llm.rb 15.0.3 → 15.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +433 -3
- data/README.md +188 -71
- data/bin/llm.rb +50 -7
- data/data/alibaba.json +912 -823
- data/data/anthropic.json +234 -187
- data/data/bedrock.json +3702 -2058
- data/data/deepinfra.json +1288 -951
- data/data/deepseek.json +87 -53
- data/data/google.json +670 -670
- data/data/mistral.json +501 -460
- data/data/moonshot.json +43 -248
- data/data/openai.json +1008 -914
- data/data/openrouter.json +14417 -0
- data/data/xai.json +213 -201
- data/data/zai.json +242 -149
- data/docs/deepdive/advanced/compaction.md +5 -5
- data/docs/deepdive/advanced/context.md +8 -6
- data/docs/deepdive/advanced/guard.md +2 -2
- data/docs/deepdive/features/builtin_tools.md +93 -22
- data/docs/deepdive/features/{repl.md → console.md} +28 -28
- data/docs/deepdive/features/database.md +3 -3
- data/docs/deepdive/fundamentals/agents.md +13 -12
- data/docs/deepdive/fundamentals/providers.md +91 -6
- data/docs/deepdive/fundamentals/skills.md +14 -6
- data/docs/deepdive/fundamentals/stream.md +4 -4
- data/docs/deepdive/fundamentals/tools.md +63 -31
- data/docs/deepdive/reference/cost.md +2 -2
- data/docs/deepdive/reference/model_registry.md +2 -2
- data/docs/deepdive/reference/tracer.md +15 -13
- data/docs/deepdive.md +2 -2
- data/lib/llm/active_record/acts_as_agent.rb +9 -5
- data/lib/llm/agent.rb +40 -15
- data/lib/llm/{repl → console}/bar.rb +3 -3
- data/lib/llm/{repl → console}/buffer.rb +24 -9
- data/lib/llm/{repl → console}/color.rb +2 -2
- data/lib/llm/{repl → console}/command.rb +12 -12
- data/lib/llm/{repl → console}/commands/exit.rb +4 -4
- data/lib/llm/{repl → console}/commands/help.rb +1 -1
- data/lib/llm/{repl/commands/compact.rb → console/commands/keep.rb} +11 -9
- data/lib/llm/{repl → console}/commands/model.rb +2 -2
- data/lib/llm/{repl → console}/input/cache.rb +2 -2
- data/lib/llm/{repl → console}/input/char.rb +2 -2
- data/lib/llm/{repl → console}/input/row.rb +1 -1
- data/lib/llm/{repl → console}/input.rb +18 -10
- data/lib/llm/console/markdown/parser.rb +78 -0
- data/lib/llm/{repl → console}/markdown/table.rb +8 -5
- data/lib/llm/{repl → console}/markdown.rb +13 -30
- data/lib/llm/console/node.rb +69 -0
- data/lib/llm/{repl → console}/status.rb +11 -11
- data/lib/llm/{repl → console}/stream.rb +36 -9
- data/lib/llm/{repl → console}/walker.rb +1 -1
- data/lib/llm/{repl → console}/window.rb +17 -17
- data/lib/llm/{repl.rb → console.rb} +39 -19
- data/lib/llm/context/deserializer.rb +2 -1
- data/lib/llm/context.rb +29 -13
- data/lib/llm/cost.rb +13 -0
- data/lib/llm/function/async/reactor.rb +20 -1
- data/lib/llm/function/fork/task.rb +14 -10
- data/lib/llm/function.rb +1 -1
- data/lib/llm/json_adapter.rb +40 -28
- data/lib/llm/message.rb +7 -0
- data/lib/llm/provider.rb +31 -10
- data/lib/llm/providers/alibaba.rb +1 -1
- data/lib/llm/providers/anthropic.rb +1 -1
- data/lib/llm/providers/bedrock/models.rb +2 -2
- data/lib/llm/providers/bedrock.rb +1 -1
- data/lib/llm/providers/deepseek.rb +1 -1
- data/lib/llm/providers/google.rb +1 -1
- data/lib/llm/providers/ollama.rb +1 -1
- data/lib/llm/providers/openai/responses.rb +2 -1
- data/lib/llm/providers/openai.rb +2 -1
- data/lib/llm/providers/openrouter.rb +87 -0
- data/lib/llm/schema/leaf.rb +34 -2
- data/lib/llm/schema.rb +4 -2
- data/lib/llm/sequel/agent.rb +9 -5
- data/lib/llm/skill.rb +7 -1
- data/lib/llm/stream.rb +8 -3
- data/lib/llm/tool/param.rb +5 -1
- data/lib/llm/tool.rb +5 -0
- data/lib/llm/tools/bundle.rb +53 -0
- data/lib/llm/tools/edit-file.rb +7 -2
- data/lib/llm/tools/exec.rb +78 -0
- data/lib/llm/tools/git.rb +27 -26
- data/lib/llm/tools/mkdir.rb +12 -19
- data/lib/llm/tools/read_file.rb +69 -9
- data/lib/llm/tools/rg.rb +20 -24
- data/lib/llm/tools/ruby.rb +17 -25
- data/lib/llm/tools/utils.rb +75 -2
- data/lib/llm/tools/write_file.rb +4 -1
- data/lib/llm/tracer/logger.rb +2 -2
- data/lib/llm/tracer/pretty_logger.rb +4 -4
- data/lib/llm/tracer/telemetry.rb +2 -2
- data/lib/llm/tracer.rb +33 -0
- data/lib/llm/transport/curb.rb +5 -3
- data/lib/llm/transport/http.rb +5 -2
- data/lib/llm/transport/persistent_http.rb +6 -4
- data/lib/llm/transport/utils.rb +8 -6
- data/lib/llm/version.rb +1 -1
- data/lib/llm.rb +18 -12
- data/llm.gemspec +8 -8
- metadata +80 -37
- data/lib/llm/repl/node.rb +0 -44
- data/lib/llm/tools/shell.rb +0 -55
data/data/openai.json
CHANGED
|
@@ -7,101 +7,6 @@
|
|
|
7
7
|
"name": "OpenAI",
|
|
8
8
|
"doc": "https://platform.openai.com/docs/models",
|
|
9
9
|
"models": {
|
|
10
|
-
"gpt-4o": {
|
|
11
|
-
"id": "gpt-4o",
|
|
12
|
-
"name": "GPT-4o",
|
|
13
|
-
"description": "Omni-era GPT for multimodal chat, practical coding, and general assistants",
|
|
14
|
-
"family": "gpt",
|
|
15
|
-
"attachment": true,
|
|
16
|
-
"reasoning": false,
|
|
17
|
-
"tool_call": true,
|
|
18
|
-
"structured_output": true,
|
|
19
|
-
"temperature": true,
|
|
20
|
-
"knowledge": "2023-09",
|
|
21
|
-
"release_date": "2024-05-13",
|
|
22
|
-
"last_updated": "2024-08-06",
|
|
23
|
-
"modalities": {
|
|
24
|
-
"input": [
|
|
25
|
-
"text",
|
|
26
|
-
"image",
|
|
27
|
-
"pdf"
|
|
28
|
-
],
|
|
29
|
-
"output": [
|
|
30
|
-
"text"
|
|
31
|
-
]
|
|
32
|
-
},
|
|
33
|
-
"open_weights": false,
|
|
34
|
-
"limit": {
|
|
35
|
-
"context": 128000,
|
|
36
|
-
"output": 16384
|
|
37
|
-
},
|
|
38
|
-
"cost": {
|
|
39
|
-
"input": 2.5,
|
|
40
|
-
"output": 10,
|
|
41
|
-
"cache_read": 1.25
|
|
42
|
-
}
|
|
43
|
-
},
|
|
44
|
-
"gpt-image-1.5": {
|
|
45
|
-
"id": "gpt-image-1.5",
|
|
46
|
-
"name": "gpt-image-1.5",
|
|
47
|
-
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
48
|
-
"family": "gpt-image",
|
|
49
|
-
"attachment": true,
|
|
50
|
-
"reasoning": false,
|
|
51
|
-
"tool_call": false,
|
|
52
|
-
"temperature": false,
|
|
53
|
-
"release_date": "2025-11-25",
|
|
54
|
-
"last_updated": "2025-11-25",
|
|
55
|
-
"modalities": {
|
|
56
|
-
"input": [
|
|
57
|
-
"text",
|
|
58
|
-
"image"
|
|
59
|
-
],
|
|
60
|
-
"output": [
|
|
61
|
-
"text",
|
|
62
|
-
"image"
|
|
63
|
-
]
|
|
64
|
-
},
|
|
65
|
-
"open_weights": false,
|
|
66
|
-
"limit": {
|
|
67
|
-
"context": 0,
|
|
68
|
-
"input": 0,
|
|
69
|
-
"output": 0
|
|
70
|
-
}
|
|
71
|
-
},
|
|
72
|
-
"gpt-5.3-chat-latest": {
|
|
73
|
-
"id": "gpt-5.3-chat-latest",
|
|
74
|
-
"name": "GPT-5.3 Chat (latest)",
|
|
75
|
-
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
|
|
76
|
-
"family": "gpt",
|
|
77
|
-
"attachment": true,
|
|
78
|
-
"reasoning": false,
|
|
79
|
-
"tool_call": true,
|
|
80
|
-
"structured_output": true,
|
|
81
|
-
"temperature": true,
|
|
82
|
-
"knowledge": "2025-08-31",
|
|
83
|
-
"release_date": "2026-03-03",
|
|
84
|
-
"last_updated": "2026-03-03",
|
|
85
|
-
"modalities": {
|
|
86
|
-
"input": [
|
|
87
|
-
"text",
|
|
88
|
-
"image"
|
|
89
|
-
],
|
|
90
|
-
"output": [
|
|
91
|
-
"text"
|
|
92
|
-
]
|
|
93
|
-
},
|
|
94
|
-
"open_weights": false,
|
|
95
|
-
"limit": {
|
|
96
|
-
"context": 128000,
|
|
97
|
-
"output": 16384
|
|
98
|
-
},
|
|
99
|
-
"cost": {
|
|
100
|
-
"input": 1.75,
|
|
101
|
-
"output": 14,
|
|
102
|
-
"cache_read": 0.175
|
|
103
|
-
}
|
|
104
|
-
},
|
|
105
10
|
"gpt-5-nano": {
|
|
106
11
|
"id": "gpt-5-nano",
|
|
107
12
|
"name": "GPT-5 Nano",
|
|
@@ -147,24 +52,23 @@
|
|
|
147
52
|
"cache_read": 0.005
|
|
148
53
|
}
|
|
149
54
|
},
|
|
150
|
-
"gpt-
|
|
151
|
-
"id": "gpt-
|
|
152
|
-
"name": "GPT-
|
|
153
|
-
"description": "
|
|
154
|
-
"family": "gpt-
|
|
55
|
+
"gpt-4.1-nano": {
|
|
56
|
+
"id": "gpt-4.1-nano",
|
|
57
|
+
"name": "GPT-4.1 nano",
|
|
58
|
+
"description": "Tiny GPT-4.1 option for classification, routing, and very high-volume tasks",
|
|
59
|
+
"family": "gpt-nano",
|
|
155
60
|
"attachment": true,
|
|
156
61
|
"reasoning": false,
|
|
157
62
|
"tool_call": true,
|
|
158
63
|
"structured_output": true,
|
|
159
64
|
"temperature": true,
|
|
160
|
-
"knowledge": "
|
|
161
|
-
"release_date": "
|
|
162
|
-
"last_updated": "
|
|
65
|
+
"knowledge": "2024-04",
|
|
66
|
+
"release_date": "2025-04-14",
|
|
67
|
+
"last_updated": "2025-04-14",
|
|
163
68
|
"modalities": {
|
|
164
69
|
"input": [
|
|
165
70
|
"text",
|
|
166
|
-
"image"
|
|
167
|
-
"pdf"
|
|
71
|
+
"image"
|
|
168
72
|
],
|
|
169
73
|
"output": [
|
|
170
74
|
"text"
|
|
@@ -172,31 +76,33 @@
|
|
|
172
76
|
},
|
|
173
77
|
"open_weights": false,
|
|
174
78
|
"limit": {
|
|
175
|
-
"context":
|
|
176
|
-
"output":
|
|
79
|
+
"context": 1047576,
|
|
80
|
+
"output": 32768
|
|
177
81
|
},
|
|
82
|
+
"status": "deprecated",
|
|
178
83
|
"cost": {
|
|
179
|
-
"input": 0.
|
|
180
|
-
"output": 0.
|
|
181
|
-
"cache_read": 0.
