GPT-6.1 Sol is OpenAI's mid tier GPT-6 model, aimed at complex coding, computer use and professional work at a lower cost than GPT-6 Astra. OpenAI released it on September 29, 2026, one week after GPT-6 Sol, and it replaced GPT-6 Sol as the recommended Sol model.
Key facts
| Item | GPT-6.1 Sol |
|---|---|
| Maker | OpenAI |
| Released | September 29, 2026 |
| API model ID | gpt-6.1-sol |
| Context window | 1,050,000 tokens (up to 922,000 input) |
| Max output | 128,000 tokens |
| Knowledge cutoff | April 30, 2026 |
| Input / output | Text and image in, text out |
| API price per million tokens | $2 input, $0.10 cached input, $10 output |
Sources: OpenAI's GPT-6.1 Sol model page, the API changelog and the GPT-6.1 Sol announcement, checked October 24, 2026.
Pricing details
Standard pricing is $2 per million input tokens and $10 per million output tokens. Cached input costs $0.10, which is 5 percent of the input rate, and cache writes cost $2.50. Prompts above 272,000 input tokens are billed at twice the input and cache rates and 1.5 times the output rate. Fast mode costs twice the standard rate, Ultrafast mode six times, and Batch and Flex half. Regional processing adds 10 percent where available, and the model supports US and EU data residency.
Where to use GPT-6.1 Sol
- API: on the OpenAI API. Use the Responses API for tool calling; Chat Completions works without tools. Reasoning effort runs from
lowtomax, withmediumas the default. Multi agent delegation is in beta. - ChatGPT: in ChatGPT Work and Codex for Plus, Pro, Business, Enterprise and Edu. OpenAI said at launch that it was not yet in the Chat tab of ChatGPT, where GPT-6 Sol runs for paid users.
How it compares
OpenAI says 6.1 Sol matches GPT-6 Astra on the DeepSWE software engineering benchmark at roughly a fifth of the cost. At the same token price, Anthropic's Claude Sonnet 5.5 is the closest rival, and Grok 4.7 is cheaper on output. Read GPT-6 Sol vs Astra vs Luna for the full lineup.
Who should use it
GPT-6.1 Sol is the sensible default for most OpenAI API work: coding agents, document work and analysis. Move up to Astra only when your tests show a clear gain, and drop to GPT-6 Luna for bulk tasks. Launch dates for all of these are on the release timeline.