> ## Documentation Index > Fetch the complete documentation index at: https://docs.tess.im/llms.txt > Use this file to discover all available pages before exploring further. # GPT-6.1 Sol > Learn about GPT-6.1 Sol, the OpenAI model that comes close to GPT-6 Astra on documents, business processes, and professional work, at a fraction of the cost. GPT-6.1 Sol is the upgrade to GPT-6 Sol, released by OpenAI on September 29, 2026, one week after the previous generation. It was built for **everyday professional work**: understanding complex documents, running processes that span several systems, and operating applications on a computer. The highlight is the balance between quality and cost. According to OpenAI, GPT-6.1 Sol **comes close to GPT-6 Astra**, the company's most capable model, at about **one-fifth of Astra's cost**. It keeps the same list price as GPT-6 Sol, with gains on almost every front. In Artificial Analysis's independent evaluation, it lands just one point behind Astra. | **Provider model** `gpt-6.1-sol` | **Context** 1.05M tokens | **Max output** 128K tokens | **Provider** OpenAI | | :- | :- | :- | :- | | **Input** Text + image | **Output** Text | **Reasoning** Yes | **Knowledge cutoff** April 30, 2026 | ## When to use GPT-6.1 Sol and when to use GPT-6 Astra OpenAI positions GPT-6.1 Sol as the new balance between capability and cost for important everyday work. Astra remains the choice for the hardest tasks, such as frontier scientific research. | | GPT-6.1 Sol | GPT-6 Astra | | - | - | - | | Best for | Documents, business processes, computer use, and everyday analysis | The most demanding work and the hardest scientific research | | Quality | Close to Astra in most published evaluations | The highest in the GPT-6 family | | Context / max output | 1.05M / 128K | 1.05M / 128K | | Usage price | One-fifth of Astra | Reference | > [!TIP] > > Start with GPT-6.1 Sol. Move up to Astra only when Sol's quality still doesn't close the deliverable. ## What changes compared with GPT-6 Sol ### Complex documents, with tables and fine print In the GDP.pdf evaluation, models answer real questions about work PDFs in finance, healthcare, legal, and seven other professional fields, with tables, charts, diagrams, and fine-print details. GPT-6.1 Sol scored **above Claude Opus 5.5 at less than half the cost per task** and came close to Astra at about one-fifth of the cost. ### Business processes from start to finish In AutomationBench, agents run complete workflows with 47 tools across sales, marketing, operations, support, finance, and HR. GPT-6.1 Sol scored **2.2 points above Claude Opus 5.5 at about one-third of the cost**, and 4.8 points above GPT-6 Sol at the same setting. ### Steadier at operating applications In long computer-use tasks (OSWorld 2.0), GPT-6.1 Sol beat GPT-6 Sol by 7 points at less than half the cost. It came within 2.1 points of Astra, at about one-seventh of the cost per task. ### Fewer factual errors On difficult questions where users had already flagged errors from earlier models, the share of answers with at least one factual error fell from 11.4% to 7.7%, a reduction of about 32% compared with GPT-6 Sol. Across all tested settings, the error rate stayed within 2 points of Astra's. ### More transparent about what it couldn't do OpenAI reports that the model better respects the restrictions set by the user and more often flags when a tool isn't working, instead of carrying on as if everything were fine. ## What the benchmarks show OpenAI published most results as comparisons with other models, in quality and cost per task. | Benchmark | GPT-6.1 Sol result | Cost per task | What it measures | | - | - | - | - | | GDP.pdf | Above Opus 5.5 and close to Astra | Less than half of Opus 5.5; about 1/5 of Astra | Professional questions about complex PDFs | | AutomationBench 1.0.6 (medium) | **+2.2 points** over Opus 5.5; +4.8 over GPT-6 Sol | About 1/3 of Opus 5.5 | Business processes with 47 tools | | OSWorld 2.0 offline (max) | **+7 points** over GPT-6 Sol; 2.1 behind Astra | About 1/7 of Astra | Computer use in long tasks | | DeepSWE v1.1 | **Matches Astra**; +6.4 over GPT-6 Sol's best | About 1/5 of Astra | Software engineering in real projects | | Terminal-Bench Science 0.1 (max) | **More than double** GPT-6 Sol | Over 75% lower than Opus 5.5 and Astra | Data analysis and scientific research | | Factuality (low) | **7.7%** of answers with errors, versus 11.4% for GPT-6 Sol | — | Factual errors on difficult questions | Artificial Analysis's independent evaluation confirms how close it is to Astra. | Model | Intelligence Index | | - | -: | | GPT-6 Astra | 53 | | **GPT-6.1 Sol** | **52** | | GPT-6 Sol | 48 | GPT-6.1 Sol ranked 10th out of 222 models in the same price range, where the median is 26. The index combines ten evaluations, including real professional work, document analysis, and business processes. > [!NOTE] > > The results in the table were published by OpenAI, measured in the company's research environment or API. Numbers for other models came from public reports. Results depend on the tools and settings used in each measurement. Validate in your own workflow before changing a production setup. ### Where it doesn't lead * **Astra stays ahead** on the hardest scientific tasks: on Terminal-Bench Science, it had the highest score among the models tested (68.1%), and OpenAI recommends Astra for this kind of research. * On Artificial Analysis, running at maximum effort, the model took longer to start answering and generated text more slowly than the average for its range. For simple tasks, lower effort levels tend to be faster. ## Where it creates the most value ### A contract or report full of fine print Contracts, expert reports, annual reports, and tenders hide