Coding and software agents
Teams running multi-step coding, debugging, repository analysis, and review workflows that need stronger reasoning than a lightweight model.
Independent tool overview
GPT-6.1 Sol: a capable model for complex work at a lower cost
GPT-6.1 Sol is OpenAI's model for complex coding, computer use, and professional work. It aims to deliver near-Astra capability at lower token prices, with access through the API, Codex, and ChatGPT Work for eligible users.
Visit the official GPT-6.1 Sol site ↗GPT-6.1 Sol
- Best for
- Complex coding, computer use, and professional workflows
- API model ID
- gpt-6.1-sol
- Context window
- 1,050,000 tokens
- Maximum output
- 128,000 tokens
- Default reasoning
- Medium; low, high, xhigh, and max also supported
- Standard API pricing
- $2 input, $0.10 cached input, and $10 output per 1M tokens for short-context requests
Overview
What GPT-6.1 Sol is
GPT-6.1 Sol is an upgrade to GPT-6 Sol for demanding work across code, documents, and applications. OpenAI positions it between GPT-6 Astra, its most capable model, and GPT-6 Luna, its lower-cost choice for focused, high-volume tasks. The right choice still depends on your own quality, latency, and cost evaluations.
Developers select the API model ID gpt-6.1-sol. It accepts text and image inputs and generates text, with a 1,050,000-token context window and a 128,000-token maximum output. The Responses API supports tool calling, while Chat Completions is available without tool calling for this model. In ChatGPT, GPT-6.1 Sol launched in Work and Codex for eligible plans, not in Chat.
Use cases
Who GPT-6.1 Sol is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Document-heavy professional work
Analysts who need to work through long documents, tables, and supporting context while checking answers against source material.
Computer-use workflows
Builders of supervised agents that interact with apps through supported computer-use tools and must handle several connected steps.
Cost-conscious complex workloads
Teams that want to compare near-Astra capability with lower standard token prices before choosing a model for repeated work.
Capabilities
Core GPT-6.1 Sol features
Long context and large outputs
A 1,050,000-token context window and up to 128,000 output tokens support substantial documents, codebases, and multi-step task context.
Adjustable reasoning effort
Choose low, medium, high, xhigh, or max reasoning effort; medium is the default. None and minimal are not supported.
Responses API tool use
Supports function calling and tools including web and file search, code interpreter, hosted shell, apply patch, computer use, MCP, and tool search through the Responses API.
Text and image understanding
Accepts text and image inputs and produces text output. Native audio and video inputs are not supported on the model card.
Structured outputs and streaming
Supports structured outputs and streaming for applications that need predictable output formats or incremental responses.
Discounted cached input
Standard cached input costs $0.10 per million tokens for short-context requests, useful when an application repeatedly reuses the same prompt prefix.
Process
How the GPT-6.1 Sol workflow works
Step 1
Choose where to use it
Select GPT-6.1 Sol in an eligible Codex or ChatGPT Work workspace, or use the model ID gpt-6.1-sol in an OpenAI API integration.
Step 2
Set reasoning and provide context
Start with medium reasoning, then adjust effort for task difficulty. Supply the relevant code, documents, images, and constraints without assuming the full context window is always economical.
Step 3
Connect tools when needed
For API agents, use the Responses API to connect supported tools and define the actions the model may take. Chat Completions does not provide tool calling for GPT-6.1 Sol.
Step 4
Evaluate quality, latency, and cost
Review results and permissions on real tasks, then compare with Astra, Sol, or Luna before standardizing a production workflow.
Cost
GPT-6.1 Sol pricing and free plan
OpenAI bills API use by tokens. Standard short-context rates are $2 per million input tokens, $0.10 per million cached input tokens, and $10 per million output tokens. Requests above 272,000 input tokens use higher long-context rates for the full request; tool calls may add separate charges. ChatGPT Work and Codex access depends on plan and workspace settings and is separate from API billing.
API Standard — short context
$2 input / $0.10 cached input / $10 output per 1M tokens
Standard processing for prompts of up to 272,000 input tokens.
- Cache writes are $2.50 per million tokens.
- Built-in tools such as web search can carry separate charges.
- API usage is billed separately from ChatGPT subscriptions.