|
|
84
|
+
"input": 0.1,
|
|
85
|
+
"output": 0.4,
|
|
86
|
+
"cache_read": 0.025
|
|
182
87
|
}
|
|
183
88
|
},
|
|
184
|
-
"gpt-
|
|
185
|
-
"id": "gpt-
|
|
186
|
-
"name": "GPT-
|
|
187
|
-
"description": "
|
|
89
|
+
"gpt-4o-2024-05-13": {
|
|
90
|
+
"id": "gpt-4o-2024-05-13",
|
|
91
|
+
"name": "GPT-4o (2024-05-13)",
|
|
92
|
+
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
|
|
188
93
|
"family": "gpt",
|
|
189
|
-
"attachment":
|
|
94
|
+
"attachment": true,
|
|
190
95
|
"reasoning": false,
|
|
191
|
-
"tool_call":
|
|
192
|
-
"structured_output":
|
|
96
|
+
"tool_call": true,
|
|
97
|
+
"structured_output": true,
|
|
193
98
|
"temperature": true,
|
|
194
|
-
"knowledge": "
|
|
195
|
-
"release_date": "
|
|
196
|
-
"last_updated": "
|
|
99
|
+
"knowledge": "2023-09",
|
|
100
|
+
"release_date": "2024-05-13",
|
|
101
|
+
"last_updated": "2024-05-13",
|
|
197
102
|
"modalities": {
|
|
198
103
|
"input": [
|
|
199
|
-
"text"
|
|
104
|
+
"text",
|
|
105
|
+
"image"
|
|
200
106
|
],
|
|
201
107
|
"output": [
|
|
202
108
|
"text"
|
|
@@ -204,29 +110,26 @@
|
|
|
204
110
|
},
|
|
205
111
|
"open_weights": false,
|
|
206
112
|
"limit": {
|
|
207
|
-
"context":
|
|
113
|
+
"context": 128000,
|
|
208
114
|
"output": 4096
|
|
209
115
|
},
|
|
210
116
|
"status": "deprecated",
|
|
211
117
|
"cost": {
|
|
212
|
-
"input":
|
|
213
|
-
"output":
|
|
214
|
-
"cache_read": 0
|
|
118
|
+
"input": 5,
|
|
119
|
+
"output": 15
|
|
215
120
|
}
|
|
216
121
|
},
|
|
217
|
-
"
|
|
218
|
-
"id": "
|
|
219
|
-
"name": "
|
|
220
|
-
"description": "
|
|
221
|
-
"family": "
|
|
122
|
+
"gpt-5-pro": {
|
|
123
|
+
"id": "gpt-5-pro",
|
|
124
|
+
"name": "GPT-5 Pro",
|
|
125
|
+
"description": "Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning",
|
|
126
|
+
"family": "gpt-pro",
|
|
222
127
|
"attachment": true,
|
|
223
128
|
"reasoning": true,
|
|
224
129
|
"reasoning_options": [
|
|
225
130
|
{
|
|
226
131
|
"type": "effort",
|
|
227
132
|
"values": [
|
|
228
|
-
"low",
|
|
229
|
-
"medium",
|
|
230
133
|
"high"
|
|
231
134
|
]
|
|
232
135
|
}
|
|
@@ -234,9 +137,9 @@
|
|
|
234
137
|
"tool_call": true,
|
|
235
138
|
"structured_output": true,
|
|
236
139
|
"temperature": false,
|
|
237
|
-
"knowledge": "
|
|
238
|
-
"release_date": "2025-
|
|
239
|
-
"last_updated": "2025-
|
|
140
|
+
"knowledge": "2024-09-30",
|
|
141
|
+
"release_date": "2025-10-06",
|
|
142
|
+
"last_updated": "2025-10-06",
|
|
240
143
|
"modalities": {
|
|
241
144
|
"input": [
|
|
242
145
|
"text",
|
|
@@ -248,38 +151,69 @@
|
|
|
248
151
|
},
|
|
249
152
|
"open_weights": false,
|
|
250
153
|
"limit": {
|
|
251
|
-
"context":
|
|
252
|
-
"
|
|
154
|
+
"context": 400000,
|
|
155
|
+
"input": 272000,
|
|
156
|
+
"output": 272000
|
|
253
157
|
},
|
|
254
|
-
"status": "deprecated",
|
|
255
158
|
"cost": {
|
|
256
|
-
"input":
|
|
257
|
-
"output":
|
|
159
|
+
"input": 15,
|
|
160
|
+
"output": 120
|
|
258
161
|
}
|
|
259
162
|
},
|
|
260
|
-
"
|
|
261
|
-
"id": "
|
|
262
|
-
"name": "
|
|
263
|
-
"description": "
|
|
264
|
-
"family": "gpt-
|
|
163
|
+
"chatgpt-image-latest": {
|
|
164
|
+
"id": "chatgpt-image-latest",
|
|
165
|
+
"name": "chatgpt-image-latest",
|
|
166
|
+
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
167
|
+
"family": "gpt-image",
|
|
168
|
+
"attachment": true,
|
|
169
|
+
"reasoning": false,
|
|
170
|
+
"tool_call": false,
|
|
171
|
+
"temperature": false,
|
|
172
|
+
"release_date": "2025-12-16",
|
|
173
|
+
"last_updated": "2025-12-16",
|
|
174
|
+
"modalities": {
|
|
175
|
+
"input": [
|
|
176
|
+
"text",
|
|
177
|
+
"image"
|
|
178
|
+
],
|
|
179
|
+
"output": [
|
|
180
|
+
"text",
|
|
181
|
+
"image"
|
|
182
|
+
]
|
|
183
|
+
},
|
|
184
|
+
"open_weights": false,
|
|
185
|
+
"limit": {
|
|
186
|
+
"context": 0,
|
|
187
|
+
"input": 0,
|
|
188
|
+
"output": 0
|
|
189
|
+
}
|
|
190
|
+
},
|
|
191
|
+
"gpt-5.6-sol": {
|
|
192
|
+
"id": "gpt-5.6-sol",
|
|
193
|
+
"name": "GPT-5.6 Sol",
|
|
194
|
+
"description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows",
|
|
195
|
+
"family": "gpt-sol",
|
|
265
196
|
"attachment": true,
|
|
266
197
|
"reasoning": true,
|
|
267
198
|
"reasoning_options": [
|
|
268
199
|
{
|
|
269
200
|
"type": "effort",
|
|
270
201
|
"values": [
|
|
202
|
+
"none",
|
|
203
|
+
"low",
|
|
271
204
|
"medium",
|
|
272
205
|
"high",
|
|
273
|
-
"xhigh"
|
|
206
|
+
"xhigh",
|
|
207
|
+
"max"
|
|
274
208
|
]
|
|
275
209
|
}
|
|
276
210
|
],
|
|
277
211
|
"tool_call": true,
|
|
278
212
|
"structured_output": true,
|
|
279
213
|
"temperature": false,
|
|
280
|
-
"knowledge": "
|
|
281
|
-
"release_date": "2026-
|
|
282
|
-
"last_updated": "2026-
|
|
214
|
+
"knowledge": "2026-02-16",
|
|
215
|
+
"release_date": "2026-07-09",
|
|
216
|
+
"last_updated": "2026-07-09",
|
|
283
217
|
"modalities": {
|
|
284
218
|
"input": [
|
|
285
219
|
"text",
|
|
@@ -296,102 +230,74 @@
|
|
|
296
230
|
"input": 922000,
|
|
297
231
|
"output": 128000
|
|
298
232
|
},
|
|
299
|
-
"
|
|
300
|
-
"
|
|
301
|
-
|
|
302
|
-
|
|
303
|
-
|
|
304
|
-
|
|
305
|
-
|
|
306
|
-
|
|
307
|
-
|
|
308
|
-
|
|
233
|
+
"experimental": {
|
|
234
|
+
"modes": {
|
|
235
|
+
"fast": {
|
|
236
|
+
"cost": {
|
|
237
|
+
"input": 8,
|
|
238
|
+
"output": 40,
|
|
239
|
+
"cache_read": 0.8,
|
|
240
|
+
"cache_write": 10
|
|
241
|
+
},
|
|
242
|
+
"provider": {
|
|
243
|
+
"body": {
|
|
244
|
+
"service_tier": "priority"
|
|
245
|
+
}
|
|
246
|
+
}
|
|
247
|
+
},
|
|
248
|
+
"pro": {
|
|
249
|
+
"provider": {
|
|
250
|
+
"body": {
|
|
251
|
+
"reasoning": {
|
|
252
|
+
"mode": "pro"
|
|
253
|
+
}
|
|
254
|
+
}
|
|
309
255
|
}
|
|
310
256
|
}
|
|
311
|
-
],
|
|
312
|
-
"context_over_200k": {
|
|
313
|
-
"input": 60,
|
|
314
|
-
"output": 270
|
|
315
257
|
}
|
|
316
|
-
}
|
|
317
|
-
},
|
|
318
|
-
"gpt-4-turbo": {
|
|
319
|
-
"id": "gpt-4-turbo",
|
|
320
|
-
"name": "GPT-4 Turbo",
|
|
321
|
-
"description": "Compact GPT model for low-latency assistance and high-volume workloads",
|
|
322
|
-
"family": "gpt",
|
|
323
|
-
"attachment": true,
|
|
324
|
-
"reasoning": false,
|
|
325
|
-
"tool_call": true,
|
|
326
|
-
"structured_output": false,
|
|
327
|
-
"temperature": true,
|
|
328
|
-
"knowledge": "2023-12",
|
|
329
|
-
"release_date": "2023-11-06",
|
|
330
|
-
"last_updated": "2024-04-09",
|
|
331
|
-
"modalities": {
|
|
332
|
-
"input": [
|
|
333
|
-
"text",
|
|
334
|
-
"image"
|
|
335
|
-
],
|
|
336
|
-
"output": [
|
|
337
|
-
"text"
|
|
338
|
-
]
|
|
339
|
-
},
|
|
340
|
-
"open_weights": false,
|
|
341
|
-
"limit": {
|
|
342
|
-
"context": 128000,
|
|
343
|
-
"output": 4096
|
|
344
|
-
},
|
|
345
|
-
"status": "deprecated",
|
|
346
|
-
"cost": {
|
|
347
|
-
"input": 10,
|
|
348
|
-
"output": 30
|
|
349
|
-
}
|
|
350
|
-
},
|
|
351
|
-
"gpt-image-1-mini": {
|
|
352
|
-
"id": "gpt-image-1-mini",
|
|
353
|
-
"name": "gpt-image-1-mini",
|
|
354
|
-
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
355
|
-
"family": "gpt-image",
|
|
356
|
-
"attachment": true,
|
|
357
|
-
"reasoning": false,
|
|
358
|
-
"tool_call": false,
|
|
359
|
-
"temperature": false,
|
|
360
|
-
"release_date": "2025-09-26",
|
|
361
|
-
"last_updated": "2025-09-26",
|
|
362
|
-
"modalities": {
|
|
363
|
-
"input": [
|
|
364
|
-
"text",
|
|
365
|
-
"image"
|
|
366
|
-
],
|
|
367
|
-
"output": [
|
|
368
|
-
"text",
|
|
369
|
-
"image"
|
|
370
|
-
]
|
|
371
258
|
},
|
|
372
|
-
"
|
|
373
|
-
|
|
374
|
-
"
|
|
375
|
-
"
|
|
376
|
-
"
|
|
259
|
+
"cost": {
|
|
260
|
+
"input": 4,
|
|
261
|
+
"output": 20,
|
|
262
|
+
"cache_read": 0.4,
|
|
263
|
+
"cache_write": 5,
|
|
264
|
+
"tiers": [
|
|
265
|
+
{
|
|
266
|
+
"input": 8,
|
|
267
|
+
"output": 30,
|
|
268
|
+
"cache_read": 0.8,
|
|
269
|
+
"cache_write": 10,
|
|
270
|
+
"tier": {
|
|
271
|
+
"type": "context",
|
|
272
|
+
"size": 272000
|
|
273
|
+
}
|
|
274
|
+
}
|
|
275
|
+
],
|
|
276
|
+
"context_over_200k": {
|
|
277
|
+
"input": 8,
|
|
278
|
+
"output": 30,
|
|
279
|
+
"cache_read": 0.8,
|
|
280
|
+
"cache_write": 10
|
|
281
|
+
}
|
|
377
282
|
}
|
|
378
283
|
},
|
|
379
|
-
"gpt-
|
|
380
|
-
"id": "gpt-
|
|
381
|
-
"name": "GPT-
|
|
284
|
+
"gpt-4o-2024-08-06": {
|
|
285
|
+
"id": "gpt-4o-2024-08-06",
|
|
286
|
+
"name": "GPT-4o (2024-08-06)",
|
|
382
287
|
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
|
|
383
288
|
"family": "gpt",
|
|
384
289
|
"attachment": true,
|
|
385
290
|
"reasoning": false,
|
|
386
291
|
"tool_call": true,
|
|
387
|
-
"structured_output":
|
|
292
|
+
"structured_output": true,
|
|
388
293
|
"temperature": true,
|
|
389
|
-
"knowledge": "2023-
|
|
390
|
-
"release_date": "
|
|
391
|
-
"last_updated": "2024-
|
|
294
|
+
"knowledge": "2023-09",
|
|
295
|
+
"release_date": "2024-08-06",
|
|
296
|
+
"last_updated": "2024-08-06",
|
|
392
297
|
"modalities": {
|
|
393
298
|
"input": [
|
|
394
|
-
"text"
|
|
299
|
+
"text",
|
|
300
|
+
"image"
|
|
395
301
|
],
|
|
396
302
|
"output": [
|
|
397
303
|
"text"
|
|
@@ -399,27 +305,26 @@
|
|
|
399
305
|
},
|
|
400
306
|
"open_weights": false,
|
|
401
307
|
"limit": {
|
|
402
|
-
"context":
|
|
403
|
-
"output":
|
|
308
|
+
"context": 128000,
|
|
309
|
+
"output": 16384
|
|
404
310
|
},
|
|
405
|
-
"status": "deprecated",
|
|
406
311
|
"cost": {
|
|
407
|
-
"input":
|
|
408
|
-
"output":
|
|
312
|
+
"input": 2.5,
|
|
313
|
+
"output": 10,
|
|
314
|
+
"cache_read": 1.25
|
|
409
315
|
}
|
|
410
316
|
},
|
|
411
|
-
"gpt-
|
|
412
|
-
"id": "gpt-
|
|
413
|
-
"name": "GPT-
|
|
414
|
-
"description": "
|
|
415
|
-
"family": "gpt-
|
|
317
|
+
"gpt-6-astra": {
|
|
318
|
+
"id": "gpt-6-astra",
|
|
319
|
+
"name": "GPT-6 Astra",
|
|
320
|
+
"description": "GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.",
|
|
321
|
+
"family": "gpt-astra",
|
|
416
322
|
"attachment": true,
|
|
417
323
|
"reasoning": true,
|
|
418
324
|
"reasoning_options": [
|
|
419
325
|
{
|
|
420
326
|
"type": "effort",
|
|
421
327
|
"values": [
|
|
422
|
-
"none",
|
|
423
328
|
"low",
|
|
424
329
|
"medium",
|
|
425
330
|
"high",
|
|
@@ -431,9 +336,9 @@
|
|
|
431
336
|
"tool_call": true,
|
|
432
337
|
"structured_output": true,
|
|
433
338
|
"temperature": false,
|
|
434
|
-
"knowledge": "2026-
|
|
435
|
-
"release_date": "2026-
|
|
436
|
-
"last_updated": "2026-
|
|
339
|
+
"knowledge": "2026-04-30",
|
|
340
|
+
"release_date": "2026-09-04",
|
|
341
|
+
"last_updated": "2026-09-04",
|
|
437
342
|
"modalities": {
|
|
438
343
|
"input": [
|
|
439
344
|
"text",
|
|
@@ -454,10 +359,10 @@
|
|
|
454
359
|
"modes": {
|
|
455
360
|
"fast": {
|
|
456
361
|
"cost": {
|
|
457
|
-
"input":
|
|
458
|
-
"output":
|
|
459
|
-
"cache_read":
|
|
460
|
-
"cache_write":
|
|
362
|
+
"input": 20,
|
|
363
|
+
"output": 100,
|
|
364
|
+
"cache_read": 2,
|
|
365
|
+
"cache_write": 25
|
|
461
366
|
},
|
|
462
367
|
"provider": {
|
|
463
368
|
"body": {
|
|
@@ -477,16 +382,16 @@
|
|
|
477
382
|
}
|
|
478
383
|
},
|
|
479
384
|
"cost": {
|
|
480
|
-
"input":
|
|
481
|
-
"output":
|
|
482
|
-
"cache_read":
|
|
483
|
-
"cache_write":
|
|
385
|
+
"input": 10,
|
|
386
|
+
"output": 50,
|
|
387
|
+
"cache_read": 1,
|
|
388
|
+
"cache_write": 12.5,
|
|
484
389
|
"tiers": [
|
|
485
390
|
{
|
|
486
|
-
"input":
|
|
487
|
-
"output":
|
|
488
|
-
"cache_read":
|
|
489
|
-
"cache_write":
|
|
391
|
+
"input": 20,
|
|
392
|
+
"output": 75,
|
|
393
|
+
"cache_read": 2,
|
|
394
|
+
"cache_write": 25,
|
|
490
395
|
"tier": {
|
|
491
396
|
"type": "context",
|
|
492
397
|
"size": 272000
|
|
@@ -494,36 +399,36 @@
|
|
|
494
399
|
}
|
|
495
400
|
],
|
|
496
401
|
"context_over_200k": {
|
|
497
|
-
"input":
|
|
498
|
-
"output":
|
|
499
|
-
"cache_read":
|
|
500
|
-
"cache_write":
|
|
402
|
+
"input": 20,
|
|
403
|
+
"output": 75,
|
|
404
|
+
"cache_read": 2,
|
|
405
|
+
"cache_write": 25
|
|
501
406
|
}
|
|
502
407
|
}
|
|
503
408
|
},
|
|
504
|
-
"
|
|
505
|
-
"id": "
|
|
506
|
-
"name": "
|
|
507
|
-
"description": "
|
|
508
|
-
"family": "
|
|
409
|
+
"gpt-5.2-pro": {
|
|
410
|
+
"id": "gpt-5.2-pro",
|
|
411
|
+
"name": "GPT-5.2 Pro",
|
|
412
|
+
"description": "Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows",
|
|
413
|
+
"family": "gpt-pro",
|
|
509
414
|
"attachment": true,
|
|
510
415
|
"reasoning": true,
|
|
511
416
|
"reasoning_options": [
|
|
512
417
|
{
|
|
513
418
|
"type": "effort",
|
|
514
419
|
"values": [
|
|
515
|
-
"low",
|
|
516
420
|
"medium",
|
|
517
|
-
"high"
|
|
421
|
+
"high",
|
|
422
|
+
"xhigh"
|
|
518
423
|
]
|
|
519
424
|
}
|
|
520
425
|
],
|
|
521
426
|
"tool_call": true,
|
|
522
|
-
"structured_output":
|
|
427
|
+
"structured_output": false,
|
|
523
428
|
"temperature": false,
|
|
524
|
-
"knowledge": "
|
|
525
|
-
"release_date": "2025-
|
|
526
|
-
"last_updated": "2025-
|
|
429
|
+
"knowledge": "2025-08-31",
|
|
430
|
+
"release_date": "2025-12-11",
|
|
431
|
+
"last_updated": "2025-12-11",
|
|
527
432
|
"modalities": {
|
|
528
433
|
"input": [
|
|
529
434
|
"text",
|
|
@@ -535,25 +440,28 @@
|
|
|
535
440
|
},
|
|
536
441
|
"open_weights": false,
|
|
537
442
|
"limit": {
|
|
538
|
-
"context":
|
|
539
|
-
"
|
|
443
|
+
"context": 400000,
|
|
444
|
+
"input": 272000,
|
|
445
|
+
"output": 128000
|
|
540
446
|
},
|
|
541
447
|
"cost": {
|
|
542
|
-
"input":
|
|
543
|
-
"output":
|
|
448
|
+
"input": 21,
|
|
449
|
+
"output": 168
|
|
544
450
|
}
|
|
545
451
|
},
|
|
546
|
-
"gpt-5.