deadlines, penalties, and conditions in tables and footnotes. GPT-6.1 Sol reads the whole document and answers based on what is written. **Expected result:** answers with the source page or clause and the points of attention highlighted. ### A process that spans several systems A lead arrives in the CRM and needs a proposal, an email, and an entry in the tracking spreadsheet. A refund request moves through support, finance, and approval. The model runs the steps in sequence and asks for confirmation at critical points. **Expected result:** the process completed end to end, with every step recorded. ### An analysis that has to be right Reports for leadership, answers to customers, and internal opinions can't contain wrong information. With fewer factual errors, GPT-6.1 Sol reduces rework during review. **Expected result:** analysis with sources cited and gaps pointed out, instead of filled in by guesswork. ## How to apply it in Tess **1. Choose the model for the job** Use GPT-6.1 Sol for documents, business processes, and everyday analysis. Save GPT-6 Astra for the hardest work, when Sol's quality still doesn't close the deliverable. **2. Provide the material and the expected format** Send the documents, spreadsheets, and examples of the result. Say who the material is for and how it will be used. **3. Set tools and approvals** Enable only the tools the process actually uses. Require confirmation before sending, publishing, or changing data. **4. Ask for a check before delivery** Instruct the agent to verify numbers, clauses, and sources, and to say what it couldn't confirm. ## How to write the request GPT-6.1 Sol performs best when the request works like an agreement, not a step-by-step script. Say what needs to come out, let the model choose the path, and define how the result will be checked. | Part of the request | What to say | | - | - | | Result | What needs to be ready at the end | | Context | Only the documents and data that matter for the task | | Limits | What the agent must not do | | Tools and approvals | What it can use on its own and what needs your approval | | Done criteria | How to know the work is complete | | Check | What it should verify before delivering | | When to stop | When it should ask instead of assuming | Not every request needs every part. A simple question needs only result and context. An agent that changes systems needs clear limits and approvals. > [!TIP] > > When sources may disagree, say which one wins. For example: the current policy outweighs an old document. That way the model flags the conflict instead of choosing on its own. ## Three playbooks to get started ### 1. Contract review **Input:** contract PDF, the company's points of attention, and an opinion template.\ **Tools:** file reading and document generation. ```text theme={null} Read the entire contract, including annexes and footnotes. List deadlines, amounts, penalties, automatic renewal, and termination conditions, with the clause and page for each item. Compare against the attached points of attention and highlight what differs. Do not interpret ambiguous clauses: flag them for legal review. If the contract doesn't answer something, write "not stated in the document". ``` **Done when:** every item has a source clause and page, differences are highlighted, and ambiguous points are set aside for legal. ### 2. End-to-end sales follow-up **Input:** list of opportunities in the CRM, proposal template, and sales policy.\ **Tools:** CRM, email, and spreadsheet integrations. ```text theme={null} For each opportunity stalled for more than 7 days, review its history in the CRM, prepare an updated proposal using the attached template, and a follow-up email in the company's tone. Do not send anything: leave the emails as drafts for approval and record the status and next step of each opportunity in the spreadsheet. After recording, reopen the spreadsheet and confirm everything was saved. ``` **Done when:** every opportunity has a proposal, a draft email, and a recorded next step, awaiting approval. ### 3. Invoice and payment reconciliation **Input:** invoice PDFs, bank statement, and accounts payable spreadsheet.\ **Tools:** file reading, data analysis, and spreadsheet generation. ```text theme={null} Match each invoice with the corresponding payment in the bank statement and in the accounts payable spreadsheet. Build a spreadsheet with what matches, what differs in amount or date, and what has no match. Do not adjust amounts: point out the difference and the source of each number. ``` **Done when:** every invoice is reconciled or flagged as divergent, with the source of each amount. ## Limitations and best practices * For the hardest tasks, especially frontier scientific research, GPT-6 Astra remains the safer choice. * OpenAI maintains reinforced safeguards for cybersecurity and biology topics. * A 1.05M-token context doesn't replace selecting information. Send only the relevant content. * The usage price is one-fifth of Astra's, but the final cost of each task depends on how much the model works and how many revisions it needs. Compare both in your workflow by the cost of an approved result. * Review numbers, clauses, and citations before publishing or sending. * Keep human approval for financial, legal, clinical, or high-impact decisions. See also: [Models and Costs](/en/models-and-cost) · [GPT-6 Sol and Luna](/en/gpt-6-sol-luna) · [GPT-6 Astra](/en/gpt-6-astra) · [Introducing GPT-6.1 Sol — OpenAI](https://openai.com/index/introducing-gpt-6-1-sol/) · [GPT-6.1 Sol — Artificial Analysis](https://artificialanalysis.ai/models/releases/gpt-6-1-sol) · [GPT-6.1 Sol — OpenAI documentation](https://developers.openai.com/api/docs/models/gpt-6.1-sol). This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.