
|
373
|
+
{
|
|
374
|
+
"type": "toggle"
|
|
375
|
+
}
|
|
376
|
+
],
|
|
377
|
+
"tool_call": true,
|
|
378
|
+
"structured_output": true,
|
|
379
|
+
"temperature": true,
|
|
380
|
+
"release_date": "2026-04-02",
|
|
381
|
+
"last_updated": "2026-04-02",
|
|
292
382
|
"modalities": {
|
|
293
383
|
"input": [
|
|
384
|
+
"text",
|
|
385
|
+
"image",
|
|
386
|
+
"audio"
|
|
387
|
+
],
|
|
388
|
+
"output": [
|
|
294
389
|
"text"
|
|
390
|
+
]
|
|
391
|
+
},
|
|
392
|
+
"open_weights": true,
|
|
393
|
+
"limit": {
|
|
394
|
+
"context": 131072,
|
|
395
|
+
"output": 8192
|
|
396
|
+
},
|
|
397
|
+
"cost": {
|
|
398
|
+
"input": 0.02,
|
|
399
|
+
"output": 0.1
|
|
400
|
+
}
|
|
401
|
+
},
|
|
402
|
+
"google/gemma-4-31B-it": {
|
|
403
|
+
"id": "google/gemma-4-31B-it",
|
|
404
|
+
"name": "Gemma 4 31B IT",
|
|
405
|
+
"description": "Largest Gemma 4 instruction model for open, self-hosted chat and reasoning",
|
|
406
|
+
"family": "gemma",
|
|
407
|
+
"attachment": true,
|
|
408
|
+
"reasoning": true,
|
|
409
|
+
"reasoning_options": [
|
|
410
|
+
{
|
|
411
|
+
"type": "toggle"
|
|
412
|
+
}
|
|
413
|
+
],
|
|
414
|
+
"tool_call": true,
|
|
415
|
+
"structured_output": true,
|
|
416
|
+
"temperature": true,
|
|
417
|
+
"release_date": "2026-04-02",
|
|
418
|
+
"last_updated": "2026-04-02",
|
|
419
|
+
"modalities": {
|
|
420
|
+
"input": [
|
|
421
|
+
"text",
|
|
422
|
+
"image",
|
|
423
|
+
"video"
|
|
295
424
|
],
|
|
296
425
|
"output": [
|
|
297
426
|
"text"
|
|
@@ -303,22 +432,577 @@
|
|
|
303
432
|
"output": 32768
|
|
304
433
|
},
|
|
305
434
|
"cost": {
|
|
306
|
-
"input": 0.
|
|
307
|
-
"output":
|
|
435
|
+
"input": 0.13,
|
|
436
|
+
"output": 0.38
|
|
308
437
|
}
|
|
309
438
|
},
|
|
310
|
-
"
|
|
311
|
-
"id": "
|
|
312
|
-
"name": "
|
|
313
|
-
"description": "
|
|
314
|
-
"family": "
|
|
439
|
+
"ByteDance/Seed-2.0-code": {
|
|
440
|
+
"id": "ByteDance/Seed-2.0-code",
|
|
441
|
+
"name": "Seed 2.0 Code",
|
|
442
|
+
"description": "ByteDance Seed coding model for multimodal software engineering and long-running agents",
|
|
443
|
+
"family": "seed",
|
|
444
|
+
"attachment": true,
|
|
445
|
+
"reasoning": true,
|
|
446
|
+
"reasoning_options": [
|
|
447
|
+
{
|
|
448
|
+
"type": "effort",
|
|
449
|
+
"values": [
|
|
450
|
+
"none",
|
|
451
|
+
"low",
|
|
452
|
+
"medium",
|
|
453
|
+
"high"
|
|
454
|
+
]
|
|
455
|
+
}
|
|
456
|
+
],
|
|
457
|
+
"tool_call": true,
|
|
458
|
+
"structured_output": true,
|
|
459
|
+
"temperature": true,
|
|
460
|
+
"release_date": "2026-02-14",
|
|
461
|
+
"last_updated": "2026-02-14",
|
|
462
|
+
"modalities": {
|
|
463
|
+
"input": [
|
|
464
|
+
"text",
|
|
465
|
+
"image"
|
|
466
|
+
],
|
|
467
|
+
"output": [
|
|
468
|
+
"text"
|
|
469
|
+
]
|
|
470
|
+
},
|
|
471
|
+
"open_weights": false,
|
|
472
|
+
"limit": {
|
|
473
|
+
"context": 256000,
|
|
474
|
+
"output": 131072
|
|
475
|
+
},
|
|
476
|
+
"cost": {
|
|
477
|
+
"input": 0.5,
|
|
478
|
+
"output": 3,
|
|
479
|
+
"cache_read": 0.1,
|
|
480
|
+
"tiers": [
|
|
481
|
+
{
|
|
482
|
+
"input": 1,
|
|
483
|
+
"output": 6,
|
|
484
|
+
"cache_read": 0.2,
|
|
485
|
+
"tier": {
|
|
486
|
+
"type": "context",
|
|
487
|
+
"size": 128000
|
|
488
|
+
}
|
|
489
|
+
}
|
|
490
|
+
]
|
|
491
|
+
}
|
|
492
|
+
},
|
|
493
|
+
"ByteDance/Seed-2.0-pro": {
|
|
494
|
+
"id": "ByteDance/Seed-2.0-pro",
|
|
495
|
+
"name": "Seed 2.0 Pro",
|
|
496
|
+
"description": "Flagship ByteDance Seed 2.0 model for complex multimodal reasoning and long-horizon agent workflows",
|
|
497
|
+
"family": "seed",
|
|
498
|
+
"attachment": true,
|
|
499
|
+
"reasoning": true,
|
|
500
|
+
"reasoning_options": [],
|
|
501
|
+
"tool_call": true,
|
|
502
|
+
"structured_output": true,
|
|
503
|
+
"temperature": true,
|
|
504
|
+
"release_date": "2026-02-14",
|
|
505
|
+
"last_updated": "2026-02-14",
|
|
506
|
+
"modalities": {
|
|
507
|
+
"input": [
|
|
508
|
+
"text",
|
|
509
|
+
"image"
|
|
510
|
+
],
|
|
511
|
+
"output": [
|
|
512
|
+
"text"
|
|
513
|
+
]
|
|
514
|
+
},
|
|
515
|
+
"open_weights": false,
|
|
516
|
+
"limit": {
|
|
517
|
+
"context": 256000,
|
|
518
|
+
"output": 128000
|
|
519
|
+
},
|
|
520
|
+
"cost": {
|
|
521
|
+
"input": 0.5,
|
|
522
|
+
"output": 3,
|
|
523
|
+
"cache_read": 0.1,
|
|
524
|
+
"tiers": [
|
|
525
|
+
{
|
|
526
|
+
"input": 1,
|
|
527
|
+
"output": 6,
|
|
528
|
+
"cache_read": 0.2,
|
|
529
|
+
"tier": {
|
|
530
|
+
"type": "context",
|
|
531
|
+
"size": 128000
|
|
532
|
+
}
|
|
533
|
+
}
|
|
534
|
+
]
|
|
535
|
+
}
|
|
536
|
+
},
|
|
537
|
+
"ByteDance/Seed-2.0-mini": {
|
|
538
|
+
"id": "ByteDance/Seed-2.0-mini",
|
|
539
|
+
"name": "Seed 2.0 Mini",
|
|
540
|
+
"description": "Lightweight ByteDance Seed 2.0 model for low-latency multimodal reasoning and high-volume tasks",
|
|
541
|
+
"family": "seed",
|
|
542
|
+
"attachment": true,
|
|
543
|
+
"reasoning": true,
|
|
544
|
+
"reasoning_options": [],
|
|
545
|
+
"tool_call": true,
|
|
546
|
+
"structured_output": true,
|
|
547
|
+
"temperature": true,
|
|
548
|
+
"release_date": "2026-02-14",
|
|
549
|
+
"last_updated": "2026-02-14",
|
|
550
|
+
"modalities": {
|
|
551
|
+
"input": [
|
|
552
|
+
"text",
|
|
553
|
+
"image"
|
|
554
|
+
],
|
|
555
|
+
"output": [
|
|
556
|
+
"text"
|
|
557
|
+
]
|
|
558
|
+
},
|
|
559
|
+
"open_weights": false,
|
|
560
|
+
"limit": {
|
|
561
|
+
"context": 256000,
|
|
562
|
+
"output": 32000
|
|
563
|
+
},
|
|
564
|
+
"cost": {
|
|
565
|
+
"input": 0.1,
|
|
566
|
+
"output": 0.4,
|
|
567
|
+
"cache_read": 0.02,
|
|
568
|
+
"tiers": [
|
|
569
|
+
{
|
|
570
|
+
"input": 0.2,
|
|
571
|
+
"output": 0.8,
|
|
572
|
+
"cache_read": 0.2,
|
|
573
|
+
"tier": {
|
|
574
|
+
"type": "context",
|
|
575
|
+
"size": 128000
|
|
576
|
+
}
|
|
577
|
+
}
|
|
578
|
+
]
|
|
579
|
+
}
|
|
580
|
+
},
|
|
581
|
+
"moonshotai/Kimi-K2.5": {
|
|
582
|
+
"id": "moonshotai/Kimi-K2.5",
|
|
583
|
+
"name": "Kimi K2.5",
|
|
584
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
585
|
+
"family": "kimi-k2",
|
|
586
|
+
"attachment": true,
|
|
587
|
+
"reasoning": true,
|
|
588
|
+
"reasoning_options": [
|
|
589
|
+
{
|
|
590
|
+
"type": "toggle"
|
|
591
|
+
}
|
|
592
|
+
],
|
|
593
|
+
"tool_call": true,
|
|
594
|
+
"interleaved": {
|
|
595
|
+
"field": "reasoning_content"
|
|
596
|
+
},
|
|
597
|
+
"structured_output": true,
|
|
598
|
+
"temperature": true,
|
|
599
|
+
"knowledge": "2025-01",
|
|
600
|
+
"release_date": "2026-01-27",
|
|
601
|
+
"last_updated": "2026-01-27",
|
|
602
|
+
"modalities": {
|
|
603
|
+
"input": [
|
|
604
|
+
"text",
|
|
605
|
+
"image",
|
|
606
|
+
"video"
|
|
607
|
+
],
|
|
608
|
+
"output": [
|
|
609
|
+
"text"
|
|
610
|
+
]
|
|
611
|
+
},
|
|
612
|
+
"open_weights": true,
|
|
613
|
+
"limit": {
|
|
614
|
+
"context": 262144,
|
|
615
|
+
"output": 32768
|
|
616
|
+
},
|
|
617
|
+
"cost": {
|
|
618
|
+
"input": 0.45,
|
|
619
|
+
"output": 2.25,
|
|
620
|
+
"cache_read": 0.07
|
|
621
|
+
}
|
|
622
|
+
},
|
|
623
|
+
"moonshotai/Kimi-K2.7-Code": {
|
|
624
|
+
"id": "moonshotai/Kimi-K2.7-Code",
|
|
625
|
+
"name": "Kimi K2.7 Code",
|
|
626
|
+
"description": "Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking",
|
|
627
|
+
"family": "kimi-k2",
|
|
628
|
+
"attachment": true,
|
|
629
|
+
"reasoning": true,
|
|
630
|
+
"reasoning_options": [
|
|
631
|
+
{
|
|
632
|
+
"type": "toggle"
|
|
633
|
+
}
|
|
634
|
+
],
|
|
635
|
+
"tool_call": true,
|
|
636
|
+
"interleaved": {
|
|
637
|
+
"field": "reasoning_content"
|
|
638
|
+
},
|
|
639
|
+
"structured_output": true,
|
|
640
|
+
"temperature": false,
|
|
641
|
+
"knowledge": "2025-01",
|
|
642
|
+
"release_date": "2026-06-12",
|
|
643
|
+
"last_updated": "2026-06-12",
|
|
644
|
+
"modalities": {
|
|
645
|
+
"input": [
|
|
646
|
+
"text",
|
|
647
|
+
"image",
|
|
648
|
+
"video"
|
|
649
|
+
],
|
|
650
|
+
"output": [
|
|
651
|
+
"text"
|
|
652
|
+
]
|
|
653
|
+
},
|
|
654
|
+
"open_weights": true,
|
|
655
|
+
"limit": {
|
|
656
|
+
"context": 262144,
|
|
657
|
+
"output": 262144
|
|
658
|
+
},
|
|
659
|
+
"cost": {
|
|
660
|
+
"input": 0.68,
|
|
661
|
+
"output": 3.4,
|
|
662
|
+
"cache_read": 0.136
|
|
663
|
+
}
|
|
664
|
+
},
|
|
665
|
+
"moonshotai/Kimi-K3": {
|
|
666
|
+
"id": "moonshotai/Kimi-K3",
|
|
667
|
+
"name": "Kimi K3",
|
|
668
|
+
"description": "Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work",
|
|
669
|
+
"family": "kimi-k3",
|
|
670
|
+
"attachment": true,
|
|
671
|
+
"reasoning": true,
|
|
672
|
+
"reasoning_options": [
|
|
673
|
+
{
|
|
674
|
+
"type": "effort",
|
|
675
|
+
"values": [
|
|
676
|
+
"low",
|
|
677
|
+
"high",
|
|
678
|
+
"max"
|
|
679
|
+
]
|
|
680
|
+
}
|
|
681
|
+
],
|
|
682
|
+
"tool_call": true,
|
|
683
|
+
"structured_output": true,
|
|
684
|
+
"temperature": false,
|
|
685
|
+
"release_date": "2026-07-16",
|
|
686
|
+
"last_updated": "2026-07-16",
|
|
687
|
+
"modalities": {
|
|
688
|
+
"input": [
|
|
689
|
+
"text",
|
|
690
|
+
"image"
|
|
691
|
+
],
|
|
692
|
+
"output": [
|
|
693
|
+
"text"
|
|
694
|
+
]
|
|
695
|
+
},
|
|
696
|
+
"open_weights": true,
|
|
697
|
+
"limit": {
|
|
698
|
+
"context": 1048576,
|
|
699
|
+
"output": 131072
|
|
700
|
+
},
|
|
701
|
+
"cost": {
|
|
702
|
+
"input": 2.85,
|
|
703
|
+
"output": 14.25,
|
|
704
|
+
"cache_read": 0.285
|
|
705
|
+
}
|
|
706
|
+
