GPT-6.1 Sol: API price, context, tools, and compatibility

Quick answer

OpenAI introduced GPT-6.1 Sol on September 29, 2026. Its API page lists the model ID gpt-6.1-sol, a 1.05-million-token context window, and standard token rates of $2 per million input tokens and $10 per million output tokens. Tool calling uses the Responses API. These are API specifications, not proof of access in ChatGPT's standard chat or any third-party app; the exact model is not verified in Chat AI's public catalog.

Download Chat AI Opens the official App Store or Google Play for your device.

GPT-6.1 Sol is a distinct API model release

OpenAI's Deployment Safety Hub says GPT-6.1 Sol was introduced September 29 as a new model in the GPT-6 family. The developer page identifies its API model ID as gpt-6.1-sol and positions it for complex coding, computer use, and professional work, describing performance as comparable to GPT-6 Astra at lower cost. That performance characterization is OpenAI's positioning, not an independent benchmark or a result on your workload.

Sources: OpenAI Developers, OpenAI Deployment Safety Hub

Standard API rates are $2 input and $10 output per million tokens

The model page lists $2.00 per million uncached input tokens and $10.00 per million output tokens. Cached input is $0.10 per million, while cache writes are $2.50. Requests above 272,000 input tokens use twice the input and cache rates and 1.5 times the output rate for the full request. Fast mode is priced at twice standard; Batch and Flex are listed at 50% below standard, and regional processing adds a 10% premium where available. A long context window can still make a task expensive; estimate from actual input, output, and processing tier rather than comparing only the base input rate.

Sources: OpenAI Developers

The model supports long inputs, images, and Responses API tools

OpenAI lists a 1,050,000-token context window and up to 128,000 output tokens. The model accepts text and image input, produces text output, and does not list audio or video support. Web search, file search, code interpreter, computer use, hosted shell, and other tools are supported through the Responses API. Chat Completions is available without tool calling. Supported reasoning effort values are low, medium, high, xhigh, and max; none and minimal are not supported. Check the migration guide before changing an integration, particularly if it relies on Chat Completions tools or an unsupported reasoning setting.

Sources: OpenAI Developers, OpenAI Developers

OpenAI applies its higher-risk safeguards; API listing is not app availability

OpenAI's safety addendum says it treats GPT-6.1 Sol as Critical capability in cybersecurity and High capability in biological and chemical domains under its Preparedness Framework, and applies the same safeguards stack as GPT-6 Astra. These are OpenAI's assessment categories and evaluations, not a guarantee that errors or misuse are impossible. Separately, a live API model ID does not establish access in regular ChatGPT or a third-party app. GPT-6.1 Sol is not listed by exact name in Chat AI's public model directory, so this article makes no claim that Chat AI includes it.

Sources: OpenAI Deployment Safety Hub, Chat AI, Chat AI

Frequently asked questions

What readers usually ask

What is the GPT-6.1 Sol API model ID?

OpenAI's model page lists gpt-6.1-sol. A published model ID does not by itself confirm that a particular app or subscription exposes it.

How much does GPT-6.1 Sol cost?

The listed standard rates are $2 per million uncached input tokens and $10 per million output tokens. Cached input, cache writes, long prompts, processing tier, and regional processing have separate rates.

Can GPT-6.1 Sol call tools through Chat Completions?

OpenAI says to use the Responses API for tool calling. Chat Completions is supported without tool calling.

Is GPT-6.1 Sol available in Chat AI?

Its exact name is not verified in Chat AI's public model directory. Check the directory for the current catalog; this API launch is not an availability announcement for Chat AI.

Evidence