|
|
547
|
-
"id": "gpt-5.
|
|
548
|
-
"name": "GPT-5.
|
|
549
|
-
"description": "
|
|
550
|
-
"family": "gpt-
|
|
452
|
+
"gpt-5.3-codex-spark": {
|
|
453
|
+
"id": "gpt-5.3-codex-spark",
|
|
454
|
+
"name": "GPT-5.3 Codex Spark",
|
|
455
|
+
"description": "Coding-optimized GPT model for repository edits, reviews, and agentic software work",
|
|
456
|
+
"family": "gpt-codex-spark",
|
|
551
457
|
"attachment": true,
|
|
552
458
|
"reasoning": true,
|
|
553
459
|
"reasoning_options": [
|
|
554
460
|
{
|
|
555
461
|
"type": "effort",
|
|
556
462
|
"values": [
|
|
463
|
+
"none",
|
|
464
|
+
"low",
|
|
557
465
|
"medium",
|
|
558
466
|
"high",
|
|
559
467
|
"xhigh"
|
|
@@ -561,15 +469,97 @@
|
|
|
561
469
|
}
|
|
562
470
|
],
|
|
563
471
|
"tool_call": true,
|
|
564
|
-
"structured_output":
|
|
472
|
+
"structured_output": true,
|
|
565
473
|
"temperature": false,
|
|
566
474
|
"knowledge": "2025-08-31",
|
|
475
|
+
"release_date": "2026-02-05",
|
|
476
|
+
"last_updated": "2026-02-05",
|
|
477
|
+
"modalities": {
|
|
478
|
+
"input": [
|
|
479
|
+
"text",
|
|
480
|
+
"image",
|
|
481
|
+
"pdf"
|
|
482
|
+
],
|
|
483
|
+
"output": [
|
|
484
|
+
"text"
|
|
485
|
+
]
|
|
486
|
+
},
|
|
487
|
+
"open_weights": false,
|
|
488
|
+
"limit": {
|
|
489
|
+
"context": 128000,
|
|
490
|
+
"input": 100000,
|
|
491
|
+
"output": 32000
|
|
492
|
+
},
|
|
493
|
+
"cost": {
|
|
494
|
+
"input": 1.75,
|
|
495
|
+
"output": 14,
|
|
496
|
+
"cache_read": 0.175
|
|
497
|
+
}
|
|
498
|
+
},
|
|
499
|
+
"gpt-4.1-mini": {
|
|
500
|
+
"id": "gpt-4.1-mini",
|
|
501
|
+
"name": "GPT-4.1 mini",
|
|
502
|
+
"description": "Affordable GPT-4.1 lane for fast coding help and structured extraction",
|
|
503
|
+
"family": "gpt-mini",
|
|
504
|
+
"attachment": true,
|
|
505
|
+
"reasoning": false,
|
|
506
|
+
"tool_call": true,
|
|
507
|
+
"structured_output": true,
|
|
508
|
+
"temperature": true,
|
|
509
|
+
"knowledge": "2024-04",
|
|
510
|
+
"release_date": "2025-04-14",
|
|
511
|
+
"last_updated": "2025-04-14",
|
|
512
|
+
"modalities": {
|
|
513
|
+
"input": [
|
|
514
|
+
"text",
|
|
515
|
+
"image",
|
|
516
|
+
"pdf"
|
|
517
|
+
],
|
|
518
|
+
"output": [
|
|
519
|
+
"text"
|
|
520
|
+
]
|
|
521
|
+
},
|
|
522
|
+
"open_weights": false,
|
|
523
|
+
"limit": {
|
|
524
|
+
"context": 1047576,
|
|
525
|
+
"output": 32768
|
|
526
|
+
},
|
|
527
|
+
"cost": {
|
|
528
|
+
"input": 0.4,
|
|
529
|
+
"output": 1.6,
|
|
530
|
+
"cache_read": 0.1
|
|
531
|
+
}
|
|
532
|
+
},
|
|
533
|
+
"gpt-5.4": {
|
|
534
|
+
"id": "gpt-5.4",
|
|
535
|
+
"name": "GPT-5.4",
|
|
536
|
+
"description": "Agent-ready GPT for coding and computer-use workflows at a lower cost",
|
|
537
|
+
"family": "gpt",
|
|
538
|
+
"attachment": true,
|
|
539
|
+
"reasoning": true,
|
|
540
|
+
"reasoning_options": [
|
|
541
|
+
{
|
|
542
|
+
"type": "effort",
|
|
543
|
+
"values": [
|
|
544
|
+
"none",
|
|
545
|
+
"low",
|
|
546
|
+
"medium",
|
|
547
|
+
"high",
|
|
548
|
+
"xhigh"
|
|
549
|
+
]
|
|
550
|
+
}
|
|
551
|
+
],
|
|
552
|
+
"tool_call": true,
|
|
553
|
+
"structured_output": true,
|
|
554
|
+
"temperature": true,
|
|
555
|
+
"knowledge": "2025-08-31",
|
|
567
556
|
"release_date": "2026-03-05",
|
|
568
557
|
"last_updated": "2026-03-05",
|
|
569
558
|
"modalities": {
|
|
570
559
|
"input": [
|
|
571
560
|
"text",
|
|
572
|
-
"image"
|
|
561
|
+
"image",
|
|
562
|
+
"pdf"
|
|
573
563
|
],
|
|
574
564
|
"output": [
|
|
575
565
|
"text"
|
|
@@ -581,29 +571,81 @@
|
|
|
581
571
|
"input": 922000,
|
|
582
572
|
"output": 128000
|
|
583
573
|
},
|
|
574
|
+
"experimental": {
|
|
575
|
+
"modes": {
|
|
576
|
+
"fast": {
|
|
577
|
+
"cost": {
|
|
578
|
+
"input": 5,
|
|
579
|
+
"output": 30,
|
|
580
|
+
"cache_read": 0.5
|
|
581
|
+
},
|
|
582
|
+
"provider": {
|
|
583
|
+
"body": {
|
|
584
|
+
"service_tier": "priority"
|
|
585
|
+
}
|
|
586
|
+
}
|
|
587
|
+
}
|
|
588
|
+
}
|
|
589
|
+
},
|
|
590
|
+
"cost": {
|
|
591
|
+
"input": 2.5,
|
|
592
|
+
"output": 15,
|
|
593
|
+
"cache_read": 0.25,
|
|
594
|
+
"tiers": [
|
|
595
|
+
{
|
|
596
|
+
"input": 5,
|
|
597
|
+
"output": 22.5,
|
|
598
|
+
"cache_read": 0.5,
|
|
599
|
+
"tier": {
|
|
600
|
+
"type": "context",
|
|
601
|
+
"size": 272000
|
|
602
|
+
}
|
|
603
|
+
}
|
|
604
|
+
],
|
|
605
|
+
"context_over_200k": {
|
|
606
|
+
"input": 5,
|
|
607
|
+
"output": 22.5,
|
|
608
|
+
"cache_read": 0.5
|
|
609
|
+
}
|
|
610
|
+
}
|
|
611
|
+
},
|
|
612
|
+
"gpt-4-turbo": {
|
|
613
|
+
"id": "gpt-4-turbo",
|
|
614
|
+
"name": "GPT-4 Turbo",
|
|
615
|
+
"description": "Compact GPT model for low-latency assistance and high-volume workloads",
|
|
616
|
+
"family": "gpt",
|
|
617
|
+
"attachment": true,
|
|
618
|
+
"reasoning": false,
|
|
619
|
+
"tool_call": true,
|
|
620
|
+
"structured_output": false,
|
|
621
|
+
"temperature": true,
|
|
622
|
+
"knowledge": "2023-12",
|
|
623
|
+
"release_date": "2023-11-06",
|
|
624
|
+
"last_updated": "2024-04-09",
|
|
625
|
+
"modalities": {
|
|
626
|
+
"input": [
|
|
627
|
+
"text",
|
|
628
|
+
"image"
|
|
629
|
+
],
|
|
630
|
+
"output": [
|
|
631
|
+
"text"
|
|
632
|
+
]
|
|
633
|
+
},
|
|
634
|
+
"open_weights": false,
|
|
635
|
+
"limit": {
|
|
636
|
+
"context": 128000,
|
|
637
|
+
"output": 4096
|
|
638
|
+
},
|
|
639
|
+
"status": "deprecated",
|
|
584
640
|
"cost": {
|
|
585
|
-
"input":
|
|
586
|
-
"output":
|
|
587
|
-
"tiers": [
|
|
588
|
-
{
|
|
589
|
-
"input": 60,
|
|
590
|
-
"output": 270,
|
|
591
|
-
"tier": {
|
|
592
|
-
"type": "context",
|
|
593
|
-
"size": 272000
|
|
594
|
-
}
|
|
595
|
-
}
|
|
596
|
-
],
|
|
597
|
-
"context_over_200k": {
|
|
598
|
-
"input": 60,
|
|
599
|
-
"output": 270
|
|
600
|
-
}
|
|
641
|
+
"input": 10,
|
|
642
|
+
"output": 30
|
|
601
643
|
}
|
|
602
644
|
},
|
|
603
|
-
"gpt-5": {
|
|
604
|
-
"id": "gpt-5",
|
|
605
|
-
"name": "GPT-5",
|
|
606
|
-
"description": "
|
|
645
|
+
"gpt-5.1": {
|
|
646
|
+
"id": "gpt-5.1",
|
|
647
|
+
"name": "GPT-5.1",
|
|
648
|
+
"description": "Sharper GPT-5 generation for coding, product work, and tool-assisted tasks",
|
|
607
649
|
"family": "gpt",
|
|
608
650
|
"attachment": true,
|
|
609
651
|
"reasoning": true,
|
|
@@ -611,7 +653,7 @@
|
|
|
611
653
|
{
|
|
612
654
|
"type": "effort",
|
|
613
655
|
"values": [
|
|
614
|
-
"
|
|
656
|
+
"none",
|
|
615
657
|
"low",
|
|
616
658
|
"medium",
|
|
617
659
|
"high"
|
|
@@ -620,10 +662,10 @@
|
|
|
620
662
|
],
|
|
621
663
|
"tool_call": true,
|
|
622
664
|
"structured_output": true,
|
|
623
|
-
"temperature":
|
|
665
|
+
"temperature": true,
|
|
624
666
|
"knowledge": "2024-09-30",
|
|
625
|
-
"release_date": "2025-
|
|
626
|
-
"last_updated": "2025-
|
|
667
|
+
"release_date": "2025-11-13",
|
|
668
|
+
"last_updated": "2025-11-13",
|
|
627
669
|
"modalities": {
|
|
628
670
|
"input": [
|
|
629
671
|
"text",
|
|
@@ -645,32 +687,29 @@
|
|
|
645
687
|
"cache_read": 0.125
|
|
646
688
|
}
|
|
647
689
|
},
|
|
648
|
-
"
|
|
649
|
-
"id": "
|
|
650
|
-
"name": "
|
|
651
|
-
"description": "
|
|
652
|
-
"family": "
|
|
690
|
+
"o1": {
|
|
691
|
+
"id": "o1",
|
|
692
|
+
"name": "o1",
|
|
693
|
+
"description": "O-series reasoning model for hard analysis, math, coding, and planning",
|
|
694
|
+
"family": "o",
|
|
653
695
|
"attachment": true,
|
|
654
696
|
"reasoning": true,
|
|
655
697
|
"reasoning_options": [
|
|
656
698
|
{
|
|
657
699
|
"type": "effort",
|
|
658
700
|
"values": [
|
|
659
|
-
"none",
|
|
660
701
|
"low",
|
|
661
702
|
"medium",
|
|
662
|
-
"high"
|
|
663
|
-
"xhigh",
|
|
664
|
-
"max"
|
|
703
|
+
"high"
|
|
665
704
|
]
|
|
666
705
|
}
|
|
667
706
|
],
|
|
668
707
|
"tool_call": true,
|
|
669
708
|
"structured_output": true,
|
|
670
709
|
"temperature": false,
|
|
671
|
-
"knowledge": "
|
|
672
|
-
"release_date": "
|
|
673
|
-
"last_updated": "
|
|
710
|
+
"knowledge": "2023-09",
|
|
711
|
+
"release_date": "2024-12-05",
|
|
712
|
+
"last_updated": "2024-12-05",
|
|
674
713
|
"modalities": {
|
|
675
714
|
"input": [
|
|
676
715
|
"text",
|
|
@@ -683,89 +722,34 @@
|
|
|
683
722
|
},
|
|
684
723
|
"open_weights": false,
|
|
685
724
|
"limit": {
|
|
686
|
-
"context":
|
|
687
|
-
"
|
|
688
|
-
"output": 128000
|
|
689
|
-
},
|
|
690
|
-
"experimental": {
|
|
691
|
-
"modes": {
|
|
692
|
-
"fast": {
|
|
693
|
-
"cost": {
|
|
694
|
-
"input": 10,
|
|
695
|
-
"output": 60,
|
|
696
|
-
"cache_read": 1,
|
|
697
|
-
"cache_write": 12.5
|
|
698
|
-
},
|
|
699
|
-
"provider": {
|
|
700
|
-
"body": {
|
|
701
|
-
"service_tier": "priority"
|
|
702
|
-
}
|
|
703
|
-
}
|
|
704
|
-
},
|
|
705
|
-
"pro": {
|
|
706
|
-
"provider": {
|
|
707
|
-
"body": {
|
|
708
|
-
"reasoning": {
|
|
709
|
-
"mode": "pro"
|
|
710
|
-
}
|
|
711
|
-
}
|
|
712
|
-
}
|
|
713
|
-
}
|
|
714
|
-
}
|
|
725
|
+
"context": 200000,
|
|
726
|
+
"output": 100000
|
|
715
727
|
},
|
|
728
|
+
"status": "deprecated",
|
|
716
729
|
"cost": {
|
|
717
|
-
"input":
|
|
718
|
-
"output":
|
|
719
|
-
"cache_read":
|
|
720
|
-
"cache_write": 6.25,
|
|
721
|
-
"tiers": [
|
|
722
|
-
{
|
|
723
|
-
"input": 10,
|
|
724
|
-
"output": 45,
|
|
725
|
-
"cache_read": 1,
|
|
726
|
-
"cache_write": 12.5,
|
|
727
|
-
"tier": {
|
|
728
|
-
"type": "context",
|
|
729
|
-
"size": 272000
|
|
730
|
-
}
|
|
731
|