API Standard — long context
$4 input / $0.20 cached input / $15 output per 1M tokens
Applies when a prompt exceeds 272,000 input tokens, with the higher rates charged for the full request.
- Cache writes are $5 per million tokens.
- Check the current pricing table for processing-tier and regional variations.
ChatGPT Work and Codex
Eligible Plus, Pro, Business, Enterprise, or Edu plan
Model availability depends on the plan, client, rollout, and workspace settings.
- GPT-6.1 Sol launched in Work and Codex, not Chat.
- Enterprise administrators may control whether the model is enabled.
- A subscription does not include API token usage.
Pricing checked . Check current pricing at the source ↗
Assessment
GPT-6.1 Sol strengths and limitations
Where it stands out
- OpenAI positions it close to Astra on complex work while standard input and output token prices are one-fifth of Astra rates.
- The 1.05M-token context window and 128K-token output limit support large, connected tasks.
- Responses API support spans search, code execution, computer use, MCP, and other agent tools.
- Cached input is half the standard cached-input price of GPT-6 Sol, which can help repeated-context workloads.
What to consider
- Near-Astra performance is an OpenAI characterization, not a guarantee for every task; run your own evaluations.
- GPT-6.1 Sol does not support none or minimal reasoning effort, so Luna may be a better fit for simpler, high-volume requests.
- Prompts above 272,000 input tokens incur higher rates for the entire request, and tool calls may cost extra.
- The model is not available in Chat at launch; Work and Codex access depends on plan and workspace controls.
- The model card lists no native audio or video input support.
Compare
GPT-6.1 Sol alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Consumer
GPT-6 Astra
Choose Astra when the hardest end-to-end tasks justify a higher token price and you want OpenAI’s most capable model.
Explore GPT-6 Astra →Consumer
GPT-6 Sol
Compare the previous Sol model when you need to evaluate behavior or preserve a pinned existing workflow.
Explore GPT-6 Sol →Consumer
GPT-6 Luna
Choose Luna for focused, high-volume tasks where lower cost matters more than maximum capability.
Explore GPT-6 Luna →Consumer
Claude Opus 5.5
Compare a different model provider for demanding coding and knowledge work before committing to one ecosystem.
Explore Claude Opus 5.5 →Questions
GPT-6.1 Sol FAQs
What is GPT-6.1 Sol?
GPT-6.1 Sol is an OpenAI model for complex coding, computer use, and professional work. Its API model ID is gpt-6.1-sol.
How is GPT-6.1 Sol different from GPT-6 Sol?
OpenAI positions 6.1 Sol as a more capable upgrade. Both list $2 per million standard input tokens and $10 per million output tokens for short-context API requests, while 6.1 Sol cuts cached input from $0.20 to $0.10 per million tokens. Test both on your own tasks before switching.
How much does GPT-6.1 Sol cost in the API?
For standard requests with no more than 272,000 input tokens, OpenAI lists $2 per million input tokens, $0.10 per million cached input tokens, and $10 per million output tokens. Longer prompts and some tools cost more.
What are the context and output limits?
The official API model card lists a 1,050,000-token context window and a 128,000-token maximum output.
Can I use GPT-6.1 Sol in ChatGPT?
At launch it is available to eligible users in ChatGPT Work and Codex, but not in Chat. Plan, rollout, and workspace settings can affect availability.
Does GPT-6.1 Sol support tool calling?
Yes, through the Responses API. OpenAI says Chat Completions is supported for this model without tool calling.
Does it accept images, audio, or video?
It accepts text and image inputs and produces text output. The API model card does not list native audio or video support.
Is GPT-6.1 Sol free?
The API model card lists no free API tier. Access in Codex or ChatGPT Work depends on an eligible plan and workspace settings; API usage is billed separately.
Bottom line
Our GPT-6.1 Sol verdict
GPT-6.1 Sol is a strong model to evaluate first when work spans code, apps, and long documents but Astra’s token price is difficult to justify. Use Astra for the hardest cases and Luna for simpler work at scale. This assessment is documentation-based; benchmark claims should be checked against your own quality, latency, and cost measurements.
Visit GPT-6.1 Sol website ↗Similar Consumer Tools
View all Consumer AI tools →
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.