OpenAI released GPT-5.1 in ChatGPT on November 12, 2025, followed by the developer/API release on November 13. The ChatGPT versions—Instant, Thinking, and Pro—were retired on March 11, 2026, so GPT-5.1 is no longer selectable there. GPT-5.1 remains documented as an API model, but developers should check the live documentation for current availability, limits, pricing, and rate limits.
When was GPT-5.1 released?
OpenAI announced GPT-5.1 for ChatGPT on November 12, 2025, in its official launch post. The developer/API announcement followed on November 13, 2025, in OpenAI’s developer announcement.
The two announcements described related but separate products: ChatGPT variants for end users and API models for developers. Launch availability was staged, beginning with paid ChatGPT users before expanding, so the initial rollout should not be confused with present-day access.
What changed in ChatGPT?
OpenAI introduced two main ChatGPT variants, each aimed at a different balance of speed and reasoning:
#1 Best Overall
| Variant | OpenAI’s launch positioning | Current ChatGPT status |
|---|---|---|
| GPT-5.1 Instant | More conversational, better at following instructions, and able to use adaptive reasoning for harder prompts. | Retired from ChatGPT on March 11, 2026. |
| GPT-5.1 Thinking | Clearer and faster on simpler tasks while spending more effort on complex problems. | Retired from ChatGPT on March 11, 2026. |
| GPT-5.1 Pro | A higher-end ChatGPT option listed in the retirement notice. | Retired from ChatGPT on March 11, 2026. |
These descriptions are OpenAI’s product characterizations, not results from an independent comparative benchmark. They explain the intended behavior at launch rather than guaranteeing a particular response time or quality for every prompt.
Is GPT-5.1 still available in ChatGPT?
No. OpenAI’s ChatGPT release notes state that GPT-5.1 Instant, Thinking, and Pro were no longer available in ChatGPT on March 11, 2026.
OpenAI said existing conversations using GPT-5.1 would continue on corresponding current models. That means an older chat may remain accessible while being answered by a newer replacement model; it does not mean GPT-5.1 is still an option in the model picker.
What did GPT-5.1 add for API developers?
The November 13 developer release focused on controllable reasoning, coding, and agentic workflows:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
- Adaptive reasoning: OpenAI said GPT-5.1 dynamically adjusts how long it thinks based on task complexity, aiming to be faster and more token-efficient on simpler requests.
- A
nonereasoning option: Developers could choose no extended reasoning for latency-sensitive work, while retaining higher reasoning settings for difficult tasks. - Coding and agentic workflows: The announcement highlighted support for workflows that call tools, modify project files, and perform multi-step operations.
- Shell and
apply_patchtools: These tools were specifically named in the launch announcement for command execution and applying code changes. - Extended prompt caching: OpenAI described cache retention of up to 24 hours at launch. Treat that as a dated launch description and confirm current behavior before designing around it.
The launch announcement said GPT-5.1 and gpt-5.1-chat-latest were available on paid API tiers, with pricing and rate limits then matching GPT-5. Those were November 2025 terms, not a promise about current API billing or access.
What does the current API documentation say?
OpenAI’s current GPT-5.1 model page, accessed September 28, 2026, describes GPT-5.1 as a coding and agentic model with configurable reasoning effort. At that time, the page listed a 400,000-token context window and a 128,000-token maximum output.
Model specifications, snapshots, endpoint support, prices, and rate limits can change. Verify every value on the live model page and your account’s billing or limits documentation before deploying a production integration. The API’s documentation is a separate availability question from GPT-5.1’s retirement in ChatGPT.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Which GPT-5.1 details matter for a developer decision?
Reasoning versus latency
Use the reasoning control as a workload decision: low or no extended reasoning can suit routine, latency-sensitive requests, while more reasoning may help complex analysis or planning. OpenAI’s adaptive-reasoning description is a product claim, not an independent speed benchmark.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Best Value
Coding and tool use
GPT-5.1 was positioned for coding and agentic systems, including shell and apply_patch workflows. Review your tool permissions, sandboxing, logging, and approval controls before allowing a model to execute commands or change files.
Context and output capacity
The model page’s listed 400,000-token context and 128,000-token maximum output can matter for large repositories, long documents, and multi-step agents. Confirm that those limits apply to the exact model snapshot and endpoint you plan to use.
Pricing and limits
Do not copy launch-era pricing or rate limits into a current estimate. Check the live GPT-5.1 documentation and your organization’s current usage tier immediately before committing to an architecture.
Quick Recap
GPT-5.1 at a glance
| Question | Answer |
|---|---|
| ChatGPT announcement | November 12, 2025 |
| Developer/API announcement | November 13, 2025 |
| ChatGPT retirement | March 11, 2026 |
| API status | Documented separately; verify live availability and terms |
| Launch focus | Adaptive reasoning, instruction following, coding, and agentic workflows |
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