-
}
|
|
732
|
-
],
|
|
733
|
-
"context_over_200k": {
|
|
734
|
-
"input": 10,
|
|
735
|
-
"output": 45,
|
|
736
|
-
"cache_read": 1,
|
|
737
|
-
"cache_write": 12.5
|
|
738
|
-
}
|
|
730
|
+
"input": 15,
|
|
731
|
+
"output": 60,
|
|
732
|
+
"cache_read": 7.5
|
|
739
733
|
}
|
|
740
734
|
},
|
|
741
|
-
"gpt-
|
|
742
|
-
"id": "gpt-
|
|
743
|
-
"name": "GPT-
|
|
744
|
-
"description": "
|
|
745
|
-
"family": "gpt
|
|
735
|
+
"gpt-4o": {
|
|
736
|
+
"id": "gpt-4o",
|
|
737
|
+
"name": "GPT-4o",
|
|
738
|
+
"description": "Omni-era GPT for multimodal chat, practical coding, and general assistants",
|
|
739
|
+
"family": "gpt",
|
|
746
740
|
"attachment": true,
|
|
747
|
-
"reasoning":
|
|
748
|
-
"reasoning_options": [
|
|
749
|
-
{
|
|
750
|
-
"type": "effort",
|
|
751
|
-
"values": [
|
|
752
|
-
"minimal",
|
|
753
|
-
"low",
|
|
754
|
-
"medium",
|
|
755
|
-
"high"
|
|
756
|
-
]
|
|
757
|
-
}
|
|
758
|
-
],
|
|
741
|
+
"reasoning": false,
|
|
759
742
|
"tool_call": true,
|
|
760
743
|
"structured_output": true,
|
|
761
|
-
"temperature":
|
|
762
|
-
"knowledge": "
|
|
763
|
-
"release_date": "
|
|
764
|
-
"last_updated": "
|
|
744
|
+
"temperature": true,
|
|
745
|
+
"knowledge": "2023-09",
|
|
746
|
+
"release_date": "2024-05-13",
|
|
747
|
+
"last_updated": "2024-08-06",
|
|
765
748
|
"modalities": {
|
|
766
749
|
"input": [
|
|
767
750
|
"text",
|
|
768
|
-
"image"
|
|
751
|
+
"image",
|
|
752
|
+
"pdf"
|
|
769
753
|
],
|
|
770
754
|
"output": [
|
|
771
755
|
"text"
|
|
@@ -773,14 +757,13 @@
|
|
|
773
757
|
},
|
|
774
758
|
"open_weights": false,
|
|
775
759
|
"limit": {
|
|
776
|
-
"context":
|
|
777
|
-
"
|
|
778
|
-
"output": 128000
|
|
760
|
+
"context": 128000,
|
|
761
|
+
"output": 16384
|
|
779
762
|
},
|
|
780
763
|
"cost": {
|
|
781
|
-
"input":
|
|
782
|
-
"output":
|
|
783
|
-
"cache_read":
|
|
764
|
+
"input": 2.5,
|
|
765
|
+
"output": 10,
|
|
766
|
+
"cache_read": 1.25
|
|
784
767
|
}
|
|
785
768
|
},
|
|
786
769
|
"gpt-5.6-luna": {
|
|
@@ -876,36 +859,6 @@
|
|
|
876
859
|
}
|
|
877
860
|
}
|
|
878
861
|
},
|
|
879
|
-
"text-embedding-ada-002": {
|
|
880
|
-
"id": "text-embedding-ada-002",
|
|
881
|
-
"name": "text-embedding-ada-002",
|
|
882
|
-
"description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
|
|
883
|
-
"family": "text-embedding",
|
|
884
|
-
"attachment": false,
|
|
885
|
-
"reasoning": false,
|
|
886
|
-
"tool_call": false,
|
|
887
|
-
"temperature": false,
|
|
888
|
-
"knowledge": "2022-12",
|
|
889
|
-
"release_date": "2022-12-15",
|
|
890
|
-
"last_updated": "2022-12-15",
|
|
891
|
-
"modalities": {
|
|
892
|
-
"input": [
|
|
893
|
-
"text"
|
|
894
|
-
],
|
|
895
|
-
"output": [
|
|
896
|
-
"text"
|
|
897
|
-
]
|
|
898
|
-
},
|
|
899
|
-
"open_weights": false,
|
|
900
|
-
"limit": {
|
|
901
|
-
"context": 8192,
|
|
902
|
-
"output": 1536
|
|
903
|
-
},
|
|
904
|
-
"cost": {
|
|
905
|
-
"input": 0.1,
|
|
906
|
-
"output": 0
|
|
907
|
-
}
|
|
908
|
-
},
|
|
909
862
|
"gpt-5.3-codex": {
|
|
910
863
|
"id": "gpt-5.3-codex",
|
|
911
864
|
"name": "GPT-5.3 Codex",
|
|
@@ -927,7 +880,7 @@
|
|
|
927
880
|
],
|
|
928
881
|
"tool_call": true,
|
|
929
882
|
"structured_output": true,
|
|
930
|
-
"temperature":
|
|
883
|
+
"temperature": true,
|
|
931
884
|
"knowledge": "2025-08-31",
|
|
932
885
|
"release_date": "2026-02-05",
|
|
933
886
|
"last_updated": "2026-02-05",
|
|
@@ -953,31 +906,19 @@
|
|
|
953
906
|
"cache_read": 0.175
|
|
954
907
|
}
|
|
955
908
|
},
|
|
956
|
-
"gpt-
|
|
957
|
-
"id": "gpt-
|
|
958
|
-
"name": "GPT-
|
|
959
|
-
"description": "
|
|
960
|
-
"family": "gpt-
|
|
909
|
+
"gpt-4o-mini": {
|
|
910
|
+
"id": "gpt-4o-mini",
|
|
911
|
+
"name": "GPT-4o mini",
|
|
912
|
+
"description": "Small omni GPT for cheap multimodal assistance and production-scale traffic",
|
|
913
|
+
"family": "gpt-mini",
|
|
961
914
|
"attachment": true,
|
|
962
|
-
"reasoning":
|
|
963
|
-
"reasoning_options": [
|
|
964
|
-
{
|
|
965
|
-
"type": "effort",
|
|
966
|
-
"values": [
|
|
967
|
-
"none",
|
|
968
|
-
"low",
|
|
969
|
-
"medium",
|
|
970
|
-
"high",
|
|
971
|
-
"xhigh"
|
|
972
|
-
]
|
|
973
|
-
}
|
|
974
|
-
],
|
|
915
|
+
"reasoning": false,
|
|
975
916
|
"tool_call": true,
|
|
976
917
|
"structured_output": true,
|
|
977
|
-
"temperature":
|
|
978
|
-
"knowledge": "
|
|
979
|
-
"release_date": "
|
|
980
|
-
"last_updated": "
|
|
918
|
+
"temperature": true,
|
|
919
|
+
"knowledge": "2023-09",
|
|
920
|
+
"release_date": "2024-07-18",
|
|
921
|
+
"last_updated": "2024-07-18",
|
|
981
922
|
"modalities": {
|
|
982
923
|
"input": [
|
|
983
924
|
"text",
|
|
@@ -991,54 +932,47 @@
|
|
|
991
932
|
"open_weights": false,
|
|
992
933
|
"limit": {
|
|
993
934
|
"context": 128000,
|
|
994
|
-
"
|
|
995
|
-
"output": 32000
|
|
935
|
+
"output": 16384
|
|
996
936
|
},
|
|
997
937
|
"cost": {
|
|
998
|
-
"input":
|
|
999
|
-
"output":
|
|
1000
|
-
"cache_read": 0.
|
|
938
|
+
"input": 0.15,
|
|
939
|
+
"output": 0.6,
|
|
940
|
+
"cache_read": 0.075
|
|
1001
941
|
}
|
|
1002
942
|
},
|
|
1003
|
-
"gpt-
|
|
1004
|
-
"id": "gpt-
|
|
1005
|
-
"name": "
|
|
1006
|
-
"description": "
|
|
1007
|
-
"family": "gpt-
|
|
943
|
+
"gpt-image-1.5": {
|
|
944
|
+
"id": "gpt-image-1.5",
|
|
945
|
+
"name": "gpt-image-1.5",
|
|
946
|
+
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
947
|
+
"family": "gpt-image",
|
|
1008
948
|
"attachment": true,
|
|
1009
949
|
"reasoning": false,
|
|
1010
|
-
"tool_call":
|
|
1011
|
-
"
|
|
1012
|
-
"
|
|
1013
|
-
"
|
|
1014
|
-
"release_date": "2025-04-14",
|
|
1015
|
-
"last_updated": "2025-04-14",
|
|
950
|
+
"tool_call": false,
|
|
951
|
+
"temperature": false,
|
|
952
|
+
"release_date": "2025-11-25",
|
|
953
|
+
"last_updated": "2025-11-25",
|
|
1016
954
|
"modalities": {
|
|
1017
955
|
"input": [
|
|
1018
956
|
"text",
|
|
1019
957
|
"image"
|
|
1020
958
|
],
|
|
1021
959
|
"output": [
|
|
1022
|
-
"text"
|
|
960
|
+
"text",
|
|
961
|
+
"image"
|
|
1023
962
|
]
|
|
1024
963
|
},
|
|
1025
964
|
"open_weights": false,
|
|
1026
965
|
"limit": {
|
|
1027
|
-
"context":
|
|
1028
|
-
"
|
|
1029
|
-
|
|
1030
|
-
"status": "deprecated",
|
|
1031
|
-
"cost": {
|
|
1032
|
-
"input": 0.1,
|
|
1033
|
-
"output": 0.4,
|
|
1034
|
-
"cache_read": 0.025
|
|
966
|
+
"context": 0,
|
|
967
|
+
"input": 0,
|
|
968
|
+
"output": 0
|
|
1035
969
|
}
|
|
1036
970
|
},
|
|
1037
|
-
"o1": {
|
|
1038
|
-
"id": "o1",
|
|
1039
|
-
"name": "o1",
|
|
971
|
+
"o1-pro": {
|
|
972
|
+
"id": "o1-pro",
|
|
973
|
+
"name": "o1-pro",
|
|
1040
974
|
"description": "O-series reasoning model for hard analysis, math, coding, and planning",
|
|
1041
|
-
"family": "o",
|
|
975
|
+
"family": "o-pro",
|
|
1042
976
|
"attachment": true,
|
|
1043
977
|
"reasoning": true,
|
|
1044
978
|
"reasoning_options": [
|
|
@@ -1055,13 +989,12 @@
|
|
|
1055
989
|
"structured_output": true,
|
|
1056
990
|
"temperature": false,
|
|
1057
991
|
"knowledge": "2023-09",
|
|
1058
|
-
"release_date": "
|
|
1059
|
-
"last_updated": "
|
|
992
|
+
"release_date": "2025-03-19",
|
|
993
|
+
"last_updated": "2025-03-19",
|
|
1060
994
|
"modalities": {
|
|
1061
995
|
"input": [
|
|
1062
996
|
"text",
|
|
1063
|
-
"image"
|
|
1064
|
-
"pdf"
|
|
997
|
+
"image"
|
|
1065
998
|
],
|
|
1066
999
|
"output": [
|
|
1067
1000
|
"text"
|
|
@@ -1074,40 +1007,28 @@
|
|
|
1074
1007
|
},
|
|
1075
1008
|
"status": "deprecated",
|
|
1076
1009
|
"cost": {
|
|
1077
|
-
"input":
|
|
1078
|
-
"output":
|
|
1079
|
-
"cache_read": 7.5
|
|
1010
|
+
"input": 150,
|
|
1011
|
+
"output": 600
|
|
1080
1012
|
}
|
|
1081
1013
|
},
|
|
1082
|
-
"gpt-
|
|
1083
|
-
"id": "gpt-
|
|
1084
|
-
"name": "GPT-
|
|
1085
|
-
"description": "
|
|
1014
|
+
"gpt-4.1": {
|
|
1015
|
+
"id": "gpt-4.1",
|
|
1016
|
+
"name": "GPT-4.1",
|
|
1017
|
+
"description": "Long-lived GPT workhorse for coding, instruction following, and production apps",
|
|
1086
1018
|
"family": "gpt",
|
|
1087
1019
|
"attachment": true,
|
|
1088
|
-
"reasoning":
|
|
1089
|
-
"reasoning_options": [
|
|
1090
|
-
{
|
|
1091
|
-
"type": "effort",
|
|
1092
|
-
"values": [
|
|
1093
|
-
"none",
|
|
1094
|
-
"low",
|
|
1095
|
-
"medium",
|
|
1096
|
-
"high",
|
|
1097
|
-
"xhigh"
|
|
1098
|
-
]
|
|
1099
|
-
}
|
|
1100
|
-
],
|
|
1020
|
+
"reasoning": false,
|
|
1101
1021
|
"tool_call": true,
|
|
1102
1022
|
"structured_output": true,
|
|
1103
|
-
"temperature":
|
|
1104
|
-
"knowledge": "
|
|
1105
|
-
"release_date": "2025-
|
|
1106
|
-
"last_updated": "2025-
|
|
1023
|
+
"temperature": true,
|
|
1024
|
+
"knowledge": "2024-04",
|
|
1025
|
+
"release_date": "2025-04-14",
|
|
1026
|
+
"last_updated": "2025-04-14",
|
|
1107
1027
|
"modalities": {
|
|
1108
1028
|
"input": [
|
|
1109
1029
|
"text",
|
|
1110
|
-
"image"
|
|
1030
|
+
"image",
|
|
1031
|
+
"pdf"
|
|
1111
1032
|
],
|
|
1112
1033
|
"output": [
|
|
1113
1034
|
"text"
|
|
@@ -1115,14 +1036,43 @@
|
|
|
1115
1036
|
},
|
|
1116
1037
|
"open_weights": false,
|
|
1117
1038
|
"limit": {
|
|
1118
|
-
"context":
|
|
1119
|
-
"
|
|
1120
|
-
"output": 128000
|
|
1039
|
+
"context": 1047576,
|
|
1040
|
+
"output": 32768
|
|
1121
1041
|
},
|
|
1122
1042
|
"cost": {
|
|
1123
|
-
"input":
|
|
1124
|
-
"output":
|
|
1125
|
-
"cache_read": 0.
