models.md +37 −84
2 2
3## Recommended models3## Recommended models
4 4
55
6 6
77gpt-5.3-codexgpt-5.4
8 8
99Most capable agentic coding model to date, combining frontier coding performance with stronger reasoning and professional knowledge capabilities.Flagship frontier model for professional work that brings the industry-leading coding capabilities of GPT-5.3-Codex together with stronger reasoning, tool use, and agentic workflows.
10 10
1111codex -m gpt-5.3-codexcodex -m gpt-5.4
12 12
13Copy command13Copy command
14 14
26 26
27API Access27API Access
28 28
2929
30 30
3131gpt-5.3-codex-sparkgpt-5.4-mini
32 32
3333Text-only research preview model optimized for near-instant, real-time coding iteration. Available to ChatGPT Pro users.Fast, efficient mini model for responsive coding tasks and subagents.
34 34
3535codex -m gpt-5.3-codex-sparkcodex -m gpt-5.4-mini
36 36
37Copy command37Copy command
38 38
50 50
51API Access51API Access
52 52
5353
54 54
5555gpt-5.2-codexgpt-5.3-codex
56 56
5757Advanced coding model for real-world engineering. Succeeded by GPT-5.3-Codex.Industry-leading coding model for complex software engineering. Its coding capabilities now also power GPT-5.4.
58 58
5959codex -m gpt-5.2-codexcodex -m gpt-5.3-codex
60 60
61Copy command61Copy command
62 62
74 74
75API Access75API Access
76 76
7777For most coding tasks in Codex, start with gpt-5.3-codex. It is available for
78ChatGPT-authenticated Codex sessions in the Codex app, CLI, IDE extension, and
79Codex Cloud. API access for GPT-5.3-Codex will come soon. The
80gpt-5.3-codex-spark model is available in research preview for ChatGPT Pro
81subscribers.
82
83## Alternative models
84
85
86
87gpt-5.2
88
89Our best general agentic model for tasks across industries and domains.
90
91codex -m gpt-5.2
92
93Copy command
94
95Show details
96
97
98
99gpt-5.1-codex-max
100
101Optimized for long-horizon, agentic coding tasks in Codex.
102
103codex -m gpt-5.1-codex-max
104
105Copy command
106
107Show details
108
109
110
111gpt-5.1
112
113Great for coding and agentic tasks across domains. Succeeded by GPT-5.2.
114
115codex -m gpt-5.1
116
117Copy command
118
119Show details
120
121
122 78
12379gpt-5.1-codexgpt-5.3-codex-spark
124 80
12581Optimized for long-running, agentic coding tasks in Codex. Succeeded by GPT-5.1-Codex-Max.Text-only research preview model optimized for near-instant, real-time coding iteration. Available to ChatGPT Pro users.
126 82
12783codex -m gpt-5.1-codexcodex -m gpt-5.3-codex-spark
128 84
129Copy command85Copy command
130 86
13187Show detailsCapability
132
133
134
135gpt-5-codex
136
137Version of GPT-5 tuned for long-running, agentic coding tasks. Succeeded by GPT-5.1-Codex.
138
139codex -m gpt-5-codex
140 88
14189Copy commandSpeed
142 90
14391Show detailsCodex CLI & SDK
144 92
14593Codex app & IDE extension
146 94
14795gpt-5-codex-miniCodex Cloud
148 96
14997Smaller, more cost-effective version of GPT-5-Codex. Succeeded by GPT-5.1-Codex-Mini.ChatGPT Credits
150 98
15199codex -m gpt-5-codexAPI Access
152 100
153101Copy commandFor most tasks in Codex, start with `gpt-5.4`. It combines strong coding,
102reasoning, native computer use, and broader professional workflows in one
103model. Use `gpt-5.4-mini` when you want a faster, lower-cost option for
104lighter coding tasks or subagents. The `gpt-5.3-codex-spark` model is
105available in research preview for ChatGPT Pro subscribers and is optimized for
106near-instant, real-time coding iteration.
154 107
155108Show details## Alternative models
156 109
157110
158 111
159112gpt-5gpt-5.2
160 113
161114Reasoning model for coding and agentic tasks across domains. Succeeded by GPT-5.1.Previous general-purpose model for coding and agentic tasks, including hard debugging tasks that benefit from deeper deliberation.
162 115
163116codex -m gpt-5codex -m gpt-5.2
164 117
165Copy command118Copy command
166 119
168 121
169## Other models122## Other models
170 123
171124Codex works best with the models listed above.When you sign in with ChatGPT, Codex works best with the models listed above.
172 125
173You can also point Codex at any model and provider that supports either the [Chat Completions](https://platform.openai.com/docs/api-reference/chat) or [Responses APIs](https://platform.openai.com/docs/api-reference/responses) to fit your specific use case.126You can also point Codex at any model and provider that supports either the [Chat Completions](https://platform.openai.com/docs/api-reference/chat) or [Responses APIs](https://platform.openai.com/docs/api-reference/responses) to fit your specific use case.
174 127
182The Codex CLI and IDE extension use the same `config.toml` [configuration file](https://developers.openai.com/codex/config-basic). To specify a model, add a `model` entry to your configuration file. If you don't specify a model, the Codex app, CLI, or IDE Extension defaults to a recommended model.135The Codex CLI and IDE extension use the same `config.toml` [configuration file](https://developers.openai.com/codex/config-basic). To specify a model, add a `model` entry to your configuration file. If you don't specify a model, the Codex app, CLI, or IDE Extension defaults to a recommended model.
183 136
184```137```
185138model = "gpt-5.2"model = "gpt-5.4"
186```139```
187 140
188### Choosing a different local model temporarily141### Choosing a different local model temporarily
192To start a new Codex CLI thread with a specific model or to specify the model for `codex exec` you can use the `--model`/`-m` flag:145To start a new Codex CLI thread with a specific model or to specify the model for `codex exec` you can use the `--model`/`-m` flag:
193 146
194```bash147```bash
195148codex -m gpt-5.3-codexcodex -m gpt-5.4
196```149```
197 150
198### Choosing your model for cloud tasks151### Choosing your model for cloud tasks