See how ChatGPT, Perplexity and Google AI Overviews describe you today
Free AI Visibility AuditGPT-5.1-Codex is OpenAI's coding-specialised version of GPT-5.1, built for software engineering work rather than general chat. It handles both interactive development sessions and long stretches of independent work on complex engineering tasks.
OpenAIVerified 24/09/2026Released 13/11/202501 / decision
Each figure sits next to the middle of the field, so it can be read as dear or cheap, wide or narrow, rather than floating on its own. Compared against the 126 models we track, not against an absolute standard.
Capability
Not measured
No published benchmark scores yet
Input / 1M tokens
$1.25
field median $0.30
Output / 1M tokens
$10.00
field median $1.25
Context
400K
field median 524K
02 / overview
Four questions, answered separately, because somebody arrives at one of them rather than at the top of the page.
The job it was built for, and the job it is not for.
GPT-5.1-Codex is a variant of GPT-5.1 tuned for software engineering and coding workflows. The job it is built for is writing, reading and changing code, either alongside a developer in a session or running on its own through a long task. It is not the model to reach for when you want general writing, research or open-ended conversation, that is what the base GPT-5.1 is for.
Where it sits in its line, and when a sibling is the better pick.
GPT-5.1-Codex first appeared in November 2025, as the coding-specialised sibling in the GPT-5.1 family. It is the right pick when the work is a codebase, particularly when a task needs to run for a long time without a human turning every crank. It is the wrong pick for anything outside engineering, where the general GPT-5.1 will behave better and you gain nothing from the specialisation.
The API, the apps it powers, and what its limits let you do.
GPT-5.1-Codex is reached through the OpenAI API and is the model behind OpenAI's Codex coding tooling. It accepts text and images, so screenshots, diagrams and design mockups can go into a prompt alongside code. The context window is large enough to hold a substantial slice of a repository, plus logs and diffs, in a single pass, and the output ceiling is high enough to return large multi-file changes rather than fragments.
What changes because this exists, or why it does not.
The practical change is duration. Agentic coding work that previously had to be chopped into small supervised steps can be handed over as one long task, with enough context to keep the whole picture in view and enough output room to return the full change set. For teams already running coding agents, this is a step up in how much you can delegate in one go rather than a new category of thing.
Follows GPT-5.1. Superseded by GPT-5.1-Codex-Max. See the whole line.
03 / evidence
Split by category, so a strong number never hides a thin evidence base. Verified means we read it on the benchmark's own published results; a provider's claim about its own model is shown and labelled rather than dropped.
No published benchmark scores for this model yet.
We publish a score only where we can link the result to where it was published. Until a benchmark result for this model exists in a source we read, this section stays empty rather than being filled with a provider's marketing figure.
How we decide what counts as evidence04 / ledger
Every published row, grouped by category, each compared with the best published score on the same benchmark. 'Is 64% good' is a question nobody can answer; '26 points behind the leader' is one anybody can.
Nothing in the ledger yet.
Each row here carries a score, the benchmark it came from, what the leading model scored on the same test, and a link to the published result. Rows appear as results are published and read.
How we decide what counts as evidence05 / capability
Where this model is strong, and against how many peers. Ranks are against models with evidence in that category, not against everything we track: ranking against models nobody tested would rank who published, not who is better.
No category scores to shape yet.
A category score is the weighted mean of the benchmarks published for it. With no published rows there is nothing to average, and an empty chart drawn at zero would say something false.
How we decide what counts as evidence06 / cost
List API rates as last read from the provider, with the source on every row, plus every change we have recorded since we started tracking it.
| Charge | Price | Unit | Read on | Source |
|---|---|---|---|---|
| Input | $1.25 | per 1M tokens | 2026-09-24 | Check |
| Output | $10.00 | per 1M tokens | 2026-09-24 | Check |
It sits at the same rate as the general GPT-5.1 line, so choosing the coding variant costs no premium over the standard model. Against the wider field it is mid-market, well above the small fast models used for classification and routing, and well below the top-tier reasoning models, though long agentic runs on a large context will consume tokens quickly whatever the unit rate.
| Workload | Input / month | Output / month | Cost |
|---|---|---|---|
| A small product team | 20M tokens | 5M tokens | $75.00 |
| A busy support assistant | 200M tokens | 40M tokens | $650.00 |
| A document pipeline | 1000M tokens | 100M tokens | $2,250.00 |
List API rates, no caching and no batch discount, which both providers offer and which change the answer a great deal. Treat these as the ceiling, not the bill.
07 / specs
As listed by the provider's own catalogue and re-read every few hours. Anything absent is absent there too.
| Context window | 400,000 tokens |
|---|---|
| Maximum output | 128,000 tokens |
| Modalities | text, image |
| Released | 13/11/2025 |
| Status | Current |
| Catalogue identifier | openai/gpt-5.1-codex |
08 / lineage
What this model replaced, what replaced it, and what else its provider has in the field.
Also from OpenAI
09 / line
Every model in this family in release order, so a page from eight months ago says in one glance that two newer ones exist.
Came before
← GPT-5.1Came after
GPT-5.1-Codex-Max →Ordered by release date and worked out from the naming, so a new member slots in as soon as its page exists. A retired model keeps its page and its place in the line.
10 / notes
What this model changes for a brand trying to be cited in AI answers, and every change we have logged since it launched.
What it changes for you
A coding-specialised model does not answer buyer questions, so it will not cite your marketing pages. Where it matters for a SaaS brand is developer-facing content: if you sell an API, an SDK or anything a developer integrates, your docs, code samples and error messages are now being read by an agent working through a long task, not skimmed by a human. Clear, complete, copy-pasteable documentation is what gets your product used correctly inside those sessions.
Buyers meet GPT-5.1-Codex mostly through developer tooling rather than a consumer chat app: OpenAI's Codex products and the API, plus any IDE extension, CI job or coding agent a team has wired up to it. It is not a model your prospects will encounter while asking a question in a chatbot, so it sits outside the usual AI search surfaces.
Nothing published here yet. Changes appear within hours of a provider announcing them.
11 / questions
The things people ask about this model, answered from what is on this page rather than from anywhere else.
Models change what gets cited. We measure whether AI answers name your brand or your competitors across every assistant, and show you what to change.