|
|
1043
|
+
"input": 2,
|
|
1044
|
+
"output": 8,
|
|
1045
|
+
"cache_read": 0.5
|
|
1046
|
+
}
|
|
1047
|
+
},
|
|
1048
|
+
"text-embedding-ada-002": {
|
|
1049
|
+
"id": "text-embedding-ada-002",
|
|
1050
|
+
"name": "text-embedding-ada-002",
|
|
1051
|
+
"description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
|
|
1052
|
+
"family": "text-embedding",
|
|
1053
|
+
"attachment": false,
|
|
1054
|
+
"reasoning": false,
|
|
1055
|
+
"tool_call": false,
|
|
1056
|
+
"temperature": false,
|
|
1057
|
+
"knowledge": "2022-12",
|
|
1058
|
+
"release_date": "2022-12-15",
|
|
1059
|
+
"last_updated": "2022-12-15",
|
|
1060
|
+
"modalities": {
|
|
1061
|
+
"input": [
|
|
1062
|
+
"text"
|
|
1063
|
+
],
|
|
1064
|
+
"output": [
|
|
1065
|
+
"text"
|
|
1066
|
+
]
|
|
1067
|
+
},
|
|
1068
|
+
"open_weights": false,
|
|
1069
|
+
"limit": {
|
|
1070
|
+
"context": 8192,
|
|
1071
|
+
"output": 1536
|
|
1072
|
+
},
|
|
1073
|
+
"cost": {
|
|
1074
|
+
"input": 0.1,
|
|
1075
|
+
"output": 0
|
|
1126
1076
|
}
|
|
1127
1077
|
},
|
|
1128
1078
|
"gpt-image-1": {
|
|
@@ -1153,18 +1103,18 @@
|
|
|
1153
1103
|
},
|
|
1154
1104
|
"status": "deprecated"
|
|
1155
1105
|
},
|
|
1156
|
-
"gpt-
|
|
1157
|
-
"id": "gpt-
|
|
1158
|
-
"name": "GPT-
|
|
1159
|
-
"description": "
|
|
1160
|
-
"family": "gpt",
|
|
1106
|
+
"gpt-5.4-nano": {
|
|
1107
|
+
"id": "gpt-5.4-nano",
|
|
1108
|
+
"name": "GPT-5.4 nano",
|
|
1109
|
+
"description": "Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation",
|
|
1110
|
+
"family": "gpt-nano",
|
|
1161
1111
|
"attachment": true,
|
|
1162
1112
|
"reasoning": true,
|
|
1163
1113
|
"reasoning_options": [
|
|
1164
1114
|
{
|
|
1165
1115
|
"type": "effort",
|
|
1166
1116
|
"values": [
|
|
1167
|
-
"
|
|
1117
|
+
"none",
|
|
1168
1118
|
"low",
|
|
1169
1119
|
"medium",
|
|
1170
1120
|
"high",
|
|
@@ -1173,34 +1123,116 @@
|
|
|
1173
1123
|
}
|
|
1174
1124
|
],
|
|
1175
1125
|
"tool_call": true,
|
|
1176
|
-
"structured_output":
|
|
1177
|
-
"temperature":
|
|
1178
|
-
"knowledge": "
|
|
1179
|
-
"release_date": "2026-
|
|
1180
|
-
"last_updated": "2026-
|
|
1126
|
+
"structured_output": true,
|
|
1127
|
+
"temperature": true,
|
|
1128
|
+
"knowledge": "2025-08-31",
|
|
1129
|
+
"release_date": "2026-03-17",
|
|
1130
|
+
"last_updated": "2026-03-17",
|
|
1181
1131
|
"modalities": {
|
|
1182
1132
|
"input": [
|
|
1183
1133
|
"text",
|
|
1184
|
-
"audio",
|
|
1185
1134
|
"image"
|
|
1186
1135
|
],
|
|
1187
1136
|
"output": [
|
|
1137
|
+
"text"
|
|
1138
|
+
]
|
|
1139
|
+
},
|
|
1140
|
+
"open_weights": false,
|
|
1141
|
+
"limit": {
|
|
1142
|
+
"context": 400000,
|
|
1143
|
+
"input": 272000,
|
|
1144
|
+
"output": 128000
|
|
1145
|
+
},
|
|
1146
|
+
"cost": {
|
|
1147
|
+
"input": 0.2,
|
|
1148
|
+
"output": 1.25,
|
|
1149
|
+
"cache_read": 0.02
|
|
1150
|
+
}
|
|
1151
|
+
},
|
|
1152
|
+
"gpt-5.5-pro": {
|
|
1153
|
+
"id": "gpt-5.5-pro",
|
|
1154
|
+
"name": "GPT-5.5 Pro",
|
|
1155
|
+
"description": "Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding",
|
|
1156
|
+
"family": "gpt-pro",
|
|
1157
|
+
"attachment": true,
|
|
1158
|
+
"reasoning": true,
|
|
1159
|
+
"reasoning_options": [
|
|
1160
|
+
{
|
|
1161
|
+
"type": "effort",
|
|
1162
|
+
"values": [
|
|
1163
|
+
"medium",
|
|
1164
|
+
"high",
|
|
1165
|
+
"xhigh"
|
|
1166
|
+
]
|
|
1167
|
+
}
|
|
1168
|
+
],
|
|
1169
|
+
"tool_call": true,
|
|
1170
|
+
"structured_output": true,
|
|
1171
|
+
"temperature": false,
|
|
1172
|
+
"knowledge": "2025-12-01",
|
|
1173
|
+
"release_date": "2026-04-23",
|
|
1174
|
+
"last_updated": "2026-04-23",
|
|
1175
|
+
"modalities": {
|
|
1176
|
+
"input": [
|
|
1188
1177
|
"text",
|
|
1189
|
-
"
|
|
1178
|
+
"image",
|
|
1179
|
+
"pdf"
|
|
1180
|
+
],
|
|
1181
|
+
"output": [
|
|
1182
|
+
"text"
|
|
1190
1183
|
]
|
|
1191
1184
|
},
|
|
1192
1185
|
"open_weights": false,
|
|
1193
1186
|
"limit": {
|
|
1194
|
-
"context":
|
|
1195
|
-
"input":
|
|
1196
|
-
"output":
|
|
1187
|
+
"context": 1050000,
|
|
1188
|
+
"input": 922000,
|
|
1189
|
+
"output": 128000
|
|
1197
1190
|
},
|
|
1198
1191
|
"cost": {
|
|
1199
|
-
"input":
|
|
1200
|
-
"output":
|
|
1201
|
-
"
|
|
1202
|
-
|
|
1203
|
-
|
|
1192
|
+
"input": 30,
|
|
1193
|
+
"output": 180,
|
|
1194
|
+
"tiers": [
|
|
1195
|
+
{
|
|
1196
|
+
"input": 60,
|
|
1197
|
+
"output": 270,
|
|
1198
|
+
"tier": {
|
|
1199
|
+
"type": "context",
|
|
1200
|
+
"size": 272000
|
|
1201
|
+
}
|
|
1202
|
+
}
|
|
1203
|
+
],
|
|
1204
|
+
"context_over_200k": {
|
|
1205
|
+
"input": 60,
|
|
1206
|
+
"output": 270
|
|
1207
|
+
}
|
|
1208
|
+
}
|
|
1209
|
+
},
|
|
1210
|
+
"gpt-image-1-mini": {
|
|
1211
|
+
"id": "gpt-image-1-mini",
|
|
1212
|
+
"name": "gpt-image-1-mini",
|
|
1213
|
+
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
1214
|
+
"family": "gpt-image",
|
|
1215
|
+
"attachment": true,
|
|
1216
|
+
"reasoning": false,
|
|
1217
|
+
"tool_call": false,
|
|
1218
|
+
"temperature": false,
|
|
1219
|
+
"release_date": "2025-09-26",
|
|
1220
|
+
"last_updated": "2025-09-26",
|
|
1221
|
+
"modalities": {
|
|
1222
|
+
"input": [
|
|
1223
|
+
"text",
|
|
1224
|
+
"image"
|
|
1225
|
+
],
|
|
1226
|
+
"output": [
|
|
1227
|
+
"text",
|
|
1228
|
+
"image"
|
|
1229
|
+
]
|
|
1230
|
+
},
|
|
1231
|
+
"open_weights": false,
|
|
1232
|
+
"limit": {
|
|
1233
|
+
"context": 0,
|
|
1234
|
+
"input": 0,
|
|
1235
|
+
"output": 0
|
|
1204
1236
|
}
|
|
1205
1237
|
},
|
|
1206
1238
|
"gpt-5.4-mini": {
|
|
@@ -1224,7 +1256,7 @@
|
|
|
1224
1256
|
],
|
|
1225
1257
|
"tool_call": true,
|
|
1226
1258
|
"structured_output": true,
|
|
1227
|
-
"temperature":
|
|
1259
|
+
"temperature": true,
|
|
1228
1260
|
"knowledge": "2025-08-31",
|
|
1229
1261
|
"release_date": "2026-03-17",
|
|
1230
1262
|
"last_updated": "2026-03-17",
|
|
@@ -1265,23 +1297,54 @@
|
|
|
1265
1297
|
"cache_read": 0.075
|
|
1266
1298
|
}
|
|
1267
1299
|
},
|
|
1268
|
-
"gpt-
|
|
1269
|
-
"id": "gpt-
|
|
1270
|
-
"name": "
|
|
1271
|
-
"description": "
|
|
1300
|
+
"gpt-image-2": {
|
|
1301
|
+
"id": "gpt-image-2",
|
|
1302
|
+
"name": "gpt-image-2",
|
|
1303
|
+
"description": "Image model for prompt-driven generation, editing, and visual design workflows",
|
|
1304
|
+
"family": "gpt-image",
|
|
1305
|
+
"attachment": true,
|
|
1306
|
+
"reasoning": false,
|
|
1307
|
+
"tool_call": false,
|
|
1308
|
+
"temperature": false,
|
|
1309
|
+
"release_date": "2026-04-21",
|
|
1310
|
+
"last_updated": "2026-04-21",
|
|
1311
|
+
"modalities": {
|
|
1312
|
+
"input": [
|
|
1313
|
+
"text",
|
|
1314
|
+
"image"
|
|
1315
|
+
],
|
|
1316
|
+
"output": [
|
|
1317
|
+
"image"
|
|
1318
|
+
]
|
|
1319
|
+
},
|
|
1320
|
+
"open_weights": false,
|
|
1321
|
+
"limit": {
|
|
1322
|
+
"context": 0,
|
|
1323
|
+
"input": 0,
|
|
1324
|
+
"output": 0
|
|
1325
|
+
},
|
|
1326
|
+
"cost": {
|
|
1327
|
+
"input": 5,
|
|
1328
|
+
"output": 30,
|
|
1329
|
+
"cache_read": 1.25
|
|
1330
|
+
}
|
|
1331
|
+
},
|
|
1332
|
+
"gpt-3.5-turbo": {
|
|
1333
|
+
"id": "gpt-3.5-turbo",
|
|
1334
|
+
"name": "GPT-3.5-turbo",
|
|
1335
|
+
"description": "Compact GPT model for low-latency assistance and high-volume workloads",
|
|
1272
1336
|
"family": "gpt",
|
|
1273
|
-
"attachment":
|
|
1337
|
+
"attachment": false,
|
|
1274
1338
|
"reasoning": false,
|
|
1275
|
-
"tool_call":
|
|
1276
|
-
"structured_output":
|
|
1339
|
+
"tool_call": false,
|
|
1340
|
+
"structured_output": false,
|
|
1277
1341
|
"temperature": true,
|
|
1278
|
-
"knowledge": "
|
|
1279
|
-
"release_date": "
|
|
1280
|
-
"last_updated": "
|
|
1342
|
+
"knowledge": "2021-09-01",
|
|
1343
|
+
"release_date": "2023-03-01",
|
|
1344
|
+
"last_updated": "2023-11-06",
|
|
1281
1345
|
"modalities": {
|
|
1282
1346
|
"input": [
|
|
1283
|
-
"text"
|
|
1284
|
-
"image"
|
|
1347
|
+
"text"
|
|
1285
1348
|
],
|
|
1286
1349
|
"output": [
|
|
1287
1350
|
"text"
|
|
@@ -1289,38 +1352,42 @@
|
|
|
1289
1352
|
},
|
|
1290
1353
|
"open_weights": false,
|
|
1291
1354
|
"limit": {
|
|
1292
|
-
"context":
|
|
1355
|
+
"context": 16385,
|
|
1293
1356
|
"output": 4096
|
|
1294
1357
|
},
|
|
1295
1358
|
"status": "deprecated",
|
|
1296
1359
|
"cost": {
|
|
1297
|
-
"input": 5,
|
|
1298
|
-
"output":
|
|
1360
|
+
"input": 0.5,
|
|
1361
|
+
"output": 1.5,
|
|
1362
|
+
"cache_read": 0
|
|
1299
1363
|
}
|
|
1300
1364
|
},
|
|
1301
|
-
"
|
|
1302
|
-
"id": "
|
|
1303
|
-
"name": "
|
|
1304
|
-
"description": "
|
|
1305
|
-
"family": "
|
|
1365
|
+
"gpt-5.6": {
|
|
1366
|
+
"id": "gpt-5.6",
|
|
1367
|
+
"name": "GPT-5.6",
|
|
1368
|
+
"description": "Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows",
|
|
1369
|
+
"family": "gpt-sol",
|
|
1306
1370
|
"attachment": true,
|
|
1307
1371
|
"reasoning": true,
|
|
1308
1372
|
"reasoning_options": [
|
|
1309
1373
|
{
|
|
1310
1374
|
"type": "effort",
|
|
1311
1375
|
"values": [
|
|
1376
|
+
"none",
|
|
1312
1377
|
"low",
|
|
1313
1378
|
"medium",
|
|
1314
|
-
"high"
|
|
1379
|
+
"high",
|
|
1380
|
+
"xhigh",
|
|
1381
|
+
"max"
|
|
1315
1382
|
]
|
|
1316
1383
|
}
|
|
1317
1384
|
],
|
|
1318
1385
|
"tool_call": true,
|
|
1319
1386
|
"structured_output": true,
|
|
1320
1387
|
"temperature": false,
|
|
1321
|
-
"knowledge": "
|
|
1322
|
-
"release_date": "
|
|
1323
|
-
"last_updated": "
|
|
1388
|
+
"knowledge": "2026-02-16",
|
|
1389
|
+
"release_date": "2026-07-09",
|
|
1390
|
+
"last_updated": "2026-07-09",
|
|
1324
1391
|
"modalities": {
|
|
1325
1392
|
"input": [
|
|
1326
1393
|
"text",
|
|
@@ -1333,41 +1400,59 @@
|
|
|
1333
1400
|
},
|
|
1334
1401
|
"open_weights": false,
|
|
1335
1402
|
"limit": {
|
|
1336
|
-
"context":
|
|
1337
|
-
"
|
|
1403
|
+
"context": 1050000,
|
|
1404
|
+
"input": 922000,
|
|
1405
|
+
"output": 128000
|
|
1406
|
+
},
|
|
1407
|
+
"experimental": {
|
|
1408
|
+
"modes": {
|
|
1409
|
+
"fast": {
|
|
1410
|
+
"cost": {
|
|
1411
|
+
"input": 8,
|
|
1412
|
+
"output": 40,
|
|
1413
|
+
"cache_read": 0.8,
|
|
1414
|
+
"cache_write": 10
|
|
1415
|
+
},
|
|
1416
|
+
"provider": {
|
|
1417
|
+
"body": {
|
|
1418
|
+
"service_tier": "priority"
|
|
1419
|
+
}
|
|
1420
|
+
}
|
|
1421
|
+
},
|
|
1422
|
+
"pro": {
|
|
1423
|
+
"provider": {
|
|
1424
|
+
"body": {
|
|
1425
|
+
"reasoning": {
|
|
1426
|
+
"mode": "pro"
|
|
1427
|
+
}
|
|
1428
|
+
}
|
|
1429
|
+
}
|
|
1430
|
+
}
|
|
1431
|
+
}
|
|
1338
1432
|
},
|
|
1339
1433
|
"cost": {
|
|
1340
|
-
"input":
|
|
1341
|
-
"output":
|
|
1342
|
-
"cache_read": 0.