},
|
|
707
|
+
"moonshotai/Kimi-K2.6": {
|
|
708
|
+
"id": "moonshotai/Kimi-K2.6",
|
|
709
|
+
"name": "Kimi K2.6",
|
|
710
|
+
"description": "Kimi multimodal agent model for visual understanding, coding, and planning",
|
|
711
|
+
"family": "kimi-k2",
|
|
712
|
+
"attachment": true,
|
|
713
|
+
"reasoning": true,
|
|
714
|
+
"reasoning_options": [
|
|
715
|
+
{
|
|
716
|
+
"type": "toggle"
|
|
717
|
+
}
|
|
718
|
+
],
|
|
719
|
+
"tool_call": true,
|
|
720
|
+
"interleaved": {
|
|
721
|
+
"field": "reasoning_content"
|
|
722
|
+
},
|
|
723
|
+
"structured_output": true,
|
|
724
|
+
"temperature": true,
|
|
725
|
+
"knowledge": "2024-04",
|
|
726
|
+
"release_date": "2026-04-21",
|
|
727
|
+
"last_updated": "2026-04-21",
|
|
728
|
+
"modalities": {
|
|
729
|
+
"input": [
|
|
730
|
+
"text",
|
|
731
|
+
"image",
|
|
732
|
+
"video"
|
|
733
|
+
],
|
|
734
|
+
"output": [
|
|
735
|
+
"text"
|
|
736
|
+
]
|
|
737
|
+
},
|
|
738
|
+
"open_weights": true,
|
|
739
|
+
"limit": {
|
|
740
|
+
"context": 262144,
|
|
741
|
+
"output": 16384
|
|
742
|
+
},
|
|
743
|
+
"cost": {
|
|
744
|
+
"input": 0.75,
|
|
745
|
+
"output": 3.5,
|
|
746
|
+
"cache_read": 0.15
|
|
747
|
+
}
|
|
748
|
+
},
|
|
749
|
+
"openai/gpt-oss-120b": {
|
|
750
|
+
"id": "openai/gpt-oss-120b",
|
|
751
|
+
"name": "GPT OSS 120B",
|
|
752
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
753
|
+
"family": "gpt-oss",
|
|
754
|
+
"attachment": false,
|
|
755
|
+
"reasoning": true,
|
|
756
|
+
"reasoning_options": [
|
|
757
|
+
{
|
|
758
|
+
"type": "effort",
|
|
759
|
+
"values": [
|
|
760
|
+
"low",
|
|
761
|
+
"medium",
|
|
762
|
+
"high"
|
|
763
|
+
]
|
|
764
|
+
}
|
|
765
|
+
],
|
|
766
|
+
"tool_call": true,
|
|
767
|
+
"structured_output": true,
|
|
768
|
+
"temperature": true,
|
|
769
|
+
"release_date": "2025-08-05",
|
|
770
|
+
"last_updated": "2025-08-05",
|
|
771
|
+
"modalities": {
|
|
772
|
+
"input": [
|
|
773
|
+
"text"
|
|
774
|
+
],
|
|
775
|
+
"output": [
|
|
776
|
+
"text"
|
|
777
|
+
]
|
|
778
|
+
},
|
|
779
|
+
"open_weights": true,
|
|
780
|
+
"limit": {
|
|
781
|
+
"context": 131072,
|
|
782
|
+
"output": 16384
|
|
783
|
+
},
|
|
784
|
+
"cost": {
|
|
785
|
+
"input": 0.037,
|
|
786
|
+
"output": 0.17
|
|
787
|
+
}
|
|
788
|
+
},
|
|
789
|
+
"openai/gpt-oss-20b": {
|
|
790
|
+
"id": "openai/gpt-oss-20b",
|
|
791
|
+
"name": "GPT OSS 20B",
|
|
792
|
+
"description": "Open-weight GPT model for self-hosted reasoning and instruction-following workloads",
|
|
793
|
+
"family": "gpt-oss",
|
|
794
|
+
"attachment": false,
|
|
795
|
+
"reasoning": true,
|
|
796
|
+
"reasoning_options": [
|
|
797
|
+
{
|
|
798
|
+
"type": "effort",
|
|
799
|
+
"values": [
|
|
800
|
+
"low",
|
|
801
|
+
"medium",
|
|
802
|
+
"high"
|
|
803
|
+
]
|
|
804
|
+
}
|
|
805
|
+
],
|
|
806
|
+
"tool_call": true,
|
|
807
|
+
"structured_output": true,
|
|
808
|
+
"temperature": true,
|
|
809
|
+
"release_date": "2025-08-05",
|
|
810
|
+
"last_updated": "2025-08-05",
|
|
811
|
+
"modalities": {
|
|
812
|
+
"input": [
|
|
813
|
+
"text"
|
|
814
|
+
],
|
|
815
|
+
"output": [
|
|
816
|
+
"text"
|
|
817
|
+
]
|
|
818
|
+
},
|
|
819
|
+
"open_weights": true,
|
|
820
|
+
"limit": {
|
|
821
|
+
"context": 131072,
|
|
822
|
+
"output": 16384
|
|
823
|
+
},
|
|
824
|
+
"cost": {
|
|
825
|
+
"input": 0.03,
|
|
826
|
+
"output": 0.14
|
|
827
|
+
}
|
|
828
|
+
},
|
|
829
|
+
"deepseek-ai/DeepSeek-V4-Pro": {
|
|
830
|
+
"id": "deepseek-ai/DeepSeek-V4-Pro",
|
|
831
|
+
"name": "DeepSeek V4 Pro",
|
|
832
|
+
"description": "Open MoE flagship with million-token context for coding and long agent runs",
|
|
833
|
+
"family": "deepseek-thinking",
|
|
834
|
+
"attachment": false,
|
|
835
|
+
"reasoning": true,
|
|
836
|
+
"reasoning_options": [
|
|
837
|
+
{
|
|
838
|
+
"type": "toggle"
|
|
839
|
+
},
|
|
840
|
+
{
|
|
841
|
+
"type": "effort",
|
|
842
|
+
"values": [
|
|
843
|
+
"low",
|
|
844
|
+
"medium",
|
|
845
|
+
"high",
|
|
846
|
+
"xhigh"
|
|
847
|
+
]
|
|
848
|
+
}
|
|
849
|
+
],
|
|
850
|
+
"tool_call": true,
|
|
851
|
+
"interleaved": {
|
|
852
|
+
"field": "reasoning_content"
|
|
853
|
+
},
|
|
854
|
+
"structured_output": true,
|
|
855
|
+
"temperature": true,
|
|
856
|
+
"knowledge": "2025-05",
|
|
857
|
+
"release_date": "2026-04-24",
|
|
858
|
+
"last_updated": "2026-04-24",
|
|
859
|
+
"modalities": {
|
|
860
|
+
"input": [
|
|
861
|
+
"text"
|
|
862
|
+
],
|
|
863
|
+
"output": [
|
|
864
|
+
"text"
|
|
865
|
+
]
|
|
866
|
+
},
|
|
867
|
+
"open_weights": true,
|
|
868
|
+
"limit": {
|
|
869
|
+
"context": 1048576,
|
|
870
|
+
"output": 16384
|
|
871
|
+
},
|
|
872
|
+
"cost": {
|
|
873
|
+
"input": 1.3,
|
|
874
|
+
"output": 2.6,
|
|
875
|
+
"cache_read": 0.1
|
|
876
|
+
}
|
|
877
|
+
},
|
|
878
|
+
"deepseek-ai/DeepSeek-V4-Flash-0731": {
|
|
879
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash-0731",
|
|
880
|
+
"name": "DeepSeek V4 Flash 0731",
|
|
881
|
+
"description": "Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding",
|
|
882
|
+
"family": "deepseek-flash",
|
|
883
|
+
"attachment": false,
|
|
884
|
+
"reasoning": true,
|
|
885
|
+
"reasoning_options": [
|
|
886
|
+
{
|
|
887
|
+
"type": "toggle"
|
|
888
|
+
}
|
|
889
|
+
],
|
|
890
|
+
"tool_call": true,
|
|
891
|
+
"structured_output": true,
|
|
892
|
+
"temperature": true,
|
|
893
|
+
"knowledge": "2025-05",
|
|
894
|
+
"release_date": "2026-07-31",
|
|
895
|
+
"last_updated": "2026-07-31",
|
|
896
|
+
"modalities": {
|
|
897
|
+
"input": [
|
|
898
|
+
"text"
|
|
899
|
+
],
|
|
900
|
+
"output": [
|
|
901
|
+
"text"
|
|
902
|
+
]
|
|
903
|
+
},
|
|
904
|
+
"open_weights": true,
|
|
905
|
+
"limit": {
|
|
906
|
+
"context": 1048576,
|
|
907
|
+
"output": 384000
|
|
908
|
+
},
|
|
909
|
+
"cost": {
|
|
910
|
+
"input": 0.08,
|
|
911
|
+
"output": 0.18,
|
|
912
|
+
"cache_read": 0.016
|
|
913
|
+
}
|
|
914
|
+
},
|
|
915
|
+
"deepseek-ai/DeepSeek-V3.2": {
|
|
916
|
+
"id": "deepseek-ai/DeepSeek-V3.2",
|
|
917
|
+
"name": "DeepSeek-V3.2",
|
|
918
|
+
"description": "DeepSeek chat model for instruction following, coding, and analysis",
|
|
919
|
+
"attachment": false,
|
|
920
|
+
"reasoning": true,
|
|
921
|
+
"reasoning_options": [
|
|
922
|
+
{
|
|
923
|
+
"type": "toggle"
|
|
924
|
+
}
|
|
925
|
+
],
|
|
926
|
+
"tool_call": true,
|
|
927
|
+
"interleaved": {
|
|
928
|
+
"field": "reasoning_content"
|
|
929
|
+
},
|
|
930
|
+
"structured_output": true,
|
|
931
|
+
"temperature": true,
|
|
932
|
+
"knowledge": "2024-12",
|
|
933
|
+
"release_date": "2025-12-02",
|
|
934
|
+
"last_updated": "2025-12-02",
|
|
935
|
+
"modalities": {
|
|
936
|
+
"input": [
|
|
937
|
+
"text"
|
|
938
|
+
],
|
|
939
|
+
"output": [
|
|
940
|
+
"text"
|
|
941
|
+
]
|
|
942
|
+
},
|
|
943
|
+
"open_weights": false,
|
|
944
|
+
"limit": {
|
|
945
|
+
"context": 163840,
|
|
946
|
+
"output": 64000
|
|
947
|
+
},
|
|
948
|
+
"cost": {
|
|
949
|
+
"input": 0.26,
|
|
950
|
+
"output": 0.38,
|
|
951
|
+
"cache_read": 0.13
|
|
952
|
+
}
|
|
953
|
+
},
|
|
954
|
+
"deepseek-ai/DeepSeek-R1-0528": {
|
|
955
|
+
"id": "deepseek-ai/DeepSeek-R1-0528",
|
|
956
|
+
"name": "DeepSeek-R1-0528",
|
|
957
|
+
"description": "DeepSeek reasoning model for multi-step analysis, math, coding, and tools",
|
|
958
|
+
"attachment": false,
|
|
959
|
+
"reasoning": true,
|
|
960
|
+
"reasoning_options": [],
|
|
961
|
+
"tool_call": true,
|
|
962
|
+
"interleaved": {
|
|
963
|
+
"field": "reasoning_content"
|
|
964
|
+
},
|
|
965
|
+
"structured_output": true,
|
|
966
|
+
"temperature": true,
|
|
967
|
+
"knowledge": "2024-07",
|
|
968
|
+
"release_date": "2025-05-28",
|
|
969
|
+
"last_updated": "2025-05-28",
|
|
970
|
+
"modalities": {
|
|
971
|
+
"input": [
|
|
972
|
+
"text"
|
|
973
|
+
],
|
|
974
|
+
"output": [
|
|
975
|
+
"text"
|
|
976
|
+
]
|
|
977
|
+
},
|
|
978
|
+
"open_weights": false,
|
|
979
|
+
"limit": {
|
|
980
|
+
"context": 163840,
|
|
981
|
+
"output": 64000
|
|
982
|
+
},
|
|
983
|
+
"cost": {
|
|
984
|
+
"input": 0.5,
|
|
985
|
+
"output": 2.15,
|
|
986
|
+
"cache_read": 0.35
|
|
987
|
+
}
|
|
988
|
+
},
|
|
989
|
+
"deepseek-ai/DeepSeek-V3.1": {
|
|
990
|
+
"id": "deepseek-ai/DeepSeek-V3.1",
|
|
991
|
+
"name": "DeepSeek-V3.1",
|
|
992
|
+
"description": "Hybrid-reasoning DeepSeek model with thinking and non-thinking modes",
|
|
993
|
+
"family": "deepseek",
|
|
315
994
|
"attachment": false,
|
|
316
|
-
"reasoning":
|
|
995
|
+
"reasoning": true,
|
|
996
|
+
"reasoning_options": [
|
|
997
|
+
{
|
|
998
|
+
"type": "toggle"
|
|
999
|
+
}
|
|
1000
|
+
],
|
|
317
1001
|
"tool_call": true,
|
|
318
1002
|
"structured_output": true,
|
|
319
1003
|
"temperature": true,
|
|
320
|
-
"release_date": "
|
|
321
|
-
"last_updated": "
|
|
1004
|
+
"release_date": "2025-08-21",
|
|
1005
|
+
"last_updated": "2025-08-21",
|
|
322
1006
|
"modalities": {
|
|
323
1007
|
"input": [
|
|
324
1008
|
"text"
|
|
@@ -327,50 +1011,29 @@
|
|
|
327
1011
|
"text"
|
|
328
1012
|
]
|
|
329
1013
|
},
|
|
330
|
-
"open_weights":
|
|
1014
|
+
"open_weights": true,
|
|
331
1015
|
"limit": {
|
|
332
|
-
"context":
|
|
333
|
-
"output":
|
|
1016
|
+
"context": 163840,
|
|
1017
|
+
"output": 8192
|
|
334
1018
|
},
|
|
335
1019
|
"cost": {
|
|
336
|
-
"input":
|
|
337
|
-
"output":
|
|
338
|
-
"cache_read": 0.
