See how ChatGPT, Perplexity and Google AI Overviews describe you today
Free AI Visibility AuditGPT-5.2 Pro is OpenAI's most advanced model, built for complex tasks that need step-by-step reasoning, with agentic coding and long context handling improved over GPT-5 Pro.
OpenAIVerified 24/09/2026Released 10/12/202501 / 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
$21.00
field median $0.30
Output / 1M tokens
$168.00
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-5.2 Pro is a frontier reasoning model from OpenAI, aimed at work that has to be thought through in stages rather than answered in one pass. Its stated strengths are agentic coding and holding performance across long inputs. It is not the model to reach for when you want cheap, high-volume classification or short conversational replies, the cost structure makes that a poor fit.
Where it sits in its line, and when a sibling is the better pick.
GPT-5.2 Pro was first seen on 10 December 2025, arriving above GPT-5 Pro in OpenAI's line-up as the top reasoning tier. It is the right pick when a task genuinely requires multi-step reasoning or an agent working through a large codebase, and the wrong pick for anything routine, where a smaller sibling will do the same job for a fraction of the spend.
The API, the apps it powers, and what its limits let you do.
GPT-5.2 Pro is reached through the OpenAI API and accepts text, images and files, so it can take a spec document, a screenshot and a source file in the same request. The large context window means a long repository, a full contract or a set of research documents can go in whole rather than being chunked, and the generous output ceiling leaves room for long code generations or extended written analysis in a single response.
What changes because this exists, or why it does not.
The practical change is that agentic coding runs which previously lost the thread over a long session have more room to stay coherent, because both the input budget and the long context performance moved together. Teams that were splitting large documents into pieces and stitching the answers back together can stop doing that. Nothing here is new in kind, it is the existing job done with more headroom.
Follows GPT-5.2. Superseded by GPT-6 Astra. 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 | $21.00 | per 1M tokens | 2026-09-24 | Check |
| Output | $168.00 | per 1M tokens | 2026-09-24 | Check |
GPT-5.2 Pro sits at the expensive end of the market, with output priced many times above input, which is typical of reasoning tiers but steep here in absolute terms. Budget for it as a model you route hard tasks to, not one you run everything through, and expect the bill to be driven by how much it writes rather than how much you feed it.
| Workload | Input / month | Output / month | Cost |
|---|---|---|---|
| A small product team | 20M tokens | 5M tokens | $1,260.00 |
| A busy support assistant | 200M tokens | 40M tokens | $10,920.00 |
| A document pipeline | 1000M tokens | 100M tokens | $37,800.00 |
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 | image, text, file |
| Released | 10/12/2025 |
| Status | Current |
| Catalogue identifier | openai/gpt-5.2-pro |
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-5.2Came after
GPT-6 Astra →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 model with this much context can read a whole documentation set, pricing page and comparison article in one go rather than a snippet, so shallow, keyword-shaped pages lose ground to pages that hold up when read in full. If your product documentation, pricing and claims contradict each other across the site, a long context model will notice. The practical move is to make your own material internally consistent and complete enough to be read end to end.
Most buyers meet GPT-5.2 Pro through ChatGPT on OpenAI's higher paid tiers, where it handles the harder prompts people bring to it. Engineering teams meet it through the OpenAI API, usually inside a coding agent or a document analysis pipeline. Its answers also surface indirectly wherever a product has built its own AI search or assistant layer on top of OpenAI models.
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.