|
|
1343
|
-
|
|
1344
|
-
|
|
1345
|
-
|
|
1346
|
-
|
|
1347
|
-
|
|
1348
|
-
|
|
1349
|
-
|
|
1350
|
-
|
|
1351
|
-
|
|
1352
|
-
|
|
1353
|
-
|
|
1354
|
-
|
|
1355
|
-
"last_updated": "2025-12-16",
|
|
1356
|
-
"modalities": {
|
|
1357
|
-
"input": [
|
|
1358
|
-
"text",
|
|
1359
|
-
"image"
|
|
1434
|
+
"input": 4,
|
|
1435
|
+
"output": 20,
|
|
1436
|
+
"cache_read": 0.4,
|
|
1437
|
+
"cache_write": 5,
|
|
1438
|
+
"tiers": [
|
|
1439
|
+
{
|
|
1440
|
+
"input": 8,
|
|
1441
|
+
"output": 30,
|
|
1442
|
+
"cache_read": 0.8,
|
|
1443
|
+
"cache_write": 10,
|
|
1444
|
+
"tier": {
|
|
1445
|
+
"type": "context",
|
|
1446
|
+
"size": 272000
|
|
1447
|
+
}
|
|
1448
|
+
}
|
|
1360
1449
|
],
|
|
1361
|
-
"
|
|
1362
|
-
"
|
|
1363
|
-
"
|
|
1364
|
-
|
|
1365
|
-
|
|
1366
|
-
|
|
1367
|
-
"limit": {
|
|
1368
|
-
"context": 0,
|
|
1369
|
-
"input": 0,
|
|
1370
|
-
"output": 0
|
|
1450
|
+
"context_over_200k": {
|
|
1451
|
+
"input": 8,
|
|
1452
|
+
"output": 30,
|
|
1453
|
+
"cache_read": 0.8,
|
|
1454
|
+
"cache_write": 10
|
|
1455
|
+
}
|
|
1371
1456
|
}
|
|
1372
1457
|
},
|
|
1373
1458
|
"text-embedding-3-small": {
|
|
@@ -1400,17 +1485,20 @@
|
|
|
1400
1485
|
"output": 0
|
|
1401
1486
|
}
|
|
1402
1487
|
},
|
|
1403
|
-
"gpt-5-
|
|
1404
|
-
"id": "gpt-5-
|
|
1405
|
-
"name": "GPT-5
|
|
1406
|
-
"description": "
|
|
1407
|
-
"family": "gpt-
|
|
1488
|
+
"gpt-5-mini": {
|
|
1489
|
+
"id": "gpt-5-mini",
|
|
1490
|
+
"name": "GPT-5 Mini",
|
|
1491
|
+
"description": "Small GPT-5 for responsive agents, coding help, and everyday automation",
|
|
1492
|
+
"family": "gpt-mini",
|
|
1408
1493
|
"attachment": true,
|
|
1409
1494
|
"reasoning": true,
|
|
1410
1495
|
"reasoning_options": [
|
|
1411
1496
|
{
|
|
1412
1497
|
"type": "effort",
|
|
1413
1498
|
"values": [
|
|
1499
|
+
"minimal",
|
|
1500
|
+
"low",
|
|
1501
|
+
"medium",
|
|
1414
1502
|
"high"
|
|
1415
1503
|
]
|
|
1416
1504
|
}
|
|
@@ -1418,9 +1506,9 @@
|
|
|
1418
1506
|
"tool_call": true,
|
|
1419
1507
|
"structured_output": true,
|
|
1420
1508
|
"temperature": false,
|
|
1421
|
-
"knowledge": "2024-
|
|
1422
|
-
"release_date": "2025-
|
|
1423
|
-
"last_updated": "2025-
|
|
1509
|
+
"knowledge": "2024-05-30",
|
|
1510
|
+
"release_date": "2025-08-07",
|
|
1511
|
+
"last_updated": "2025-08-07",
|
|
1424
1512
|
"modalities": {
|
|
1425
1513
|
"input": [
|
|
1426
1514
|
"text",
|
|
@@ -1434,26 +1522,25 @@
|
|
|
1434
1522
|
"limit": {
|
|
1435
1523
|
"context": 400000,
|
|
1436
1524
|
"input": 272000,
|
|
1437
|
-
"output":
|
|
1525
|
+
"output": 128000
|
|
1438
1526
|
},
|
|
1439
1527
|
"cost": {
|
|
1440
|
-
"input":
|
|
1441
|
-
"output":
|
|
1528
|
+
"input": 0.25,
|
|
1529
|
+
"output": 2,
|
|
1530
|
+
"cache_read": 0.025
|
|
1442
1531
|
}
|
|
1443
1532
|
},
|
|
1444
|
-
"gpt-5.
|
|
1445
|
-
"id": "gpt-5.
|
|
1446
|
-
"name": "GPT-5.
|
|
1447
|
-
"description": "
|
|
1448
|
-
"family": "gpt",
|
|
1533
|
+
"gpt-5.4-pro": {
|
|
1534
|
+
"id": "gpt-5.4-pro",
|
|
1535
|
+
"name": "GPT-5.4 Pro",
|
|
1536
|
+
"description": "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks",
|
|
1537
|
+
"family": "gpt-pro",
|
|
1449
1538
|
"attachment": true,
|
|
1450
1539
|
"reasoning": true,
|
|
1451
1540
|
"reasoning_options": [
|
|
1452
1541
|
{
|
|
1453
1542
|
"type": "effort",
|
|
1454
1543
|
"values": [
|
|
1455
|
-
"none",
|
|
1456
|
-
"low",
|
|
1457
1544
|
"medium",
|
|
1458
1545
|
"high",
|
|
1459
1546
|
"xhigh"
|
|
@@ -1461,16 +1548,15 @@
|
|
|
1461
1548
|
}
|
|
1462
1549
|
],
|
|
1463
1550
|
"tool_call": true,
|
|
1464
|
-
"structured_output":
|
|
1551
|
+
"structured_output": false,
|
|
1465
1552
|
"temperature": false,
|
|
1466
|
-
"knowledge": "2025-
|
|
1467
|
-
"release_date": "2026-
|
|
1468
|
-
"last_updated": "2026-
|
|
1553
|
+
"knowledge": "2025-08-31",
|
|
1554
|
+
"release_date": "2026-03-05",
|
|
1555
|
+
"last_updated": "2026-03-05",
|
|
1469
1556
|
"modalities": {
|
|
1470
1557
|
"input": [
|
|
1471
1558
|
"text",
|
|
1472
|
-
"image"
|
|
1473
|
-
"pdf"
|
|
1559
|
+
"image"
|
|
1474
1560
|
],
|
|
1475
1561
|
"output": [
|
|
1476
1562
|
"text"
|
|
@@ -1482,31 +1568,13 @@
|
|
|
1482
1568
|
"input": 922000,
|
|
1483
1569
|
"output": 128000
|
|
1484
1570
|
},
|
|
1485
|
-
"experimental": {
|
|
1486
|
-
"modes": {
|
|
1487
|
-
"fast": {
|
|
1488
|
-
"cost": {
|
|
1489
|
-
"input": 12.5,
|
|
1490
|
-
"output": 75,
|
|
1491
|
-
"cache_read": 1.25
|
|
1492
|
-
},
|
|
1493
|
-
"provider": {
|
|
1494
|
-
"body": {
|
|
1495
|
-
"service_tier": "priority"
|
|
1496
|
-
}
|
|
1497
|
-
}
|
|
1498
|
-
}
|
|
1499
|
-
}
|
|
1500
|
-
},
|
|
1501
1571
|
"cost": {
|
|
1502
|
-
"input":
|
|
1503
|
-
"output":
|
|
1504
|
-
"cache_read": 0.5,
|
|
1572
|
+
"input": 30,
|
|
1573
|
+
"output": 180,
|
|
1505
1574
|
"tiers": [
|
|
1506
1575
|
{
|
|
1507
|
-
"input":
|
|
1508
|
-
"output":
|
|
1509
|
-
"cache_read": 1,
|
|
1576
|
+
"input": 60,
|
|
1577
|
+
"output": 270,
|
|
1510
1578
|
"tier": {
|
|
1511
1579
|
"type": "context",
|
|
1512
1580
|
"size": 272000
|
|
@@ -1514,63 +1582,26 @@
|
|
|
1514
1582
|
}
|
|
1515
1583
|
],
|
|
1516
1584
|
"context_over_200k": {
|
|
1517
|
-
"input":
|
|
1518
|
-
"output":
|
|
1519
|
-
"cache_read": 1
|
|
1585
|
+
"input": 60,
|
|
1586
|
+
"output": 270
|
|
1520
1587
|
}
|
|
1521
1588
|
}
|
|
1522
1589
|
},
|
|
1523
|
-
"
|
|
1524
|
-
"id": "
|
|
1525
|
-
"name": "
|
|
1526
|
-
"description": "
|
|
1527
|
-
"family": "
|
|
1528
|
-
"attachment":
|
|
1529
|
-
"reasoning": false,
|
|
1530
|
-
"tool_call": true,
|
|
1531
|
-
"structured_output": true,
|
|
1532
|
-
"temperature": true,
|
|
1533
|
-
"knowledge": "2024-04",
|
|
1534
|
-
"release_date": "2025-04-14",
|
|
1535
|
-
"last_updated": "2025-04-14",
|
|
1536
|
-
"modalities": {
|
|
1537
|
-
"input": [
|
|
1538
|
-
"text",
|
|
1539
|
-
"image",
|
|
1540
|
-
"pdf"
|
|
1541
|
-
],
|
|
1542
|
-
"output": [
|
|
1543
|
-
"text"
|
|
1544
|
-
]
|
|
1545
|
-
},
|
|
1546
|
-
"open_weights": false,
|
|
1547
|
-
"limit": {
|
|
1548
|
-
"context": 1047576,
|
|
1549
|
-
"output": 32768
|
|
1550
|
-
},
|
|
1551
|
-
"cost": {
|
|
1552
|
-
"input": 2,
|
|
1553
|
-
"output": 8,
|
|
1554
|
-
"cache_read": 0.5
|
|
1555
|
-
}
|
|
1556
|
-
},
|
|
1557
|
-
"gpt-4o-2024-08-06": {
|
|
1558
|
-
"id": "gpt-4o-2024-08-06",
|
|
1559
|
-
"name": "GPT-4o (2024-08-06)",
|
|
1560
|
-
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
|
|
1561
|
-
"family": "gpt",
|
|
1562
|
-
"attachment": true,
|
|
1590
|
+
"text-embedding-3-large": {
|
|
1591
|
+
"id": "text-embedding-3-large",
|
|
1592
|
+
"name": "text-embedding-3-large",
|
|
1593
|
+
"description": "Embedding model for semantic search, retrieval, clustering, and ranking pipelines",
|
|
1594
|
+
"family": "text-embedding",
|
|
1595
|
+
"attachment": false,
|
|
1563
1596
|
"reasoning": false,
|
|
1564
|
-
"tool_call":
|
|
1565
|
-
"
|
|
1566
|
-
"
|
|
1567
|
-
"
|
|
1568
|
-
"
|
|
1569
|
-
"last_updated": "2024-08-06",
|
|
1597
|
+
"tool_call": false,
|
|
1598
|
+
"temperature": false,
|
|
1599
|
+
"knowledge": "2024-01",
|
|
1600
|
+
"release_date": "2024-01-25",
|
|
1601
|
+
"last_updated": "2024-01-25",
|
|
1570
1602
|
"modalities": {
|
|
1571
1603
|
"input": [
|
|
1572
|
-
"text"
|
|
1573
|
-
"image"
|
|
1604
|
+
"text"
|
|
1574
1605
|
],
|
|
1575
1606
|
"output": [
|
|
1576
1607
|
"text"
|
|
@@ -1578,13 +1609,12 @@
|
|
|
1578
1609
|
},
|
|
1579
1610
|
"open_weights": false,
|
|
1580
1611
|
"limit": {
|
|
1581
|
-
"context":
|
|
1582
|
-
"output":
|
|
1612
|
+
"context": 8191,
|
|
1613
|
+
"output": 3072
|
|
1583
1614
|
},
|
|
1584
1615
|
"cost": {
|
|
1585
|
-
"input":
|
|
1586
|
-
"output":
|
|
1587
|
-
"cache_read": 1.25
|
|
1616
|
+
"input": 0.13,
|
|
1617
|
+
"output": 0
|
|
1588
1618
|
}
|
|
1589
1619
|
},
|
|
1590
1620
|
"gpt-5.6-terra": {
|
|
@@ -1680,29 +1710,63 @@
|
|
|
1680
1710
|
}
|
|
1681
1711
|
}
|
|
1682
1712
|
},
|
|
1683
|
-
"
|
|
1684
|
-
"id": "
|
|
1685
|
-
"name": "
|
|
1686
|
-
"description": "
|
|
1687
|
-
"family": "
|
|
1713
|
+
"gpt-4": {
|
|
1714
|
+
"id": "gpt-4",
|
|
1715
|
+
"name": "GPT-4",
|
|
1716
|
+
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
|
|
1717
|
+
"family": "gpt",
|
|
1718
|
+
"attachment": true,
|
|
1719
|
+
"reasoning": false,
|
|
1720
|
+
"tool_call": true,
|
|
1721
|
+
"structured_output": false,
|
|
1722
|
+
"temperature": true,
|
|
1723
|
+
"knowledge": "2023-11",
|
|
1724
|
+
"release_date": "2023-11-06",
|
|
1725
|
+
"last_updated": "2024-04-09",
|
|
1726
|
+
"modalities": {
|
|
1727
|
+
"input": [
|
|
1728
|
+
"text"
|
|
1729
|
+
],
|
|
1730
|
+
"output": [
|
|
1731
|
+
"text"
|
|
1732
|
+
]
|
|
1733
|
+
},
|
|
1734
|
+
"open_weights": false,
|
|
1735
|
+
"limit": {
|
|
1736
|
+
"context": 8192,
|
|
1737
|
+
"output": 8192
|
|
1738
|
+
},
|
|
1739
|
+
"status": "deprecated",
|
|
1740
|
+
"cost": {
|
|
1741
|
+
"input": 30,
|
|
1742
|
+
"output": 60
|
|
1743
|
+
}
|
|
1744
|
+
},
|
|
1745
|
+
"gpt-5.2": {
|
|
1746
|
+
"id": "gpt-5.2",
|
|
1747
|
+
"name": "GPT-5.2",
|
|
1748
|
+
"description": "Reliable GPT generation for broad coding, writing, and tool-assisted product work",
|
|
1749
|
+
"family": "gpt",
|
|
1688
1750
|
"attachment": true,
|
|
1689
1751
|
"reasoning": true,
|
|
1690
1752
|
"reasoning_options": [
|
|
1691
1753
|
{
|
|
1692
1754
|
"type": "effort",
|
|
1693
1755
|
"values": [
|
|
1756
|
+
"none",
|
|
1694
1757
|
"low",
|
|
1695
1758
|
"medium",
|
|
1696
|
-
"high"
|
|
1759
|
+
"high",
|
|
1760
|
+
"xhigh"
|
|
1697
1761
|
]
|
|
1698
1762
|
}
|
|
1699
1763
|
],
|
|
1700
1764
|
"tool_call": true,
|
|
1701
1765
|
"structured_output": true,
|
|
1702
|
-
"temperature":
|
|
1703
|
-
"knowledge": "
|
|
1704
|
-
"release_date": "2025-
|
|
1705
|
-
"last_updated": "2025-
|
|
1766
|
+
"temperature": true,
|
|
1767
|
+
"knowledge": "2025-08-31",
|
|
1768
|
+
"release_date": "2025-12-11",
|
|
1769
|
+
"last_updated": "2025-12-11",
|
|
1706
1770
|
"modalities": {
|
|
1707
1771
|
"input": [
|
|
1708
1772
|
"text",
|
|
@@ -1714,20 +1778,20 @@
|
|
|
1714
1778
|
},
|
|
1715
1779
|
"open_weights": false,
|
|
1716
1780
|
"limit": {
|
|
1717
|
-
"context":
|
|
1718
|
-
"
|
|
1781
|
+
"context": 400000,
|
|
1782
|
+
"input": 272000,
|
|
1783
|
+
"output": 128000
|
|
1719
1784
|
},
|
|
1720
|
-
"status": "deprecated",
|
|
1721
1785
|
"cost": {
|
|
1722
|
-
"input": 1.
