ai/list101

GPT-6.2 Sol

openai.com

OpenAI model for everyday writing, coding and multistep workflows using text, images and connected tools.

Llm Models
Visit website
Pricing
Paid
Available on
OpenAI API, ChatGPT Work, Codex
Category
Llm Models

What is GPT-6.2 Sol?

GPT-6.2 Sol is OpenAI's versatile model for everyday technical and nontechnical work. Its intended tasks include writing, coding and agent workflows that involve multiple steps. It serves developers and knowledge workers seeking a general-purpose assistant.

Sol accepts text and images and produces text. It is available through the OpenAI API and eligible ChatGPT Work and Codex accounts, with tool support for applications that need to retrieve information or act in software environments.

Key features

Everyday writing and coding

Handles technical and nontechnical requests, including writing tasks and software work.

Text and image inputs

Use written instructions and visual context together, with responses delivered as text.

Multistep tool workflows

Responses API tools include web search, file search, computer use, code interpreter, hosted shell, patch application and MCP.

Large context

The API supports a 1,050,000-token context window, with maximum input of 922,000 tokens and maximum output of 128,000 tokens.

Structured application responses

Supports streaming, function calling and structured outputs for interactive interfaces and downstream processing.

Who is it for?

Knowledge workers drafting content

Provide notes and an intended audience to request a draft, revision or concise explanation.

Developers handling routine changes

Share relevant code and requirements to request an implementation proposal or debugging help.

Application builders connecting tools

Design an assistant that gathers relevant information before composing a response.

How to use GPT-6.2 Sol

  1. Choose ChatGPT Work, Codex or an OpenAI API integration.
  2. For subscriber access, update the desktop app or Codex CLI and select GPT-6.2 Sol when available.
  3. For API access, follow the developer quickstart and set the model identifier to gpt-6.2-sol.
  4. Provide instructions and relevant text or images; configure tools needed by the workflow.
  5. Review the generated response and refine the request as needed.

GPT-6.2 Sol pricing

Paid

API requests use metered token billing. Work and Codex access follows eligible subscription allowances. Processing tiers, tool use and cache writes can affect the total.

Standard API, short context

USD 2 input, USD 0.10 cached input and USD 10 output per million tokens
Metered token consumption rather than a monthly model subscription

Applies to requests with up to 272,000 input tokens. Larger requests double input and cache rates and increase output rates by 50% for the entire request. Tools and cache writes can incur additional charges.

Check current pricing ↗

What to know before you use it

  • Native audio and video modalities are unsupported.
  • Fine-tuning is unsupported, and reasoning effort cannot be set to none.
  • Subscriber access depends on plan, workspace permissions and rollout. Sol is not added to regular Chat by this release.

Frequently asked questions

Is Sol intended only for programming?

No. OpenAI positions it for everyday technical and nontechnical work, including writing and multistep workflows.

Can Sol return structured data?

Yes. Its API supports structured outputs and function calling.

Why might Sol be missing from my model picker?

Availability depends on your plan, workspace permissions and rollout progress. Updating and restarting the desktop app, or updating Codex CLI, may be necessary.