|
|
339
|
-
"tiers": [
|
|
340
|
-
{
|
|
341
|
-
"input": 5,
|
|
342
|
-
"output": 15,
|
|
343
|
-
"cache_read": 1,
|
|
344
|
-
"tier": {
|
|
345
|
-
"type": "context",
|
|
346
|
-
"size": 32000
|
|
347
|
-
}
|
|
348
|
-
},
|
|
349
|
-
{
|
|
350
|
-
"input": 6.25,
|
|
351
|
-
"output": 18.5,
|
|
352
|
-
"cache_read": 1.25,
|
|
353
|
-
"tier": {
|
|
354
|
-
"type": "context",
|
|
355
|
-
"size": 128000
|
|
356
|
-
}
|
|
357
|
-
}
|
|
358
|
-
]
|
|
1020
|
+
"input": 0.25,
|
|
1021
|
+
"output": 0.95,
|
|
1022
|
+
"cache_read": 0.13
|
|
359
1023
|
}
|
|
360
1024
|
},
|
|
361
|
-
"
|
|
362
|
-
"id": "
|
|
363
|
-
"name": "
|
|
364
|
-
"description": "
|
|
365
|
-
"family": "
|
|
1025
|
+
"deepseek-ai/DeepSeek-V3-0324": {
|
|
1026
|
+
"id": "deepseek-ai/DeepSeek-V3-0324",
|
|
1027
|
+
"name": "DeepSeek V3 0324",
|
|
1028
|
+
"description": "March 2025 checkpoint of DeepSeek-V3 with improved reasoning and coding",
|
|
1029
|
+
"family": "deepseek",
|
|
366
1030
|
"attachment": false,
|
|
367
1031
|
"reasoning": false,
|
|
368
1032
|
"tool_call": true,
|
|
369
1033
|
"structured_output": true,
|
|
370
1034
|
"temperature": true,
|
|
371
|
-
"
|
|
372
|
-
"
|
|
373
|
-
"last_updated": "2025-09-23",
|
|
1035
|
+
"release_date": "2025-03-24",
|
|
1036
|
+
"last_updated": "2025-03-24",
|
|
374
1037
|
"modalities": {
|
|
375
1038
|
"input": [
|
|
376
1039
|
"text"
|
|
@@ -379,55 +1042,32 @@
|
|
|
379
1042
|
"text"
|
|
380
1043
|
]
|
|
381
1044
|
},
|
|
382
|
-
"open_weights":
|
|
1045
|
+
"open_weights": true,
|
|
383
1046
|
"limit": {
|
|
384
|
-
"context":
|
|
385
|
-
"output":
|
|
1047
|
+
"context": 163840,
|
|
1048
|
+
"output": 163840
|
|
386
1049
|
},
|
|
387
1050
|
"cost": {
|
|
388
|
-
"input":
|
|
389
|
-
"output":
|
|
390
|
-
"cache_read": 0.
|
|
391
|
-
"tiers": [
|
|
392
|
-
{
|
|
393
|
-
"input": 2.4,
|
|
394
|
-
"output": 12,
|
|
395
|
-
"cache_read": 0.48,
|
|
396
|
-
"tier": {
|
|
397
|
-
"type": "context",
|
|
398
|
-
"size": 32000
|
|
399
|
-
}
|
|
400
|
-
},
|
|
401
|
-
{
|
|
402
|
-
"input": 3,
|
|
403
|
-
"output": 15,
|
|
404
|
-
"cache_read": 0.6,
|
|
405
|
-
"tier": {
|
|
406
|
-
"type": "context",
|
|
407
|
-
"size": 128000
|
|
408
|
-
}
|
|
409
|
-
}
|
|
410
|
-
]
|
|
1051
|
+
"input": 0.24,
|
|
1052
|
+
"output": 0.9,
|
|
1053
|
+
"cache_read": 0.135
|
|
411
1054
|
}
|
|
412
1055
|
},
|
|
413
|
-
"
|
|
414
|
-
"id": "
|
|
415
|
-
"name": "
|
|
416
|
-
"description": "
|
|
417
|
-
"family": "
|
|
418
|
-
"attachment":
|
|
419
|
-
"reasoning":
|
|
420
|
-
"reasoning_options": [],
|
|
1056
|
+
"deepseek-ai/DeepSeek-V3": {
|
|
1057
|
+
"id": "deepseek-ai/DeepSeek-V3",
|
|
1058
|
+
"name": "DeepSeek-V3",
|
|
1059
|
+
"description": "Open DeepSeek MoE chat model for coding, math, and general reasoning",
|
|
1060
|
+
"family": "deepseek",
|
|
1061
|
+
"attachment": false,
|
|
1062
|
+
"reasoning": false,
|
|
421
1063
|
"tool_call": true,
|
|
422
1064
|
"structured_output": true,
|
|
423
1065
|
"temperature": true,
|
|
424
|
-
"release_date": "
|
|
425
|
-
"last_updated": "
|
|
1066
|
+
"release_date": "2024-12-26",
|
|
1067
|
+
"last_updated": "2024-12-26",
|
|
426
1068
|
"modalities": {
|
|
427
1069
|
"input": [
|
|
428
|
-
"text"
|
|
429
|
-
"image",
|
|
430
|
-
"video"
|
|
1070
|
+
"text"
|
|
431
1071
|
],
|
|
432
1072
|
"output": [
|
|
433
1073
|
"text"
|
|
@@ -435,30 +1075,44 @@
|
|
|
435
1075
|
},
|
|
436
1076
|
"open_weights": true,
|
|
437
1077
|
"limit": {
|
|
438
|
-
"context":
|
|
439
|
-
"output":
|
|
1078
|
+
"context": 163840,
|
|
1079
|
+
"output": 8192
|
|
440
1080
|
},
|
|
441
1081
|
"cost": {
|
|
442
|
-
"input": 0.
|
|
443
|
-
"output": 0.
|
|
1082
|
+
"input": 0.32,
|
|
1083
|
+
"output": 0.89
|
|
444
1084
|
}
|
|
445
1085
|
},
|
|
446
|
-
"
|
|
447
|
-
"id": "
|
|
448
|
-
"name": "
|
|
449
|
-
"description": "
|
|
450
|
-
"family": "
|
|
1086
|
+
"deepseek-ai/DeepSeek-V4-Flash": {
|
|
1087
|
+
"id": "deepseek-ai/DeepSeek-V4-Flash",
|
|
1088
|
+
"name": "DeepSeek V4 Flash",
|
|
1089
|
+
"description": "Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work",
|
|
1090
|
+
"family": "deepseek-flash",
|
|
451
1091
|
"attachment": false,
|
|
452
1092
|
"reasoning": true,
|
|
453
1093
|
"reasoning_options": [
|
|
454
1094
|
{
|
|
455
1095
|
"type": "toggle"
|
|
1096
|
+
},
|
|
1097
|
+
{
|
|
1098
|
+
"type": "effort",
|
|
1099
|
+
"values": [
|
|
1100
|
+
"low",
|
|
1101
|
+
"medium",
|
|
1102
|
+
"high",
|
|
1103
|
+
"xhigh"
|
|
1104
|
+
]
|
|
456
1105
|
}
|
|
457
1106
|
],
|
|
458
1107
|
"tool_call": true,
|
|
1108
|
+
"interleaved": {
|
|
1109
|
+
"field": "reasoning_content"
|
|
1110
|
+
},
|
|
1111
|
+
"structured_output": true,
|
|
459
1112
|
"temperature": true,
|
|
460
|
-
"
|
|
461
|
-
"
|
|
1113
|
+
"knowledge": "2025-05",
|
|
1114
|
+
"release_date": "2026-04-24",
|
|
1115
|
+
"last_updated": "2026-04-24",
|
|
462
1116
|
"modalities": {
|
|
463
1117
|
"input": [
|
|
464
1118
|
"text"
|
|
@@ -469,30 +1123,32 @@
|
|
|
469
1123
|
},
|
|
470
1124
|
"open_weights": true,
|
|
471
1125
|
"limit": {
|
|
472
|
-
"context":
|
|
473
|
-
"output":
|
|
1126
|
+
"context": 1048576,
|
|
1127
|
+
"output": 16384
|
|
474
1128
|
},
|
|
475
1129
|
"cost": {
|
|
476
|
-
"input": 0.
|
|
477
|
-
"output": 0.