|
|
1723
|
-
"output":
|
|
1724
|
-
"cache_read": 0.
|
|
1786
|
+
"input": 1.75,
|
|
1787
|
+
"output": 14,
|
|
1788
|
+
"cache_read": 0.175
|
|
1725
1789
|
}
|
|
1726
1790
|
},
|
|
1727
|
-
"gpt-5
|
|
1728
|
-
"id": "gpt-5
|
|
1729
|
-
"name": "GPT-5
|
|
1730
|
-
"description": "
|
|
1791
|
+
"gpt-5": {
|
|
1792
|
+
"id": "gpt-5",
|
|
1793
|
+
"name": "GPT-5",
|
|
1794
|
+
"description": "Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows",
|
|
1731
1795
|
"family": "gpt",
|
|
1732
1796
|
"attachment": true,
|
|
1733
1797
|
"reasoning": true,
|
|
@@ -1735,25 +1799,23 @@
|
|
|
1735
1799
|
{
|
|
1736
1800
|
"type": "effort",
|
|
1737
1801
|
"values": [
|
|
1738
|
-
"
|
|
1802
|
+
"minimal",
|
|
1739
1803
|
"low",
|
|
1740
1804
|
"medium",
|
|
1741
|
-
"high"
|
|
1742
|
-
"xhigh"
|
|
1805
|
+
"high"
|
|
1743
1806
|
]
|
|
1744
1807
|
}
|
|
1745
1808
|
],
|
|
1746
1809
|
"tool_call": true,
|
|
1747
1810
|
"structured_output": true,
|
|
1748
1811
|
"temperature": false,
|
|
1749
|
-
"knowledge": "
|
|
1750
|
-
"release_date": "
|
|
1751
|
-
"last_updated": "
|
|
1812
|
+
"knowledge": "2024-09-30",
|
|
1813
|
+
"release_date": "2025-08-07",
|
|
1814
|
+
"last_updated": "2025-08-07",
|
|
1752
1815
|
"modalities": {
|
|
1753
1816
|
"input": [
|
|
1754
1817
|
"text",
|
|
1755
|
-
"image"
|
|
1756
|
-
"pdf"
|
|
1818
|
+
"image"
|
|
1757
1819
|
],
|
|
1758
1820
|
"output": [
|
|
1759
1821
|
"text"
|
|
@@ -1761,74 +1823,41 @@
|
|
|
1761
1823
|
},
|
|
1762
1824
|
"open_weights": false,
|
|
1763
1825
|
"limit": {
|
|
1764
|
-
"context":
|
|
1765
|
-
"input":
|
|
1826
|
+
"context": 400000,
|
|
1827
|
+
"input": 272000,
|
|
1766
1828
|
"output": 128000
|
|
1767
1829
|
},
|
|
1768
|
-
"experimental": {
|
|
1769
|
-
"modes": {
|
|
1770
|
-
"fast": {
|
|
1771
|
-
"cost": {
|
|
1772
|
-
"input": 5,
|
|
1773
|
-
"output": 30,
|
|
1774
|
-
"cache_read": 0.5
|
|
1775
|
-
},
|
|
1776
|
-
"provider": {
|
|
1777
|
-
"body": {
|
|
1778
|
-
"service_tier": "priority"
|
|
1779
|
-
}
|
|
1780
|
-
}
|
|
1781
|
-
}
|
|
1782
|
-
}
|
|
1783
|
-
},
|
|
1784
1830
|
"cost": {
|
|
1785
|
-
"input":
|
|
1786
|
-
"output":
|
|
1787
|
-
"cache_read": 0.
|
|
1788
|
-
"tiers": [
|
|
1789
|
-
{
|
|
1790
|
-
"input": 5,
|
|
1791
|
-
"output": 22.5,
|
|
1792
|
-
"cache_read": 0.5,
|
|
1793
|
-
"tier": {
|
|
1794
|
-
"type": "context",
|
|
1795
|
-
"size": 272000
|
|
1796
|
-
}
|
|
1797
|
-
}
|
|
1798
|
-
],
|
|
1799
|
-
"context_over_200k": {
|
|
1800
|
-
"input": 5,
|
|
1801
|
-
"output": 22.5,
|
|
1802
|
-
"cache_read": 0.5
|
|
1803
|
-
}
|
|
1831
|
+
"input": 1.25,
|
|
1832
|
+
"output": 10,
|
|
1833
|
+
"cache_read": 0.125
|
|
1804
1834
|
}
|
|
1805
1835
|
},
|
|
1806
|
-
"
|
|
1807
|
-
"id": "
|
|
1808
|
-
"name": "
|
|
1809
|
-
"description": "
|
|
1810
|
-
"family": "
|
|
1811
|
-
"attachment":
|
|
1836
|
+
"gpt-5.2-chat-latest": {
|
|
1837
|
+
"id": "gpt-5.2-chat-latest",
|
|
1838
|
+
"name": "GPT-5.2 Chat",
|
|
1839
|
+
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
|
|
1840
|
+
"family": "gpt-codex",
|
|
1841
|
+
"attachment": true,
|
|
1812
1842
|
"reasoning": true,
|
|
1813
1843
|
"reasoning_options": [
|
|
1814
1844
|
{
|
|
1815
1845
|
"type": "effort",
|
|
1816
1846
|
"values": [
|
|
1817
|
-
"
|
|
1818
|
-
"medium",
|
|
1819
|
-
"high"
|
|
1847
|
+
"medium"
|
|
1820
1848
|
]
|
|
1821
1849
|
}
|
|
1822
1850
|
],
|
|
1823
1851
|
"tool_call": true,
|
|
1824
1852
|
"structured_output": true,
|
|
1825
1853
|
"temperature": false,
|
|
1826
|
-
"knowledge": "
|
|
1827
|
-
"release_date": "
|
|
1828
|
-
"last_updated": "2025-
|
|
1854
|
+
"knowledge": "2025-08-31",
|
|
1855
|
+
"release_date": "2025-12-11",
|
|
1856
|
+
"last_updated": "2025-12-11",
|
|
1829
1857
|
"modalities": {
|
|
1830
1858
|
"input": [
|
|
1831
|
-
"text"
|
|
1859
|
+
"text",
|
|
1860
|
+
"image"
|
|
1832
1861
|
],
|
|
1833
1862
|
"output": [
|
|
1834
1863
|
"text"
|
|
@@ -1836,29 +1865,39 @@
|
|
|
1836
1865
|
},
|
|
1837
1866
|
"open_weights": false,
|
|
1838
1867
|
"limit": {
|
|
1839
|
-
"context":
|
|
1840
|
-
"output":
|
|
1868
|
+
"context": 128000,
|
|
1869
|
+
"output": 16384
|
|
1841
1870
|
},
|
|
1842
1871
|
"status": "deprecated",
|
|
1843
1872
|
"cost": {
|
|
1844
|
-
"input": 1.
|
|
1845
|
-
"output":
|
|
1846
|
-
"cache_read": 0.
|
|
1873
|
+
"input": 1.75,
|
|
1874
|
+
"output": 14,
|
|
1875
|
+
"cache_read": 0.175
|
|
1847
1876
|
}
|
|
1848
1877
|
},
|
|
1849
|
-
"
|
|
1850
|
-
"id": "
|
|
1851
|
-
"name": "
|
|
1852
|
-
"description": "
|
|
1853
|
-
"family": "
|
|
1878
|
+
"o4-mini": {
|
|
1879
|
+
"id": "o4-mini",
|
|
1880
|
+
"name": "o4-mini",
|
|
1881
|
+
"description": "Fast o-series model for compact reasoning, coding, and tool use",
|
|
1882
|
+
"family": "o-mini",
|
|
1854
1883
|
"attachment": true,
|
|
1855
|
-
"reasoning":
|
|
1884
|
+
"reasoning": true,
|
|
1885
|
+
"reasoning_options": [
|
|
1886
|
+
{
|
|
1887
|
+
"type": "effort",
|
|
1888
|
+
"values": [
|
|
1889
|
+
"low",
|
|
1890
|
+
"medium",
|
|
1891
|
+
"high"
|
|
1892
|
+
]
|
|
1893
|
+
}
|
|
1894
|
+
],
|
|
1856
1895
|
"tool_call": true,
|
|
1857
1896
|
"structured_output": true,
|
|
1858
|
-
"temperature":
|
|
1859
|
-
"knowledge": "
|
|
1860
|
-
"release_date": "
|
|
1861
|
-
"last_updated": "
|
|
1897
|
+
"temperature": false,
|
|
1898
|
+
"knowledge": "2024-05",
|
|
1899
|
+
"release_date": "2025-04-16",
|
|
1900
|
+
"last_updated": "2025-04-16",
|
|
1862
1901
|
"modalities": {
|
|
1863
1902
|
"input": [
|
|
1864
1903
|
"text",
|
|
@@ -1870,70 +1909,92 @@
|
|
|
1870
1909
|
},
|
|
1871
1910
|
"open_weights": false,
|
|
1872
1911
|
"limit": {
|
|
1873
|
-
"context":
|
|
1874
|
-
"output":
|
|
1912
|
+
"context": 200000,
|
|
1913
|
+
"output": 100000
|
|
1875
1914
|
},
|
|
1915
|
+
"status": "deprecated",
|
|
1876
1916
|
"cost": {
|
|
1877
|
-
"input":
|
|
1878
|
-
"output":
|
|
1879
|
-
"cache_read":
|
|
1917
|
+
"input": 1.1,
|
|
1918
|
+
"output": 4.4,
|
|
1919
|
+
"cache_read": 0.275
|
|
1880
1920
|
}
|
|
1881
1921
|
},
|
|
1882
|
-
"
|
|
1883
|
-
"id": "
|
|
1884
|
-
"name": "
|
|
1885
|
-
"description": "
|
|
1886
|
-
"family": "
|
|
1887
|
-
"attachment":
|
|
1888
|
-
"reasoning":
|
|
1889
|
-
"
|
|
1922
|
+
"gpt-realtime-2.1": {
|
|
1923
|
+
"id": "gpt-realtime-2.1",
|
|
1924
|
+
"name": "GPT-Realtime-2.1",
|
|
1925
|
+
"description": "Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior",
|
|
1926
|
+
"family": "gpt",
|
|
1927
|
+
"attachment": true,
|
|
1928
|
+
"reasoning": true,
|
|
1929
|
+
"reasoning_options": [
|
|
1930
|
+
{
|
|
1931
|
+
"type": "effort",
|
|
1932
|
+
"values": [
|
|
1933
|
+
"minimal",
|
|
1934
|
+
"low",
|
|
1935
|
+
"medium",
|
|
1936
|
+
"high",
|
|
1937
|
+
"xhigh"
|
|
1938
|
+
]
|
|
1939
|
+
}
|
|
1940
|
+
],
|
|
1941
|
+
"tool_call": true,
|
|
1942
|
+
"structured_output": false,
|
|
1890
1943
|
"temperature": false,
|
|
1891
|
-
"knowledge": "2024-
|
|
1892
|
-
"release_date": "
|
|
1893
|
-
"last_updated": "
|
|
1944
|
+
"knowledge": "2024-09-30",
|
|
1945
|
+
"release_date": "2026-07-06",
|
|
1946
|
+
"last_updated": "2026-07-06",
|
|
1894
1947
|
"modalities": {
|
|
1895
1948
|
"input": [
|
|
1896
|
-
"text"
|
|
1949
|
+
"text",
|
|
1950
|
+
"audio",
|
|
1951
|
+
"image"
|
|
1897
1952
|
],
|
|
1898
1953
|
"output": [
|
|
1899
|
-
"text"
|
|
1954
|
+
"text",
|
|
1955
|
+
"audio"
|
|
1900
1956
|
]
|
|
1901
1957
|
},
|
|
1902
1958
|
"open_weights": false,
|
|
1903
1959
|
"limit": {
|
|
1904
|
-
"context":
|
|
1905
|
-
"
|
|
1960
|
+
"context": 128000,
|
|
1961
|
+
"input": 96000,
|
|
1962
|
+
"output": 32000
|
|
1906
1963
|
},
|
|
1907
1964
|
"cost": {
|
|
1908
|
-
"input":
|
|
1909
|
-
"output":
|
|
1965
|
+
"input": 4,
|
|
1966
|
+
"output": 24,
|
|
1967
|
+
"cache_read": 0.4,
|
|
1968
|
+
"input_audio": 32,
|
|
1969
|
+
"output_audio": 64
|
|
1910
1970
|
}
|
|
1911
1971
|
},
|
|
1912
|
-
"
|
|
1913
|
-
"id": "
|
|
1914
|
-
"name": "
|
|
1915
|
-
"description": "
|
|
1916
|
-
"family": "
|
|
1917
|
-
"attachment":
|
|
1972
|
+
"o3-mini": {
|
|
1973
|
+
"id": "o3-mini",
|
|
1974
|
+
"name": "o3-mini",
|
|
1975
|
+
"description": "Smaller o-series reasoner for economical coding, math, and planning tasks",
|
|
1976
|
+
"family": "o-mini",
|
|
1977
|
+
"attachment": false,
|
|
1918
1978
|
"reasoning": true,
|
|
1919
1979
|
"reasoning_options": [
|
|
1920
1980
|
{
|
|
1921
1981
|
"type": "effort",
|
|
1922
1982
|
"values": [
|
|
1923
|
-
"
|
|
1983
|
+
"low",
|
|
1984
|
+
"medium",
|
|
1985
|
+
"high"
|
|
1924
1986
|
]
|
|
1925
1987
|
}
|
|
1926
1988
|
],
|
|
1927
1989
|
"tool_call": true,
|
|
1928
1990
|
"structured_output": true,
|
|
1929
1991
|
"temperature": false,
|
|
1930
|
-
"knowledge": "
|
|
1931
|
-
"release_date": "
|
|
1932
|
-
"last_updated": "2025-
|
|
1992
|
+
"knowledge": "2024-05",
|
|
1993
|
+
"release_date": "2024-12-20",
|
|
1994
|
+
"last_updated": "2025-01-29",
|
|
1933
1995
|
"modalities": {
|
|
1934
1996
|
"input": [
|
|
1935
|
-
"text"
|
|
1936
|
-
"image"
|
|
1997
|
+
"text"
|
|
1937
1998
|
],
|
|
1938
1999
|
"output": [
|
|
1939
2000
|
"text"
|
|
@@ -1941,42 +2002,44 @@
|
|
|
1941
2002
|
},
|
|
1942
2003
|
"open_weights": false,
|
|
1943
2004
|
"limit": {
|
|
1944
|
-
"context":
|
|
1945
|
-
"output":
|
|
2005
|
+
"context": 200000,
|
|
2006
|
+
"output": 100000
|
|
1946
2007
|
},
|
|
2008
|
+
"status": "deprecated",
|
|
1947
2009
|
"cost": {
|
|
1948
|
-
"input": 1.
