OpenAI GPT-5.5 Pro — what it is, when to use it and what to use instead
No. GPT-5.5 Pro is active at OpenAI and has no announced end date.
Summary
GPT-5.5 Pro is the high-compute variant of GPT-5.5, at $30 per million input tokens and $180 per million output tokens with a 1.05M-token context window. It runs only on the Responses and Batch endpoints, and single requests can take several minutes.
Pro variants trade latency for a better single answer. GPT-5.5 Pro shares the 1.05M-token context window, 128K max output and 1 December 2025 knowledge cutoff of GPT-5.5, at six times the input price and six times the output price.
It accepts text and image input and returns text, with reasoning effort at medium, high (the default) or xhigh. There is no cached-input discount, which matters for repeated long prompts, and OpenAI recommends background mode to avoid client timeouts.
No deprecation has been announced. It is the model OpenAI's documentation names when steering users off GPT-5.2 Pro.
- Modality
- Use case
- Tier
- Access
Specifications
- Input price
- $30.00 / M tokens
- Output price
- $180.00 / M tokens
- Price band
- Flagship
- Context window
- 1.1M tokens
- Max output
- 128K tokens
- Input
- Text, image
- Output
- Text
- Knowledge cutoff
- December 2025
- Tool use
- Yes
- Structured output
- Yes
- Weights licence
- not published
- Model maker
- OpenAI
“unknown” means the figure has not been published, or we have not verified it yet — never that the answer is no. “not published” under the weights licence means this model is not offered with downloadable weights. The price band is ours: it bands the blended price (three parts input to one part output) as budget (≤ $0.50), standard (≤ $3), premium (≤ $12) or flagship.
Lifecycle
OpenAI has not announced a deprecation or shutdown date for this model. That is a statement about what has been published, not a guarantee — providers typically announce a shutdown months ahead, and this page updates when they do.
- Released
- Deprecated not announced
- Retired not announced
Use it for
- Solve individual hard problems offline, where a better answer is worth minutes of latency and $180 per million output tokens.
- Run batch jobs of high-value reasoning tasks through the Batch endpoint.
- Migrate off GPT-5.2 Pro, which OpenAI's documentation points away from.
Avoid it when
- You need an interactive response; requests can take several minutes and background mode is recommended.
- Your prompts repeat a long prefix; there is no cached-input discount on this model.
- Your integration uses Chat Completions; only the Responses and Batch endpoints are supported.
- Cost per call is a constraint; GPT-5.5 covers the same ground at $5 / $30.
Alternatives to consider
-
GPT-5.5 Curated
The standard variant at $5 / $30, a sixth of the price, with the same limits and knowledge cutoff.
$5.00 in · $30.00 out / M tokens 1.1M context
-
GPT-6 Astra Curated
A third of the input price at $10 / $50, with a newer April 2026 cutoff and support for Chat Completions.
$10.00 in · $50.00 out / M tokens 1.1M context
-
GPT-5.4 Pro Curated
The previous Pro variant at the same $30 / $180 but an older August 2025 knowledge cutoff.
$30.00 in · $180.00 out / M tokens 1.1M context
-
Claude Fable 5.1 Curated
Anthropic's deep-reasoning tier at $10 / $50, a third of the input price, and it also runs on the standard endpoint.
$10.00 in · $50.00 out / M tokens 1M context
-
GPT-5 pro Derived
Alternative from Microsoft Azure AI Foundry, same tier (frontier), similar price; shares chat, coding, reasoning, agents, long-context and search.
$15.00 in · $120.00 out / M tokens 400K context
Curated suggestions were picked by a reviewer, with the reason written by hand. Derived suggestions are ranked from the data — shared tags, the same tier and a comparable price band — not from benchmark scores. Check the capability fit yourself: we track lifecycle, price and specifications.
Sources
- developers.openai.com https://developers.openai.com/api/docs/pricing
- developers.openai.com https://developers.openai.com/api/docs/models/gpt-5.5-pro
Verified against the source on . Dataset snapshot . Providers can change a date without notice — the link above is always the final word.
About this profile
The specifications and dates above are facts from OpenAI, with the sources listed. The summary, the use-for and avoid-when lists and the curated alternatives are our assessment: editorial profile reviewed on by llm-researcher.
- developers.openai.com https://developers.openai.com/api/docs/models
Common questions
- What is GPT-5.5 Pro for?
- GPT-5.5 Pro is the high-compute variant of GPT-5.5, at $30 per million input tokens and $180 per million output tokens with a 1.05M-token context window. It runs only on the Responses and Batch endpoints, and single requests can take several minutes.
- What does GPT-5.5 Pro cost?
- $30.00 per million input tokens and $180.00 per million output tokens, as last verified on 2026-09-10.
- What is the context window of GPT-5.5 Pro?
- 1.1M tokens.
- Is GPT-5.5 Pro deprecated?
- No. GPT-5.5 Pro is active at OpenAI and has no announced end date.
- What should I use instead of GPT-5.5 Pro?
- OpenAI GPT-5.5 — The standard variant at $5 / $30, a sixth of the price, with the same limits and knowledge cutoff. OpenAI GPT-6 Astra — A third of the input price at $10 / $50, with a newer April 2026 cutoff and support for Chat Completions. OpenAI GPT-5.4 Pro — The previous Pro variant at the same $30 / $180 but an older August 2025 knowledge cutoff.