|
|
1130
|
+
"input": 0.09,
|
|
1131
|
+
"output": 0.18,
|
|
1132
|
+
"cache_read": 0.018
|
|
478
1133
|
}
|
|
479
1134
|
},
|
|
480
|
-
"
|
|
481
|
-
"id": "
|
|
482
|
-
"name": "
|
|
483
|
-
"description": "
|
|
484
|
-
"
|
|
485
|
-
"attachment": false,
|
|
1135
|
+
"stepfun-ai/Step-3.7-Flash": {
|
|
1136
|
+
"id": "stepfun-ai/Step-3.7-Flash",
|
|
1137
|
+
"name": "Step 3.7 Flash",
|
|
1138
|
+
"description": "Newer StepFun flash model for faster agents, coding, and multimodal prompts",
|
|
1139
|
+
"attachment": true,
|
|
486
1140
|
"reasoning": true,
|
|
487
1141
|
"reasoning_options": [],
|
|
488
1142
|
"tool_call": true,
|
|
489
|
-
"structured_output": true,
|
|
490
1143
|
"temperature": true,
|
|
491
|
-
"
|
|
492
|
-
"
|
|
1144
|
+
"knowledge": "2026-03-01",
|
|
1145
|
+
"release_date": "2026-05-29",
|
|
1146
|
+
"last_updated": "2026-05-29",
|
|
493
1147
|
"modalities": {
|
|
494
1148
|
"input": [
|
|
495
|
-
"text"
|
|
1149
|
+
"text",
|
|
1150
|
+
"image",
|
|
1151
|
+
"video"
|
|
496
1152
|
],
|
|
497
1153
|
"output": [
|
|
498
1154
|
"text"
|
|
@@ -500,34 +1156,30 @@
|
|
|
500
1156
|
},
|
|
501
1157
|
"open_weights": true,
|
|
502
1158
|
"limit": {
|
|
503
|
-
"context":
|
|
504
|
-
"output":
|
|
1159
|
+
"context": 262144,
|
|
1160
|
+
"output": 256000
|
|
505
1161
|
},
|
|
506
|
-
"status": "deprecated",
|
|
507
1162
|
"cost": {
|
|
508
|
-
"input": 0.
|
|
509
|
-
"output":
|
|
1163
|
+
"input": 0.2,
|
|
1164
|
+
"output": 1.15,
|
|
1165
|
+
"cache_read": 0.04
|
|
510
1166
|
}
|
|
511
1167
|
},
|
|
512
|
-
"
|
|
513
|
-
"id": "
|
|
514
|
-
"name": "
|
|
515
|
-
"description": "
|
|
516
|
-
"family": "
|
|
517
|
-
"attachment":
|
|
518
|
-
"reasoning":
|
|
519
|
-
"reasoning_options": [],
|
|
1168
|
+
"Qwen/Qwen3-235B-A22B-Instruct-2507": {
|
|
1169
|
+
"id": "Qwen/Qwen3-235B-A22B-Instruct-2507",
|
|
1170
|
+
"name": "Qwen3 235B-A22B Instruct 2507",
|
|
1171
|
+
"description": "Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use",
|
|
1172
|
+
"family": "qwen",
|
|
1173
|
+
"attachment": false,
|
|
1174
|
+
"reasoning": false,
|
|
520
1175
|
"tool_call": true,
|
|
521
1176
|
"structured_output": true,
|
|
522
1177
|
"temperature": true,
|
|
523
|
-
"release_date": "
|
|
524
|
-
"last_updated": "
|
|
1178
|
+
"release_date": "2025-07-21",
|
|
1179
|
+
"last_updated": "2025-07-21",
|
|
525
1180
|
"modalities": {
|
|
526
1181
|
"input": [
|
|
527
|
-
"text"
|
|
528
|
-
"image",
|
|
529
|
-
"video",
|
|
530
|
-
"audio"
|
|
1182
|
+
"text"
|
|
531
1183
|
],
|
|
532
1184
|
"output": [
|
|
533
1185
|
"text"
|
|
@@ -536,29 +1188,32 @@
|
|
|
536
1188
|
"open_weights": true,
|
|
537
1189
|
"limit": {
|
|
538
1190
|
"context": 262144,
|
|
539
|
-
"output":
|
|
1191
|
+
"output": 16384
|
|
540
1192
|
},
|
|
541
|
-
"status": "deprecated",
|
|
542
1193
|
"cost": {
|
|
543
|
-
"input": 0.
|
|
544
|
-
"output": 0.
|
|
1194
|
+
"input": 0.09,
|
|
1195
|
+
"output": 0.55
|
|
545
1196
|
}
|
|
546
1197
|
},
|
|
547
|
-
"
|
|
548
|
-
"id": "
|
|
549
|
-
"name": "
|
|
550
|
-
"description": "
|
|
551
|
-
"family": "
|
|
1198
|
+
"Qwen/Qwen3.6-27B": {
|
|
1199
|
+
"id": "Qwen/Qwen3.6-27B",
|
|
1200
|
+
"name": "Qwen3.6 27B",
|
|
1201
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1202
|
+
"family": "qwen",
|
|
552
1203
|
"attachment": true,
|
|
553
|
-
"reasoning":
|
|
1204
|
+
"reasoning": true,
|
|
1205
|
+
"reasoning_options": [],
|
|
554
1206
|
"tool_call": true,
|
|
555
1207
|
"structured_output": true,
|
|
556
|
-
"
|
|
557
|
-
"
|
|
1208
|
+
"temperature": true,
|
|
1209
|
+
"release_date": "2026-04-22",
|
|
1210
|
+
"last_updated": "2026-04-22",
|
|
558
1211
|
"modalities": {
|
|
559
1212
|
"input": [
|
|
560
1213
|
"text",
|
|
561
|
-
"image"
|
|
1214
|
+
"image",
|
|
1215
|
+
"video",
|
|
1216
|
+
"audio"
|
|
562
1217
|
],
|
|
563
1218
|
"output": [
|
|
564
1219
|
"text"
|
|
@@ -566,29 +1221,32 @@
|
|
|
566
1221
|
},
|
|
567
1222
|
"open_weights": true,
|
|
568
1223
|
"limit": {
|
|
569
|
-
"context":
|
|
570
|
-
"output":
|
|
1224
|
+
"context": 262144,
|
|
1225
|
+
"output": 65536
|
|
571
1226
|
},
|
|
572
1227
|
"cost": {
|
|
573
|
-
"input": 0.
|
|
574
|
-
"output":
|
|
1228
|
+
"input": 0.32,
|
|
1229
|
+
"output": 3.2
|
|
575
1230
|
}
|
|
576
1231
|
},
|
|
577
|
-
"
|
|
578
|
-
"id": "
|
|
579
|
-
"name": "
|
|
580
|
-
"description": "
|
|
581
|
-
"family": "
|
|
1232
|
+
"Qwen/Qwen3.5-9B": {
|
|
1233
|
+
"id": "Qwen/Qwen3.5-9B",
|
|
1234
|
+
"name": "Qwen3.5 9B",
|
|
1235
|
+
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1236
|
+
"family": "qwen",
|
|
582
1237
|
"attachment": true,
|
|
583
|
-
"reasoning":
|
|
584
|
-
"
|
|
1238
|
+
"reasoning": true,
|
|
1239
|
+
"reasoning_options": [],
|
|
1240
|
+
"tool_call": true,
|
|
585
1241
|
"structured_output": true,
|
|
586
|
-
"
|
|
587
|
-
"
|
|
1242
|
+
"temperature": true,
|
|
1243
|
+
"release_date": "2026-02-23",
|
|
1244
|
+
"last_updated": "2026-02-23",
|
|
588
1245
|
"modalities": {
|
|
589
1246
|
"input": [
|
|
590
1247
|
"text",
|
|
591
|
-
"image"
|
|
1248
|
+
"image",
|
|
1249
|
+
"video"
|
|
592
1250
|
],
|
|
593
1251
|
"output": [
|
|
594
1252
|
"text"
|
|
@@ -596,25 +1254,27 @@
|
|
|
596
1254
|
},
|
|
597
1255
|
"open_weights": true,
|
|
598
1256
|
"limit": {
|
|
599
|
-
"context":
|
|
600
|
-
"output":
|
|
1257
|
+
"context": 262144,
|
|
1258
|
+
"output": 65536
|
|
601
1259
|
},
|
|
602
1260
|
"cost": {
|
|
603
|
-
"input": 0.
|
|
604
|
-
"output": 0.
|
|
1261
|
+
"input": 0.1,
|
|
1262
|
+
"output": 0.15
|
|
605
1263
|
}
|
|
606
1264
|
},
|
|
607
|
-
"
|
|
608
|
-
"id": "
|
|
609
|
-
"name": "
|
|
610
|
-
"description": "
|
|
611
|
-
"family": "
|
|
1265
|
+
"Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo": {
|
|
1266
|
+
"id": "Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo",
|
|
1267
|
+
"name": "Qwen3 Coder 480B A35B Instruct Turbo",
|
|
1268
|
+
"description": "Qwen coding model for software agents, repository edits, and code reasoning",
|
|
1269
|
+
"family": "qwen",
|
|
612
1270
|
"attachment": false,
|
|
613
1271
|
"reasoning": false,
|
|
614
1272
|
"tool_call": true,
|
|
615
1273
|
"structured_output": true,
|
|
616
|
-
"
|
|
617
|
-
"
|
|
1274
|
+
"temperature": true,
|
|
1275
|
+
"knowledge": "2025-04",
|
|
1276
|
+
"release_date": "2025-07-23",
|
|
1277
|
+
"last_updated": "2025-07-23",
|
|
618
1278
|
"modalities": {
|
|
619
1279
|
"input": [
|
|
620
1280
|
"text"
|
|
@@ -625,79 +1285,63 @@
|
|
|
625
1285
|
},
|
|
626
1286
|
"open_weights": true,
|
|
627
1287
|
"limit": {
|
|
628
|
-
"context":
|
|
629
|
-
"output":
|
|
1288
|
+
"context": 262144,
|
|
1289
|
+
"output": 66536
|
|
630
1290
|
},
|
|
631
1291
|
"cost": {
|
|
632
|
-
"input": 0.
|
|
633
|
-
"output":
|
|
1292
|
+
"input": 0.3,
|
|
1293
|
+
"output": 1,
|
|
1294
|
+
"cache_read": 0.1
|
|
634
1295
|
}
|
|
635
1296
|
},
|
|
636
|
-
"
|
|
637
|
-
"id": "
|
|
638
|
-
"name": "
|
|
639
|
-
"description": "
|
|
640
|
-
"
|
|
1297
|
+
"Qwen/Qwen3.5-122B-A10B": {
|
|
1298
|
+
"id": "Qwen/Qwen3.5-122B-A10B",
|
|
1299
|
+
"name": "Qwen3.5 122B-A10B",
|
|
1300
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1301
|
+
"family": "qwen",
|
|
1302
|
+
"attachment": true,
|
|
641
1303
|
"reasoning": true,
|
|
642
1304
|
"reasoning_options": [],
|
|
643
1305
|
"tool_call": true,
|
|
644
|
-
"interleaved": {
|
|
645
|
-
"field": "reasoning_content"
|
|
646
|
-
},
|
|
647
1306
|
"structured_output": true,
|
|
648
1307
|
"temperature": true,
|
|
649
|
-
"
|
|
650
|
-
"
|
|
651
|
-
"last_updated": "2025-05-28",
|
|
1308
|
+
"release_date": "2026-02-23",
|
|
1309
|
+
"last_updated": "2026-02-23",
|
|
652
1310
|
"modalities": {
|
|
653
1311
|
"input": [
|
|
654
|
-
"text"
|
|
1312
|
+
"text",
|
|
1313
|
+
"image",
|
|
1314
|
+
"video",
|
|
1315
|
+
"audio"
|
|
655
1316
|
],
|
|
656
1317
|
"output": [
|
|
657
1318
|
"text"
|
|
658
1319
|
]
|
|
659
1320
|
},
|
|
660
|
-
"open_weights":
|
|
1321
|
+
"open_weights": true,
|
|
661
1322
|
"limit": {
|
|
662
|
-
"context":
|
|
663
|
-
"output":
|
|
1323
|
+
"context": 262144,
|
|
1324
|
+
"output": 65536
|
|
664
1325
|
},
|
|
665
1326
|
"cost": {
|
|
666
|
-
"input": 0.
|
|
667
|
-
"output": 2.