|
|
1949
|
-
"output":
|
|
1950
|
-
"cache_read": 0.
|
|
2010
|
+
"input": 1.1,
|
|
2011
|
+
"output": 4.4,
|
|
2012
|
+
"cache_read": 0.55
|
|
1951
2013
|
}
|
|
1952
2014
|
},
|
|
1953
|
-
"
|
|
1954
|
-
"id": "
|
|
1955
|
-
"name": "
|
|
1956
|
-
"description": "
|
|
1957
|
-
"family": "
|
|
2015
|
+
"o3": {
|
|
2016
|
+
"id": "o3",
|
|
2017
|
+
"name": "o3",
|
|
2018
|
+
"description": "Deliberate o-series reasoner for hard math, coding, and multi-step analysis",
|
|
2019
|
+
"family": "o",
|
|
1958
2020
|
"attachment": true,
|
|
1959
2021
|
"reasoning": true,
|
|
1960
2022
|
"reasoning_options": [
|
|
1961
2023
|
{
|
|
1962
2024
|
"type": "effort",
|
|
1963
2025
|
"values": [
|
|
2026
|
+
"low",
|
|
1964
2027
|
"medium",
|
|
1965
|
-
"high"
|
|
1966
|
-
"xhigh"
|
|
2028
|
+
"high"
|
|
1967
2029
|
]
|
|
1968
2030
|
}
|
|
1969
2031
|
],
|
|
1970
2032
|
"tool_call": true,
|
|
1971
|
-
"structured_output":
|
|
2033
|
+
"structured_output": true,
|
|
1972
2034
|
"temperature": false,
|
|
1973
|
-
"knowledge": "
|
|
1974
|
-
"release_date": "2025-
|
|
1975
|
-
"last_updated": "2025-
|
|
2035
|
+
"knowledge": "2024-05",
|
|
2036
|
+
"release_date": "2025-04-16",
|
|
2037
|
+
"last_updated": "2025-04-16",
|
|
1976
2038
|
"modalities": {
|
|
1977
2039
|
"input": [
|
|
1978
2040
|
"text",
|
|
1979
|
-
"image"
|
|
2041
|
+
"image",
|
|
2042
|
+
"pdf"
|
|
1980
2043
|
],
|
|
1981
2044
|
"output": [
|
|
1982
2045
|
"text"
|
|
@@ -1984,65 +2047,74 @@
|
|
|
1984
2047
|
},
|
|
1985
2048
|
"open_weights": false,
|
|
1986
2049
|
"limit": {
|
|
1987
|
-
"context":
|
|
1988
|
-
"
|
|
1989
|
-
"output": 128000
|
|
2050
|
+
"context": 200000,
|
|
2051
|
+
"output": 100000
|
|
1990
2052
|
},
|
|
1991
2053
|
"cost": {
|
|
1992
|
-
"input":
|
|
1993
|
-
"output":
|
|
2054
|
+
"input": 2,
|
|
2055
|
+
"output": 8,
|
|
2056
|
+
"cache_read": 0.5
|
|
1994
2057
|
}
|
|
1995
2058
|
},
|
|
1996
|
-
"
|
|
1997
|
-
"id": "
|
|
1998
|
-
"name": "
|
|
1999
|
-
"description": "
|
|
2000
|
-
"family": "
|
|
2059
|
+
"o3-pro": {
|
|
2060
|
+
"id": "o3-pro",
|
|
2061
|
+
"name": "o3-pro",
|
|
2062
|
+
"description": "High-effort o3 tier for difficult technical reasoning and careful answers",
|
|
2063
|
+
"family": "o-pro",
|
|
2001
2064
|
"attachment": true,
|
|
2002
|
-
"reasoning":
|
|
2003
|
-
"
|
|
2065
|
+
"reasoning": true,
|
|
2066
|
+
"reasoning_options": [
|
|
2067
|
+
{
|
|
2068
|
+
"type": "effort",
|
|
2069
|
+
"values": [
|
|
2070
|
+
"low",
|
|
2071
|
+
"medium",
|
|
2072
|
+
"high"
|
|
2073
|
+
]
|
|
2074
|
+
}
|
|
2075
|
+
],
|
|
2076
|
+
"tool_call": true,
|
|
2077
|
+
"structured_output": true,
|
|
2004
2078
|
"temperature": false,
|
|
2005
|
-
"
|
|
2006
|
-
"
|
|
2079
|
+
"knowledge": "2024-05",
|
|
2080
|
+
"release_date": "2025-06-10",
|
|
2081
|
+
"last_updated": "2025-06-10",
|
|
2007
2082
|
"modalities": {
|
|
2008
2083
|
"input": [
|
|
2009
2084
|
"text",
|
|
2010
2085
|
"image"
|
|
2011
2086
|
],
|
|
2012
2087
|
"output": [
|
|
2013
|
-
"
|
|
2088
|
+
"text"
|
|
2014
2089
|
]
|
|
2015
2090
|
},
|
|
2016
2091
|
"open_weights": false,
|
|
2017
2092
|
"limit": {
|
|
2018
|
-
"context":
|
|
2019
|
-
"
|
|
2020
|
-
"output": 0
|
|
2093
|
+
"context": 200000,
|
|
2094
|
+
"output": 100000
|
|
2021
2095
|
},
|
|
2022
2096
|
"cost": {
|
|
2023
|
-
"input":
|
|
2024
|
-
"output":
|
|
2025
|
-
"cache_read": 1.25
|
|
2097
|
+
"input": 20,
|
|
2098
|
+
"output": 80
|
|
2026
2099
|
}
|
|
2027
2100
|
},
|
|
2028
|
-
"gpt-
|
|
2029
|
-
"id": "gpt-
|
|
2030
|
-
"name": "GPT-
|
|
2031
|
-
"description": "
|
|
2032
|
-
"family": "gpt
|
|
2101
|
+
"gpt-5.3-chat-latest": {
|
|
2102
|
+
"id": "gpt-5.3-chat-latest",
|
|
2103
|
+
"name": "GPT-5.3 Chat (latest)",
|
|
2104
|
+
"description": "Chat-tuned GPT model for conversational assistance, writing, and tool workflows",
|
|
2105
|
+
"family": "gpt",
|
|
2033
2106
|
"attachment": true,
|
|
2034
2107
|
"reasoning": false,
|
|
2035
2108
|
"tool_call": true,
|
|
2036
2109
|
"structured_output": true,
|
|
2037
2110
|
"temperature": true,
|
|
2038
|
-
"knowledge": "
|
|
2039
|
-
"release_date": "
|
|
2040
|
-
"last_updated": "
|
|
2111
|
+
"knowledge": "2025-08-31",
|
|
2112
|
+
"release_date": "2026-03-03",
|
|
2113
|
+
"last_updated": "2026-03-03",
|
|
2041
2114
|
"modalities": {
|
|
2042
2115
|
"input": [
|
|
2043
2116
|
"text",
|
|
2044
|
-
"image"
|
|
2045
|
-
"pdf"
|
|
2117
|
+
"image"
|
|
2046
2118
|
],
|
|
2047
2119
|
"output": [
|
|
2048
2120
|
"text"
|
|
@@ -2050,20 +2122,21 @@
|
|
|
2050
2122
|
},
|
|
2051
2123
|
"open_weights": false,
|
|
2052
2124
|
"limit": {
|
|
2053
|
-
"context":
|
|
2054
|
-
"output":
|
|
2125
|
+
"context": 128000,
|
|
2126
|
+
"output": 16384
|
|
2055
2127
|
},
|
|
2128
|
+
"status": "deprecated",
|
|
2056
2129
|
"cost": {
|
|
2057
|
-
"input":
|
|
2058
|
-
"output":
|
|
2059
|
-
"cache_read": 0.
|
|
2130
|
+
"input": 1.75,
|
|
2131
|
+
"output": 14,
|
|
2132
|
+
"cache_read": 0.175
|
|
2060
2133
|
}
|
|
2061
2134
|
},
|
|
2062
|
-
"gpt-5.
|
|
2063
|
-
"id": "gpt-5.
|
|
2064
|
-
"name": "GPT-5.
|
|
2065
|
-
"description": "
|
|
2066
|
-
"family": "gpt
|
|
2135
|
+
"gpt-5.5": {
|
|
2136
|
+
"id": "gpt-5.5",
|
|
2137
|
+
"name": "GPT-5.5",
|
|
2138
|
+
"description": "Default frontier GPT for coding, computer use, research, and knowledge work",
|
|
2139
|
+
"family": "gpt",
|
|
2067
2140
|
"attachment": true,
|
|
2068
2141
|
"reasoning": true,
|
|
2069
2142
|
"reasoning_options": [
|
|
@@ -2081,13 +2154,14 @@
|
|
|
2081
2154
|
"tool_call": true,
|
|
2082
2155
|
"structured_output": true,
|
|
2083
2156
|
"temperature": false,
|
|
2084
|
-
"knowledge": "2025-
|
|
2085
|
-
"release_date": "2026-
|
|
2086
|
-
"last_updated": "2026-
|
|
2157
|
+
"knowledge": "2025-12-01",
|
|
2158
|
+
"release_date": "2026-04-23",
|
|
2159
|
+
"last_updated": "2026-04-23",
|
|
2087
2160
|
"modalities": {
|
|
2088
2161
|
"input": [
|
|
2089
2162
|
"text",
|
|
2090
|
-
"image"
|
|
2163
|
+
"image",
|
|
2164
|
+
"pdf"
|
|
2091
2165
|
],
|
|
2092
2166
|
"output": [
|
|
2093
2167
|
"text"
|
|
@@ -2095,40 +2169,61 @@
|
|
|
2095
2169
|
},
|
|
2096
2170
|
"open_weights": false,
|
|
2097
2171
|
"limit": {
|
|
2098
|
-
"context":
|
|
2099
|
-
"input":
|
|
2172
|
+
"context": 1050000,
|
|
2173
|
+
"input": 922000,
|
|
2100
2174
|
"output": 128000
|
|
2101
2175
|
},
|
|
2176
|
+
"experimental": {
|
|
2177
|
+
"modes": {
|
|
2178
|
+
"fast": {
|
|
2179
|
+
"cost": {
|
|
2180
|
+
"input": 12.5,
|
|
2181
|
+
"output": 75,
|
|
2182
|
+
"cache_read": 1.25
|
|
2183
|
+
},
|
|
2184
|
+
"provider": {
|
|
2185
|
+
"body": {
|
|
2186
|
+
"service_tier": "priority"
|
|
2187
|
+
}
|
|
2188
|
+
}
|
|
2189
|
+
}
|
|
2190
|
+
}
|
|
2191
|
+
},
|
|
2102
2192
|
"cost": {
|
|
2103
|
-
"input":
|
|
2104
|
-
"output":
|
|
2105
|
-
"cache_read": 0.
|
|
2193
|
+
"input": 5,
|
|
2194
|
+
"output": 30,
|
|
2195
|
+
"cache_read": 0.5,
|
|
2196
|
+
"tiers": [
|
|
2197
|
+
{
|
|
2198
|
+
"input": 10,
|
|
2199
|
+
"output": 45,
|
|
2200
|
+
"cache_read": 1,
|
|
2201
|
+
"tier": {
|
|
2202
|
+
"type": "context",
|
|
2203
|
+
"size": 272000
|
|
2204
|
+
}
|
|
2205
|
+
}
|
|
2206
|
+
],
|
|
2207
|
+
"context_over_200k": {
|
|
2208
|
+
"input": 10,
|
|
2209
|
+
"output": 45,
|
|
2210
|
+
"cache_read": 1
|
|
2211
|
+
}
|
|
2106
2212
|
}
|
|
2107
2213
|
},
|
|
2108
|
-
"gpt-
|
|
2109
|
-
"id": "gpt-
|
|
2110
|
-
"name": "GPT-
|
|
2111
|
-
"description": "
|
|
2214
|
+
"gpt-4o-2024-11-20": {
|
|
2215
|
+
"id": "gpt-4o-2024-11-20",
|
|
2216
|
+
"name": "GPT-4o (2024-11-20)",
|
|
2217
|
+
"description": "GPT model for general reasoning, writing, coding, and tool-assisted tasks",
|
|
2112
2218
|
"family": "gpt",
|
|
2113
2219
|
"attachment": true,
|
|
2114
|
-
"reasoning":
|
|
2115
|
-
"reasoning_options": [
|
|
2116
|
-
{
|
|
2117
|
-
"type": "effort",
|
|
2118
|
-
"values": [
|
|
2119
|
-
"none",
|
|
2120
|
-
"low",
|
|
2121
|
-
"medium",
|
|
2122
|
-
"high"
|
|
2123
|
-
]
|
|
2124
|
-
}
|
|
2125
|
-
],
|
|
2220
|
+
"reasoning": false,
|
|
2126
2221
|
"tool_call": true,
|
|
2127
2222
|
"structured_output": true,
|
|
2128
|
-
"temperature":
|
|
2129
|
-
"knowledge": "
|
|
2130
|
-
"release_date": "
|
|
2131
|
-
"last_updated": "
|
|
2223
|
+
"temperature": true,
|
|
2224
|
+
"knowledge": "2023-09",
|
|
2225
|
+
"release_date": "2024-11-20",
|
|
2226
|
+
"last_updated": "2024-11-20",
|
|
2132
2227
|
"modalities": {
|
|
2133
2228
|
"input": [
|
|
2134
2229
|
"text",
|
|
@@ -2140,14 +2235,13 @@
|
|
|
2140
2235
|
},
|
|
2141
2236
|
"open_weights": false,
|
|
2142
2237
|
"limit": {
|
|
2143
|
-
"context":
|
|
2144
|
-
"
|
|
2145
|
-
"output": 128000
|
|
2238
|
+
"context": 128000,
|
|
2239
|
+
"output": 16384
|
|
2146
2240
|
},
|
|
2147
2241
|
"cost": {
|
|
2148
|
-
"input":
|
|
2242
|
+
"input": 2.5,
|
|
2149
2243
|
"output": 10,
|
|
2150
|
-
"cache_read":
|
|
2244
|
+
"cache_read": 1.25
|
|
2151
2245
|
}
|
|
2152
2246
|
}
|
|
2153
2247
|
}
|