|
|
668
|
-
"cache_read": 0.35
|
|
1327
|
+
"input": 0.29,
|
|
1328
|
+
"output": 2.4
|
|
669
1329
|
}
|
|
670
1330
|
},
|
|
671
|
-
"
|
|
672
|
-
"id": "
|
|
673
|
-
"name": "
|
|
674
|
-
"description": "
|
|
675
|
-
"family": "
|
|
1331
|
+
"Qwen/Qwen3-32B": {
|
|
1332
|
+
"id": "Qwen/Qwen3-32B",
|
|
1333
|
+
"name": "Qwen3 32B",
|
|
1334
|
+
"description": "Dense open Qwen model for self-hosted chat, reasoning, and coding",
|
|
1335
|
+
"family": "qwen",
|
|
676
1336
|
"attachment": false,
|
|
677
1337
|
"reasoning": true,
|
|
678
|
-
"reasoning_options": [
|
|
679
|
-
{
|
|
680
|
-
"type": "toggle"
|
|
681
|
-
},
|
|
682
|
-
{
|
|
683
|
-
"type": "effort",
|
|
684
|
-
"values": [
|
|
685
|
-
"low",
|
|
686
|
-
"medium",
|
|
687
|
-
"high",
|
|
688
|
-
"xhigh"
|
|
689
|
-
]
|
|
690
|
-
}
|
|
691
|
-
],
|
|
1338
|
+
"reasoning_options": [],
|
|
692
1339
|
"tool_call": true,
|
|
693
|
-
"interleaved": {
|
|
694
|
-
"field": "reasoning_content"
|
|
695
|
-
},
|
|
696
1340
|
"structured_output": true,
|
|
697
1341
|
"temperature": true,
|
|
698
|
-
"knowledge": "2025-
|
|
699
|
-
"release_date": "
|
|
700
|
-
"last_updated": "
|
|
1342
|
+
"knowledge": "2025-04",
|
|
1343
|
+
"release_date": "2025-04",
|
|
1344
|
+
"last_updated": "2025-04",
|
|
701
1345
|
"modalities": {
|
|
702
1346
|
"input": [
|
|
703
1347
|
"text"
|
|
@@ -708,84 +1352,60 @@
|
|
|
708
1352
|
},
|
|
709
1353
|
"open_weights": true,
|
|
710
1354
|
"limit": {
|
|
711
|
-
"context":
|
|
1355
|
+
"context": 40960,
|
|
712
1356
|
"output": 16384
|
|
713
1357
|
},
|
|
714
1358
|
"cost": {
|
|
715
|
-
"input":
|
|
716
|
-
"output":
|
|
717
|
-
"cache_read": 0.1
|
|
1359
|
+
"input": 0.08,
|
|
1360
|
+
"output": 0.28
|
|
718
1361
|
}
|
|
719
1362
|
},
|
|
720
|
-
"
|
|
721
|
-
"id": "
|
|
722
|
-
"name": "
|
|
723
|
-
"description": "
|
|
724
|
-
"family": "
|
|
725
|
-
"attachment":
|
|
726
|
-
"reasoning":
|
|
727
|
-
"reasoning_options": [
|
|
728
|
-
{
|
|
729
|
-
"type": "toggle"
|
|
730
|
-
},
|
|
731
|
-
{
|
|
732
|
-
"type": "effort",
|
|
733
|
-
"values": [
|
|
734
|
-
"low",
|
|
735
|
-
"medium",
|
|
736
|
-
"high",
|
|
737
|
-
"xhigh"
|
|
738
|
-
]
|
|
739
|
-
}
|
|
740
|
-
],
|
|
1363
|
+
"Qwen/Qwen3.8-Max": {
|
|
1364
|
+
"id": "Qwen/Qwen3.8-Max",
|
|
1365
|
+
"name": "Qwen3.8 Max",
|
|
1366
|
+
"description": "2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows",
|
|
1367
|
+
"family": "qwen",
|
|
1368
|
+
"attachment": true,
|
|
1369
|
+
"reasoning": false,
|
|
741
1370
|
"tool_call": true,
|
|
742
|
-
"interleaved": {
|
|
743
|
-
"field": "reasoning_content"
|
|
744
|
-
},
|
|
745
1371
|
"structured_output": true,
|
|
746
1372
|
"temperature": true,
|
|
747
|
-
"
|
|
748
|
-
"
|
|
749
|
-
"last_updated": "2026-04-24",
|
|
1373
|
+
"release_date": "2026-08-03",
|
|
1374
|
+
"last_updated": "2026-08-03",
|
|
750
1375
|
"modalities": {
|
|
751
1376
|
"input": [
|
|
752
|
-
"text"
|
|
1377
|
+
"text",
|
|
1378
|
+
"image",
|
|
1379
|
+
"video",
|
|
1380
|
+
"pdf"
|
|
753
1381
|
],
|
|
754
1382
|
"output": [
|
|
755
1383
|
"text"
|
|
756
1384
|
]
|
|
757
1385
|
},
|
|
758
|
-
"open_weights":
|
|
1386
|
+
"open_weights": false,
|
|
759
1387
|
"limit": {
|
|
760
|
-
"context":
|
|
761
|
-
"output":
|
|
1388
|
+
"context": 256000,
|
|
1389
|
+
"output": 131072
|
|
762
1390
|
},
|
|
763
1391
|
"cost": {
|
|
764
|
-
"input":
|
|
765
|
-
"output":
|
|
766
|
-
"cache_read": 0.
|
|
1392
|
+
"input": 1.65,
|
|
1393
|
+
"output": 4.951,
|
|
1394
|
+
"cache_read": 0.206
|
|
767
1395
|
}
|
|
768
1396
|
},
|
|
769
|
-
"
|
|
770
|
-
"id": "
|
|
771
|
-
"name": "
|
|
772
|
-
"description": "
|
|
1397
|
+
"Qwen/Qwen3.7-Max": {
|
|
1398
|
+
"id": "Qwen/Qwen3.7-Max",
|
|
1399
|
+
"name": "Qwen3.7 Max",
|
|
1400
|
+
"description": "Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks",
|
|
1401
|
+
"family": "qwen",
|
|
773
1402
|
"attachment": false,
|
|
774
|
-
"reasoning":
|
|
775
|
-
"reasoning_options": [
|
|
776
|
-
{
|
|
777
|
-
"type": "toggle"
|
|
778
|
-
}
|
|
779
|
-
],
|
|
1403
|
+
"reasoning": false,
|
|
780
1404
|
"tool_call": true,
|
|
781
|
-
"interleaved": {
|
|
782
|
-
"field": "reasoning_content"
|
|
783
|
-
},
|
|
784
1405
|
"structured_output": true,
|
|
785
1406
|
"temperature": true,
|
|
786
|
-
"
|
|
787
|
-
"
|
|
788
|
-
"last_updated": "2025-12-02",
|
|
1407
|
+
"release_date": "2026-05-21",
|
|
1408
|
+
"last_updated": "2026-05-21",
|
|
789
1409
|
"modalities": {
|
|
790
1410
|
"input": [
|
|
791
1411
|
"text"
|
|
@@ -796,34 +1416,54 @@
|
|
|
796
1416
|
},
|
|
797
1417
|
"open_weights": false,
|
|
798
1418
|
"limit": {
|
|
799
|
-
"context":
|
|
800
|
-
"output":
|
|
1419
|
+
"context": 256000,
|
|
1420
|
+
"output": 65536
|
|
801
1421
|
},
|
|
802
1422
|
"cost": {
|
|
803
|
-
"input":
|
|
804
|
-
"output":
|
|
805
|
-
"cache_read": 0.
|
|
1423
|
+
"input": 2.5,
|
|
1424
|
+
"output": 7.5,
|
|
1425
|
+
"cache_read": 0.5,
|
|
1426
|
+
"tiers": [
|
|
1427
|
+
{
|
|
1428
|
+
"input": 5,
|
|
1429
|
+
"output": 15,
|
|
1430
|
+
"cache_read": 1,
|
|
1431
|
+
"tier": {
|
|
1432
|
+
"type": "context",
|
|
1433
|
+
"size": 32000
|
|
1434
|
+
}
|
|
1435
|
+
},
|
|
1436
|
+
{
|
|
1437
|
+
"input": 6.25,
|
|
1438
|
+
"output": 18.5,
|
|
1439
|
+
"cache_read": 1.25,
|
|
1440
|
+
"tier": {
|
|
1441
|
+
"type": "context",
|
|
1442
|
+
"size": 128000
|
|
1443
|
+
}
|
|
1444
|
+
}
|
|
1445
|
+
]
|
|
806
1446
|
}
|
|
807
1447
|
},
|
|
808
|
-
"
|
|
809
|
-
"id": "
|
|
810
|
-
"name": "
|
|
811
|
-
"description": "
|
|
812
|
-
"family": "
|
|
813
|
-
"attachment":
|
|
1448
|
+
"Qwen/Qwen3.5-27B": {
|
|
1449
|
+
"id": "Qwen/Qwen3.5-27B",
|
|
1450
|
+
"name": "Qwen3.5 27B",
|
|
1451
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1452
|
+
"family": "qwen",
|
|
1453
|
+
"attachment": true,
|
|
814
1454
|
"reasoning": true,
|
|
815
1455
|
"reasoning_options": [],
|
|
816
1456
|
"tool_call": true,
|
|
817
|
-
"
|
|
818
|
-
"field": "reasoning_content"
|
|
819
|
-
},
|
|
1457
|
+
"structured_output": true,
|
|
820
1458
|
"temperature": true,
|
|
821
|
-
"
|
|
822
|
-
"
|
|
823
|
-
"last_updated": "2026-02-12",
|
|
1459
|
+
"release_date": "2026-02-23",
|
|
1460
|
+
"last_updated": "2026-02-23",
|
|
824
1461
|
"modalities": {
|
|
825
1462
|
"input": [
|
|
826
|
-
"text"
|
|
1463
|
+
"text",
|
|
1464
|
+
"image",
|
|
1465
|
+
"video",
|
|
1466
|
+
"audio"
|
|
827
1467
|
],
|
|
828
1468
|
"output": [
|
|
829
1469
|
"text"
|
|
@@ -831,31 +1471,33 @@
|
|
|
831
1471
|
},
|
|
832
1472
|
"open_weights": true,
|
|
833
1473
|
"limit": {
|
|
834
|
-
"context":
|
|
835
|
-
"output":
|
|
1474
|
+
"context": 262144,
|
|
1475
|
+
"output": 65536
|
|
836
1476
|
},
|
|
837
|
-
"status": "deprecated",
|
|
838
1477
|
"cost": {
|
|
839
|
-
"input": 0.
|
|
840
|
-
"output":
|
|
841
|
-
"cache_read": 0.03
|
|
1478
|
+
"input": 0.26,
|
|
1479
|
+
"output": 2.6
|
|
842
1480
|
}
|
|
843
1481
|
},
|
|
844
|
-
"
|
|
845
|
-
"id": "
|
|
846
|
-
"name": "
|
|
847
|
-
"description": "
|
|
848
|
-
"family": "
|
|
849
|
-
"attachment":
|
|
1482
|
+
"Qwen/Qwen3.5-35B-A3B": {
|
|
1483
|
+
"id": "Qwen/Qwen3.5-35B-A3B",
|
|
1484
|
+
"name": "Qwen 3.5 35B A3B",
|
|
1485
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1486
|
+
"family": "qwen",
|
|
1487
|
+
"attachment": true,
|
|
850
1488
|
"reasoning": true,
|
|
851
1489
|
"reasoning_options": [],
|
|
852
1490
|
"tool_call": true,
|
|
1491
|
+
"structured_output": true,
|
|
853
1492
|
"temperature": true,
|
|
854
|
-
"
|
|
855
|
-
"
|
|
1493
|
+
"knowledge": "2025-01",
|
|
1494
|
+
"release_date": "2026-02-01",
|
|
1495
|
+
"last_updated": "2026-04-20",
|
|
856
1496
|
"modalities": {
|
|
857
1497
|
"input": [
|
|
858
|
-
"text"
|
|
1498
|
+
"text",
|
|
1499
|
+
"image",
|
|
1500
|
+
"video"
|
|
859
1501
|
],
|
|
860
1502
|
"output": [
|
|
861
1503
|
"text"
|
|
@@ -863,65 +1505,80 @@
|
|
|
863
1505
|
},
|
|
864
1506
|
"open_weights": true,
|
|
865
1507
|
"limit": {
|
|
866
|
-
"context":
|
|
867
|
-
"output":
|
|
1508
|
+
"context": 262144,
|
|
1509
|
+
"output": 81920
|
|
868
1510
|
},
|
|
869
1511
|
"cost": {
|
|
870
|
-
"input": 0.
|
|
1512
|
+
"input": 0.14,
|
|
871
1513
|
"output": 1,
|
|
872
1514
|
"cache_read": 0.05
|
|
873
1515
|
}
|
|
874
1516
|
},
|
|
875
|
-
"
|
|
876
|
-
"id": "
|
|
877
|
-
"name": "
|
|
878
|
-
"description": "
|
|
879
|
-
"family": "
|
|
880
|
-
"attachment":
|
|
881
|
-
"reasoning":
|
|
882
|
-
"reasoning_options": [],
|
|
1517
|
+
"Qwen/Qwen3-Max": {
|
|
1518
|
+
"id": "Qwen/Qwen3-Max",
|
|
1519
|
+
"name": "Qwen3 Max",
|
|
1520
|
+
"description": "Flagship Qwen3 model for coding agents, complex reasoning, and tool use",
|
|
1521
|
+
"family": "qwen",
|
|
1522
|
+
"attachment": false,
|
|
1523
|
+
"reasoning": false,
|
|
883
1524
|
"tool_call": true,
|
|
1525
|
+
"structured_output": true,
|
|
884
1526
|
"temperature": true,
|
|
885
|
-
"
|
|
886
|
-
"
|
|
1527
|
+
"knowledge": "2025-04",
|
|
1528
|
+
"release_date": "2025-09-23",
|
|
1529
|
+
"last_updated": "2025-09-23",
|
|
887
1530
|
"modalities": {
|
|
888
1531
|
"input": [
|
|
889
|
-
"text"
|
|
890
|
-
"image",
|
|
891
|
-
"video"
|
|
1532
|
+
"text"
|
|
892
1533
|
],
|
|
893
1534
|
"output": [
|
|
894
1535
|
"text"
|
|
895
1536
|
]
|
|
896
1537
|
},
|
|
897
|
-
"open_weights":
|
|
1538
|
+
"open_weights": false,
|
|
898
1539
|
"limit": {
|
|
899
|
-
"context":
|
|
900
|
-
"output":
|
|
1540
|
+
"context": 256000,
|
|
1541
|
+
"output": 65536
|
|
901
1542
|
},
|
|
902
1543
|
"cost": {
|
|
903
|
-
"input":
|
|
904
|
-
"output":
|
|
905
|
-
"cache_read": 0.
|
|
1544
|
+
"input": 1.2,
|
|
1545
|
+
"output": 6,
|
|
1546
|
+
"cache_read": 0.24,
|
|
1547
|
+
"tiers": [
|
|
1548
|
+
{
|
|
1549
|
+
"input": 2.4,
|
|
1550
|
+
"output": 12,
|
|
1551
|
+
"cache_read": 0.48,
|
|
1552
|
+
"tier": {
|
|
1553
|
+
"type": "context",
|
|
1554
|
+
"size": 32000
|
|
1555
|
+
}
|
|
1556
|
+
},
|
|
1557
|
+
{
|
|
1558
|
+
"input": 3,
|
|
1559
|
+
"output": 15,
|
|
1560
|
+
"cache_read": 0.6,
|
|
1561
|
+
"tier": {
|
|
1562
|
+
"type": "context",
|
|
1563
|
+
"size": 128000
|
|
1564
|
+
}
|
|
1565
|
+
}
|
|
1566
|
+
]
|
|
906
1567
|
}
|
|
907
1568
|
},
|
|
908
|
-
"
|
|
909
|
-
"id": "
|
|
910
|
-
"name": "
|
|
911
|
-
"description": "
|
|
912
|
-
"family": "
|
|
1569
|
+
"Qwen/Qwen3-30B-A3B": {
|
|
1570
|
+
"id": "Qwen/Qwen3-30B-A3B",
|
|
1571
|
+
"name": "Qwen3 30B A3B",
|
|
1572
|
+
"description": "Sparse MoE Qwen model with 3B active parameters for efficient chat and reasoning",
|
|
1573
|
+
"family": "qwen",
|
|
913
1574
|
"attachment": false,
|
|
914
1575
|
"reasoning": true,
|
|
915
1576
|
"reasoning_options": [],
|
|
916
1577
|
"tool_call": true,
|
|
917
|
-
"interleaved": {
|
|
918
|
-
"field": "reasoning_content"
|
|
919
|
-
},
|
|
920
1578
|
"structured_output": true,
|
|
921
1579
|
"temperature": true,
|
|
922
|
-
"
|
|
923
|
-
"
|
|
924
|
-
"last_updated": "2026-01-19",
|
|
1580
|
+
"release_date": "2025-04-28",
|
|
1581
|
+
"last_updated": "2025-04-28",
|
|
925
1582
|
"modalities": {
|
|
926
1583
|
"input": [
|
|
927
1584
|
"text"
|
|
@@ -932,39 +1589,33 @@
|
|
|
932
1589
|
},
|
|
933
1590
|
"open_weights": true,
|
|
934
1591
|
"limit": {
|
|
935
|
-
"context":
|
|
1592
|
+
"context": 40960,
|
|
936
1593
|
"output": 16384
|
|
937
1594
|
},
|
|
938
1595
|
"cost": {
|
|
939
|
-
"input": 0.
|
|
940
|
-
"output": 0.
|
|
941
|
-
"cache_read": 0.01
|
|
1596
|
+
"input": 0.12,
|
|
1597
|
+
"output": 0.5
|
|
942
1598
|
}
|
|
943
1599
|
},
|
|
944
|
-
"
|
|
945
|
-
"id": "
|
|
946
|
-
"name": "
|
|
947
|
-
"description": "
|
|
948
|
-
"family": "
|
|
949
|
-
"attachment":
|
|
1600
|
+
"Qwen/Qwen3.5-397B-A17B": {
|
|
1601
|
+
"id": "Qwen/Qwen3.5-397B-A17B",
|
|
1602
|
+
"name": "Qwen 3.5 397B A17B",
|
|
1603
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1604
|
+
"family": "qwen",
|
|
1605
|
+
"attachment": true,
|
|
950
1606
|
"reasoning": true,
|
|
951
|
-
"reasoning_options": [
|
|
952
|
-
{
|
|
953
|
-
"type": "toggle"
|
|
954
|
-
}
|
|
955
|
-
],
|
|
1607
|
+
"reasoning_options": [],
|
|
956
1608
|
"tool_call": true,
|
|
957
|
-
"interleaved": {
|
|
958
|
-
"field": "reasoning_content"
|
|
959
|
-
},
|
|
960
1609
|
"structured_output": true,
|
|
961
1610
|
"temperature": true,
|
|
962
|
-
"knowledge": "2025-
|
|
963
|
-
"release_date": "
|
|
964
|
-
"last_updated": "
|
|
1611
|
+
"knowledge": "2025-01",
|
|
1612
|
+
"release_date": "2026-02-01",
|
|
1613
|
+
"last_updated": "2026-04-20",
|
|
965
1614
|
"modalities": {
|
|
966
1615
|
"input": [
|
|
967
|
-
"text"
|
|
1616
|
+
"text",
|
|
1617
|
+
"image",
|
|
1618
|
+
"video"
|
|
968
1619
|
],
|
|
969
1620
|
"output": [
|
|
970
1621
|
"text"
|
|
@@ -972,47 +1623,33 @@
|
|
|
972
1623
|
},
|
|
973
1624
|
"open_weights": true,
|
|
974
1625
|
"limit": {
|
|
975
|
-
"context":
|
|
976
|
-
"output":
|
|
1626
|
+
"context": 262144,
|
|
1627
|
+
"output": 81920
|
|
977
1628
|
},
|
|
978
1629
|
"cost": {
|
|
979
|
-
"input": 0.
|
|
980
|
-
"output":
|
|
981
|
-
"cache_read": 0.
|
|
1630
|
+
"input": 0.45,
|
|
1631
|
+
"output": 3,
|
|
1632
|
+
"cache_read": 0.22
|
|
982
1633
|
}
|
|
983
1634
|
},
|
|
984
|
-
"
|
|
985
|
-
"id": "
|
|
986
|
-
"name": "
|
|
987
|
-
"description": "
|
|
988
|
-
"family": "
|
|
989
|
-
"attachment":
|
|
1635
|
+
"Qwen/Qwen3.6-35B-A3B": {
|
|
1636
|
+
"id": "Qwen/Qwen3.6-35B-A3B",
|
|
1637
|
+
"name": "Qwen3.6 35B A3B",
|
|
1638
|
+
"description": "Qwen vision-language model for visual reasoning, documents, and agent tasks",
|
|
1639
|
+
"family": "qwen",
|
|
1640
|
+
"attachment": true,
|
|
990
1641
|
"reasoning": true,
|
|
991
|
-
"reasoning_options": [
|
|
992
|
-
{
|
|
993
|
-
"type": "toggle"
|
|
994
|
-
},
|
|
995
|
-
{
|
|
996
|
-
"type": "effort",
|
|
997
|
-
"values": [
|
|
998
|
-
"low",
|
|
999
|
-
"medium",
|
|
1000
|
-
"high",
|
|
1001
|
-
"xhigh"
|
|
1002
|
-
]
|
|
1003
|
-
}
|
|
1004
|
-
],
|
|
1642
|
+
"reasoning_options": [],
|
|
1005
1643
|
"tool_call": true,
|
|
1006
|
-
"interleaved": {
|
|
1007
|
-
"field": "reasoning_content"
|
|
1008
|
-
},
|
|
1009
1644
|
"structured_output": true,
|
|
1010
1645
|
"temperature": true,
|
|
1011
|
-
"release_date": "2026-
|
|
1012
|
-
"last_updated": "2026-
|
|
1646
|
+
"release_date": "2026-04-01",
|
|
1647
|
+
"last_updated": "2026-04-01",
|
|
1013
1648
|
"modalities": {
|
|
1014
1649
|
"input": [
|
|
1015
|
-
"text"
|
|
1650
|
+
"text",
|
|
1651
|
+
"image",
|
|
1652
|
+
"video"
|
|
1016
1653
|
],
|
|
1017
1654
|
"output": [
|
|
1018
1655
|
"text"
|
|
@@ -1020,36 +1657,27 @@
|
|
|
1020
1657
|
},
|
|
1021
1658
|
"open_weights": true,
|
|
1022
1659
|
"limit": {
|
|
1023
|
-
"context":
|
|
1024
|
-
"output":
|
|
1660
|
+
"context": 262144,
|
|
1661
|
+
"output": 81920
|
|
1025
1662
|
},
|
|
1026
1663
|
"cost": {
|
|
1027
|
-
"input": 0.
|
|
1028
|
-
"output":
|
|
1029
|
-
"cache_read": 0.18
|
|
1664
|
+
"input": 0.1,
|
|
1665
|
+
"output": 0.95
|
|
1030
1666
|
}
|
|
1031
1667
|
},
|
|
1032
|
-
"
|
|
1033
|
-
"id": "
|
|
1034
|
-
"name": "
|
|
1035
|
-
"description": "
|
|
1036
|
-
"family": "
|
|
1668
|
+
"Qwen/Qwen3-Next-80B-A3B-Instruct": {
|
|
1669
|
+
"id": "Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
1670
|
+
"name": "Qwen3-Next 80B-A3B Instruct",
|
|
1671
|
+
"description": "Qwen instruction model for multilingual chat, reasoning, and tool use",
|
|
1672
|
+
"family": "qwen",
|
|
1037
1673
|
"attachment": false,
|
|
1038
|
-
"reasoning":
|
|
1039
|
-
"reasoning_options": [
|
|
1040
|
-
{
|
|
1041
|
-
"type": "toggle"
|
|
1042
|
-
}
|
|
1043
|
-
],
|
|
1674
|
+
"reasoning": false,
|
|
1044
1675
|
"tool_call": true,
|
|
1045
|
-
"interleaved": {
|
|
1046
|
-
"field": "reasoning_content"
|
|
1047
|
-
},
|
|
1048
1676
|
"structured_output": true,
|
|
1049
1677
|
"temperature": true,
|
|
1050
|
-
"knowledge": "2025-
|
|
1051
|
-
"release_date": "
|
|
1052
|
-
"last_updated": "
|
|
1678
|
+
"knowledge": "2025-04",
|
|
1679
|
+
"release_date": "2025-09",
|
|
1680
|
+
"last_updated": "2025-09",
|
|
1053
1681
|
"modalities": {
|
|
1054
1682
|
"input": [
|
|
1055
1683
|
"text"
|
|
@@ -1060,18 +1688,17 @@
|
|
|
1060
1688
|
},
|
|
1061
1689
|
"open_weights": true,
|
|
1062
1690
|
"limit": {
|
|
1063
|
-
"context":
|
|
1064
|
-
"output":
|
|
1691
|
+
"context": 262144,
|
|
1692
|
+
"output": 32768
|
|
1065
1693
|
},
|
|
1066
1694
|
"cost": {
|
|
1067
|
-
"input": 0.
|
|
1068
|
-
"output":
|
|
1069
|
-
"cache_read": 0.12
|
|
1695
|
+
"input": 0.09,
|
|
1696
|
+
"output": 1.1
|
|
1070
1697
|
}
|
|
1071
1698
|
},
|
|
1072
|
-
"zai-org/GLM-
|
|
1073
|
-
"id": "zai-org/GLM-
|
|
1074
|
-
"name": "GLM-
|
|
1699
|
+
"zai-org/GLM-5.1": {
|
|
1700
|
+
"id": "zai-org/GLM-5.1",
|
|
1701
|
+
"name": "GLM-5.1",
|
|
1075
1702
|
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1076
1703
|
"family": "glm",
|
|
1077
1704
|
"attachment": false,
|
|
@@ -1088,8 +1715,8 @@
|
|
|
1088
1715
|
"structured_output": true,
|
|
1089
1716
|
"temperature": true,
|
|
1090
1717
|
"knowledge": "2025-04",
|
|
1091
|
-
"release_date": "
|
|
1092
|
-
"last_updated": "
|
|
1718
|
+
"release_date": "2026-04-07",
|
|
1719
|
+
"last_updated": "2026-04-07",
|
|
1093
1720
|
"modalities": {
|
|
1094
1721
|
"input": [
|
|
1095
1722
|
"text"
|
|
@@ -1104,14 +1731,14 @@
|
|
|
1104
1731
|
"output": 16384
|
|
1105
1732
|
},
|
|
1106
1733
|
"cost": {
|
|
1107
|
-
"input":
|
|
1108
|
-
"output":
|
|
1109
|
-
"cache_read": 0.
|
|
1734
|
+
"input": 1.05,
|
|
1735
|
+
"output": 3.5,
|
|
1736
|
+
"cache_read": 0.205
|
|
1110
1737
|
}
|
|
1111
1738
|
},
|
|
1112
|
-
"zai-org/GLM-
|
|
1113
|
-
"id": "zai-org/GLM-
|
|
1114
|
-
"name": "GLM-
|
|
1739
|
+
"zai-org/GLM-4.6": {
|
|
1740
|
+
"id": "zai-org/GLM-4.6",
|
|
1741
|
+
"name": "GLM-4.6",
|
|
1115
1742
|
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1116
1743
|
"family": "glm",
|
|
1117
1744
|
"attachment": false,
|
|
@@ -1128,8 +1755,8 @@
|
|
|
1128
1755
|
"structured_output": true,
|
|
1129
1756
|
"temperature": true,
|
|
1130
1757
|
"knowledge": "2025-04",
|
|
1131
|
-
"release_date": "
|
|
1132
|
-
"last_updated": "
|
|
1758
|
+
"release_date": "2025-09-30",
|
|
1759
|
+
"last_updated": "2025-09-30",
|
|
1133
1760
|
"modalities": {
|
|
1134
1761
|
"input": [
|
|
1135
1762
|
"text"
|
|
@@ -1141,20 +1768,20 @@
|
|
|
1141
1768
|
"open_weights": true,
|
|
1142
1769
|
"limit": {
|
|
1143
1770
|
"context": 202752,
|
|
1144
|
-
"output":
|
|
1771
|
+
"output": 131072
|
|
1145
1772
|
},
|
|
1146
1773
|
"cost": {
|
|
1147
|
-
"input":
|
|
1148
|
-
"output":
|
|
1149
|
-
"cache_read": 0.
|
|
1774
|
+
"input": 0.5,
|
|
1775
|
+
"output": 2,
|
|
1776
|
+
"cache_read": 0.1
|
|
1150
1777
|
}
|
|
1151
1778
|
},
|
|
1152
|
-
"
|
|
1153
|
-
"id": "
|
|
1154
|
-
"name": "
|
|
1155
|
-
"description": "
|
|
1156
|
-
"family": "
|
|
1157
|
-
"attachment":
|
|
1779
|
+
"zai-org/GLM-5": {
|
|
1780
|
+
"id": "zai-org/GLM-5",
|
|
1781
|
+
"name": "GLM-5",
|
|
1782
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1783
|
+
"family": "glm",
|
|
1784
|
+
"attachment": false,
|
|
1158
1785
|
"reasoning": true,
|
|
1159
1786
|
"reasoning_options": [
|
|
1160
1787
|
{
|
|
@@ -1167,14 +1794,12 @@
|
|
|
1167
1794
|
},
|
|
1168
1795
|
"structured_output": true,
|
|
1169
1796
|
"temperature": true,
|
|
1170
|
-
"knowledge": "
|
|
1171
|
-
"release_date": "2026-
|
|
1172
|
-
"last_updated": "2026-
|
|
1797
|
+
"knowledge": "2025-12",
|
|
1798
|
+
"release_date": "2026-02-12",
|
|
1799
|
+
"last_updated": "2026-02-12",
|
|
1173
1800
|
"modalities": {
|
|
1174
1801
|
"input": [
|
|
1175
|
-
"text"
|
|
1176
|
-
"image",
|
|
1177
|
-
"video"
|
|
1802
|
+
"text"
|
|
1178
1803
|
],
|
|
1179
1804
|
"output": [
|
|
1180
1805
|
"text"
|
|
@@ -1182,21 +1807,21 @@
|
|
|
1182
1807
|
},
|
|
1183
1808
|
"open_weights": true,
|
|
1184
1809
|
"limit": {
|
|
1185
|
-
"context":
|
|
1810
|
+
"context": 202752,
|
|
1186
1811
|
"output": 16384
|
|
1187
1812
|
},
|
|
1188
1813
|
"cost": {
|
|
1189
|
-
"input": 0.
|
|
1190
|
-
"output":
|
|
1191
|
-
"cache_read": 0.
|
|
1814
|
+
"input": 0.6,
|
|
1815
|
+
"output": 2.08,
|
|
1816
|
+
"cache_read": 0.12
|
|
1192
1817
|
}
|
|
1193
|
-
},
|
|
1194
|
-
"
|
|
1195
|
-
"id": "
|
|
1196
|
-
"name": "
|
|
1197
|
-
"description": "
|
|
1198
|
-
"family": "
|
|
1199
|
-
"attachment":
|
|
1818
|
+
},
|
|
1819
|
+
"zai-org/GLM-4.7": {
|
|
1820
|
+
"id": "zai-org/GLM-4.7",
|
|
1821
|
+
"name": "GLM-4.7",
|
|
1822
|
+
"description": "Flagship GLM model for hybrid reasoning, coding, and agentic engineering",
|
|
1823
|
+
"family": "glm",
|
|
1824
|
+
"attachment": false,
|
|
1200
1825
|
"reasoning": true,
|
|
1201
1826
|
"reasoning_options": [
|
|
1202
1827
|
{
|
|
@@ -1209,14 +1834,12 @@
|
|
|
1209
1834
|
},
|
|
1210
1835
|
"structured_output": true,
|
|
1211
1836
|
"temperature": true,
|
|
1212
|
-
"knowledge": "2025-
|
|
1213
|
-
"release_date": "
|
|
1214
|
-
"last_updated": "
|
|
1837
|
+
"knowledge": "2025-04",
|
|
1838
|
+
"release_date": "2025-12-22",
|
|
1839
|
+
"last_updated": "2025-12-22",
|
|
1215
1840
|
"modalities": {
|
|
1216
1841
|
"input": [
|
|
1217
|
-
"text"
|
|
1218
|
-
"image",
|
|
1219
|
-
"video"
|
|
1842
|
+
"text"
|
|
1220
1843
|
],
|
|
1221
1844
|
"output": [
|
|
1222
1845
|
"text"
|
|
@@ -1224,25 +1847,34 @@
|
|
|
1224
1847
|
},
|
|
1225
1848
|
"open_weights": true,
|
|
1226
1849
|
"limit": {
|
|
1227
|
-
"context":
|
|
1228
|
-
"output":
|
|
1850
|
+
"context": 202752,
|
|
1851
|
+
"output": 16384
|
|
1229
1852
|
},
|
|
1230
1853
|
"cost": {
|
|
1231
|
-
"input": 0.
|
|
1232
|
-
"output":
|
|
1233
|
-
"cache_read": 0.
|
|
1854
|
+
"input": 0.4,
|
|
1855
|
+
"output": 1.75,
|
|
1856
|
+
"cache_read": 0.08
|
|
1234
1857
|
}
|
|
1235
1858
|
},
|
|
1236
|
-
"
|
|
1237
|
-
"id": "
|
|
1238
|
-
"name": "
|
|
1239
|
-
"description": "
|
|
1240
|
-
"family": "
|
|
1241
|
-
"attachment":
|
|
1859
|
+
"zai-org/GLM-5.2": {
|
|
1860
|
+
"id": "zai-org/GLM-5.2",
|
|
1861
|
+
"name": "GLM-5.2",
|
|
1862
|
+
"description": "Open flagship GLM for long-horizon coding agents and million-token context work",
|
|
1863
|
+
"family": "glm",
|
|
1864
|
+
"attachment": false,
|
|
1242
1865
|
"reasoning": true,
|
|
1243
1866
|
"reasoning_options": [
|
|
1244
1867
|
{
|
|
1245
1868
|
"type": "toggle"
|
|
1869
|
+
},
|
|
1870
|
+
{
|
|
1871
|
+
"type": "effort",
|
|
1872
|
+
"values": [
|
|
1873
|
+
"low",
|
|
1874
|
+
"medium",
|
|
1875
|
+
"high",
|
|
1876
|
+
"xhigh"
|
|
1877
|
+
]
|
|
1246
1878
|
}
|
|
1247
1879
|
],
|
|
1248
1880
|
"tool_call": true,
|
|
@@ -1250,15 +1882,12 @@
|
|
|
1250
1882
|
"field": "reasoning_content"
|
|
1251
1883
|
},
|
|
1252
1884
|
"structured_output": true,
|
|
1253
|
-
"temperature":
|
|
1254
|
-
"
|
|
1255
|
-
"
|
|
1256
|
-
"last_updated": "2026-06-12",
|
|
1885
|
+
"temperature": true,
|
|
1886
|
+
"release_date": "2026-06-13",
|
|
1887
|
+
"last_updated": "2026-06-13",
|
|
1257
1888
|
"modalities": {
|
|
1258
1889
|
"input": [
|
|
1259
|
-
"text"
|
|
1260
|
-
"image",
|
|
1261
|
-
"video"
|
|
1890
|
+
"text"
|
|
1262
1891
|
],
|
|
1263
1892
|
"output": [
|
|
1264
1893
|
"text"
|
|
@@ -1266,40 +1895,35 @@
|
|
|
1266
1895
|
},
|
|
1267
1896
|
"open_weights": true,
|
|
1268
1897
|
"limit": {
|
|
1269
|
-
"context":
|
|
1270
|
-
"output":
|
|
1898
|
+
"context": 1048576,
|
|
1899
|
+
"output": 32768
|
|
1271
1900
|
},
|
|
1272
1901
|
"cost": {
|
|
1273
|
-
"input": 0.
|
|
1274
|
-
"output":
|
|
1275
|
-
"cache_read": 0.
|
|
1902
|
+
"input": 0.75,
|
|
1903
|
+
"output": 2.4,
|
|
1904
|
+
"cache_read": 0.14
|
|
1276
1905
|
}
|
|
1277
1906
|
},
|
|
1278
|
-
"
|
|
1279
|
-
"id": "
|
|
1280
|
-
"name": "
|
|
1281
|
-
"description": "
|
|
1282
|
-
"family": "
|
|
1283
|
-
"attachment":
|
|
1907
|
+
"zai-org/GLM-4.7-Flash": {
|
|
1908
|
+
"id": "zai-org/GLM-4.7-Flash",
|
|
1909
|
+
"name": "GLM-4.7-Flash",
|
|
1910
|
+
"description": "Efficient GLM model for fast reasoning, coding, and agent workflows",
|
|
1911
|
+
"family": "glm-flash",
|
|
1912
|
+
"attachment": false,
|
|
1284
1913
|
"reasoning": true,
|
|
1285
|
-
"reasoning_options": [
|
|
1286
|
-
{
|
|
1287
|
-
"type": "toggle"
|
|
1288
|
-
}
|
|
1289
|
-
],
|
|
1914
|
+
"reasoning_options": [],
|
|
1290
1915
|
"tool_call": true,
|
|
1291
1916
|
"interleaved": {
|
|
1292
1917
|
"field": "reasoning_content"
|
|
1293
1918
|
},
|
|
1294
1919
|
"structured_output": true,
|
|
1295
1920
|
"temperature": true,
|
|
1296
|
-
"knowledge": "
|
|
1297
|
-
"release_date": "2026-
|
|
1298
|
-
"last_updated": "2026-
|
|
1921
|
+
"knowledge": "2025-04",
|
|
1922
|
+
"release_date": "2026-01-19",
|
|
1923
|
+
"last_updated": "2026-01-19",
|
|
1299
1924
|
"modalities": {
|
|
1300
1925
|
"input": [
|
|
1301
|
-
"text"
|
|
1302
|
-
"audio"
|
|
1926
|
+
"text"
|
|
1303
1927
|
],
|
|
1304
1928
|
"output": [
|
|
1305
1929
|
"text"
|
|
@@ -1307,42 +1931,32 @@
|
|
|
1307
1931
|
},
|
|
1308
1932
|
"open_weights": true,
|
|
1309
1933
|
"limit": {
|
|
1310
|
-
"context":
|
|
1934
|
+
"context": 202752,
|
|
1311
1935
|
"output": 16384
|
|
1312
1936
|
},
|
|
1313
1937
|
"cost": {
|
|
1314
|
-
"input":
|
|
1315
|
-
"output":
|
|
1316
|
-
"cache_read": 0.
|
|
1938
|
+
"input": 0.06,
|
|
1939
|
+
"output": 0.4,
|
|
1940
|
+
"cache_read": 0.01
|
|
1317
1941
|
}
|
|
1318
1942
|
},
|
|
1319
|
-
"
|
|
1320
|
-
"id": "
|
|
1321
|
-
"name": "
|
|
1322
|
-
"description": "
|
|
1323
|
-
"family": "
|
|
1943
|
+
"thinkingmachines/Inkling": {
|
|
1944
|
+
"id": "thinkingmachines/Inkling",
|
|
1945
|
+
"name": "Inkling",
|
|
1946
|
+
"description": "Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio",
|
|
1947
|
+
"family": "ling",
|
|
1324
1948
|
"attachment": true,
|
|
1325
1949
|
"reasoning": true,
|
|
1326
|
-
"reasoning_options": [
|
|
1327
|
-
{
|
|
1328
|
-
"type": "toggle"
|
|
1329
|
-
}
|
|
1330
|
-
],
|
|
1950
|
+
"reasoning_options": [],
|
|
1331
1951
|
"tool_call": true,
|
|
1332
|
-
"interleaved": {
|
|
1333
|
-
"field": "reasoning_content"
|
|
1334
|
-
},
|
|
1335
|
-
"structured_output": true,
|
|
1336
1952
|
"temperature": true,
|
|
1337
|
-
"
|
|
1338
|
-
"
|
|
1339
|
-
"last_updated": "2026-04-22",
|
|
1953
|
+
"release_date": "2026-07-15",
|
|
1954
|
+
"last_updated": "2026-07-15",
|
|
1340
1955
|
"modalities": {
|
|
1341
1956
|
"input": [
|
|
1342
1957
|
"text",
|
|
1343
1958
|
"image",
|
|
1344
|
-
"audio"
|
|
1345
|
-
"video"
|
|
1959
|
+
"audio"
|
|
1346
1960
|
],
|
|
1347
1961
|
"output": [
|
|
1348
1962
|
"text"
|
|
@@ -1350,36 +1964,33 @@
|
|
|
1350
1964
|
},
|
|
1351
1965
|
"open_weights": true,
|
|
1352
1966
|
"limit": {
|
|
1353
|
-
"context":
|
|
1354
|
-
"output":
|
|
1967
|
+
"context": 524288,
|
|
1968
|
+
"output": 1048576
|
|
1355
1969
|
},
|
|
1356
1970
|
"cost": {
|
|
1357
|
-
"input": 0.
|
|
1358
|
-
"output":
|
|
1359
|
-
"cache_read": 0.
|
|
1971
|
+
"input": 0.95,
|
|
1972
|
+
"output": 4.05,
|
|
1973
|
+
"cache_read": 0.16
|
|
1360
1974
|
}
|
|
1361
1975
|
},
|
|
1362
|
-
"
|
|
1363
|
-
"id": "
|
|
1364
|
-
"name": "
|
|
1365
|
-
"description": "
|
|
1366
|
-
"family": "
|
|
1976
|
+
"thinkingmachines/Inkling-Small": {
|
|
1977
|
+
"id": "thinkingmachines/Inkling-Small",
|
|
1978
|
+
"name": "Inkling Small",
|
|
1979
|
+
"description": "Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio",
|
|
1980
|
+
"family": "ling",
|
|
1367
1981
|
"attachment": true,
|
|
1368
1982
|
"reasoning": true,
|
|
1369
|
-
"reasoning_options": [
|
|
1370
|
-
{
|
|
1371
|
-
"type": "toggle"
|
|
1372
|
-
}
|
|
1373
|
-
],
|
|
1983
|
+
"reasoning_options": [],
|
|
1374
1984
|
"tool_call": true,
|
|
1375
1985
|
"structured_output": true,
|
|
1376
1986
|
"temperature": true,
|
|
1377
|
-
"release_date": "2026-
|
|
1378
|
-
"last_updated": "2026-
|
|
1987
|
+
"release_date": "2026-07-30",
|
|
1988
|
+
"last_updated": "2026-07-30",
|
|
1379
1989
|
"modalities": {
|
|
1380
1990
|
"input": [
|
|
1381
1991
|
"text",
|
|
1382
|
-
"image"
|
|
1992
|
+
"image",
|
|
1993
|
+
"audio"
|
|
1383
1994
|
],
|
|
1384
1995
|
"output": [
|
|
1385
1996
|
"text"
|
|
@@ -1387,36 +1998,30 @@
|
|
|
1387
1998
|
},
|
|
1388
1999
|
"open_weights": true,
|
|
1389
2000
|
"limit": {
|
|
1390
|
-
"context":
|
|
1391
|
-
"output":
|
|
2001
|
+
"context": 524288,
|
|
2002
|
+
"output": 1048576
|
|
1392
2003
|
},
|
|
1393
2004
|
"cost": {
|
|
1394
|
-
"input": 0.
|
|
1395
|
-
"output":
|
|
2005
|
+
"input": 0.45,
|
|
2006
|
+
"output": 1.2,
|
|
2007
|
+
"cache_read": 0.1
|
|
1396
2008
|
}
|
|
1397
2009
|
},
|
|
1398
|
-
"
|
|
1399
|
-
"id": "
|
|
1400
|
-
"name": "
|
|
1401
|
-
"description": "
|
|
1402
|
-
"family": "
|
|
2010
|
+
"meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8": {
|
|
2011
|
+
"id": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
|
|
2012
|
+
"name": "Llama 4 Maverick 17B FP8",
|
|
2013
|
+
"description": "Open multimodal Llama model for strong reasoning and fast responses",
|
|
2014
|
+
"family": "llama",
|
|
1403
2015
|
"attachment": true,
|
|
1404
|
-
"reasoning":
|
|
1405
|
-
"
|
|
1406
|
-
{
|
|
1407
|
-
"type": "toggle"
|
|
1408
|
-
}
|
|
1409
|
-
],
|
|
1410
|
-
"tool_call": true,
|
|
2016
|
+
"reasoning": false,
|
|
2017
|
+
"tool_call": false,
|
|
1411
2018
|
"structured_output": true,
|
|
1412
|
-
"
|
|
1413
|
-
"
|
|
1414
|
-
"last_updated": "2026-04-02",
|
|
2019
|
+
"release_date": "2025-04-05",
|
|
2020
|
+
"last_updated": "2025-04-05",
|
|
1415
2021
|
"modalities": {
|
|
1416
2022
|
"input": [
|
|
1417
2023
|
"text",
|
|
1418
|
-
"image"
|
|
1419
|
-
"video"
|
|
2024
|
+
"image"
|
|
1420
2025
|
],
|
|
1421
2026
|
"output": [
|
|
1422
2027
|
"text"
|
|
@@ -1424,36 +2029,25 @@
|
|
|
1424
2029
|
},
|
|
1425
2030
|
"open_weights": true,
|
|
1426
2031
|
"limit": {
|
|
1427
|
-
"context":
|
|
1428
|
-
"output":
|
|
2032
|
+
"context": 1048576,
|
|
2033
|
+
"output": 16384
|
|
1429
2034
|
},
|
|
1430
2035
|
"cost": {
|
|
1431
|
-
"input": 0.
|
|
1432
|
-
"output": 0.
|
|
2036
|
+
"input": 0.2,
|
|
2037
|
+
"output": 0.8
|
|
1433
2038
|
}
|
|
1434
2039
|
},
|
|
1435
|
-
"
|
|
1436
|
-
"id": "
|
|
1437
|
-
"name": "
|
|
1438
|
-
"description": "
|
|
1439
|
-
"family": "
|
|
2040
|
+
"meta-llama/Llama-3.3-70B-Instruct-Turbo": {
|
|
2041
|
+
"id": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
|
|
2042
|
+
"name": "Llama 3.3 70B Turbo",
|
|
2043
|
+
"description": "Compact Llama instruction model for fast chat and local deployment",
|
|
2044
|
+
"family": "llama",
|
|
1440
2045
|
"attachment": false,
|
|
1441
|
-
"reasoning":
|
|
1442
|
-
"reasoning_options": [
|
|
1443
|
-
{
|
|
1444
|
-
"type": "effort",
|
|
1445
|
-
"values": [
|
|
1446
|
-
"low",
|
|
1447
|
-
"medium",
|
|
1448
|
-
"high"
|
|
1449
|
-
]
|
|
1450
|
-
}
|
|
1451
|
-
],
|
|
2046
|
+
"reasoning": false,
|
|
1452
2047
|
"tool_call": true,
|
|
1453
2048
|
"structured_output": true,
|
|
1454
|
-
"
|
|
1455
|
-
"
|
|
1456
|
-
"last_updated": "2025-08-05",
|
|
2049
|
+
"release_date": "2024-12-06",
|
|
2050
|
+
"last_updated": "2024-12-06",
|
|
1457
2051
|
"modalities": {
|
|
1458
2052
|
"input": [
|
|
1459
2053
|
"text"
|
|
@@ -1468,35 +2062,25 @@
|
|
|
1468
2062
|
"output": 16384
|
|
1469
2063
|
},
|
|
1470
2064
|
"cost": {
|
|
1471
|
-
"input": 0.
|
|
1472
|
-
"output": 0.
|
|
2065
|
+
"input": 0.1,
|
|
2066
|
+
"output": 0.32
|
|
1473
2067
|
}
|
|
1474
2068
|
},
|
|
1475
|
-
"
|
|
1476
|
-
"id": "
|
|
1477
|
-
"name": "
|
|
1478
|
-
"description": "Open
|
|
1479
|
-
"family": "
|
|
1480
|
-
"attachment":
|
|
1481
|
-
"reasoning":
|
|
1482
|
-
"reasoning_options": [
|
|
1483
|
-
{
|
|
1484
|
-
"type": "effort",
|
|
1485
|
-
"values": [
|
|
1486
|
-
"low",
|
|
1487
|
-
"medium",
|
|
1488
|
-
"high"
|
|
1489
|
-
]
|
|
1490
|
-
}
|
|
1491
|
-
],
|
|
2069
|
+
"meta-llama/Llama-4-Scout-17B-16E-Instruct": {
|
|
2070
|
+
"id": "meta-llama/Llama-4-Scout-17B-16E-Instruct",
|
|
2071
|
+
"name": "Llama 4 Scout 17B",
|
|
2072
|
+
"description": "Open multimodal Llama model for long-context analysis and efficient agents",
|
|
2073
|
+
"family": "llama",
|
|
2074
|
+
"attachment": true,
|
|
2075
|
+
"reasoning": false,
|
|
1492
2076
|
"tool_call": true,
|
|
1493
2077
|
"structured_output": true,
|
|
1494
|
-
"
|
|
1495
|
-
"
|
|
1496
|
-
"last_updated": "2025-08-05",
|
|
2078
|
+
"release_date": "2025-04-05",
|
|
2079
|
+
"last_updated": "2025-04-05",
|
|
1497
2080
|
"modalities": {
|
|
1498
2081
|
"input": [
|
|
1499
|
-
"text"
|
|
2082
|
+
"text",
|
|
2083
|
+
"image"
|
|
1500
2084
|
],
|
|
1501
2085
|
"output": [
|
|
1502
2086
|
"text"
|
|
@@ -1504,12 +2088,12 @@
|
|
|
1504
2088
|
},
|
|
1505
2089
|
"open_weights": true,
|
|
1506
2090
|
"limit": {
|
|
1507
|
-
"context":
|
|
2091
|
+
"context": 327680,
|
|
1508
2092
|
"output": 16384
|
|
1509
2093
|
},
|
|
1510
2094
|
"cost": {
|
|
1511
|
-
"input": 0.
|
|
1512
|
-
"output": 0.
|
|
2095
|
+
"input": 0.1,
|
|
2096
|
+
"output": 0.3
|
|
1513
2097
|
}
|
|
1514
2098
|
}
|
|
1515
2099
|
}
|