Updated

A polished brushed-aluminium sphere wrapped in three thin concentric floating halo rings on a dark background, abstract hero artwork for GPT-6.1 Sol Pro

OpenAI

GPT-6.1 Sol Pro

GPT-6.1 Sol in OpenAI's pro reasoning mode, for hard problems where accuracy beats speed.

Chat with GPT-6.1 Sol Pro in Super AgentView pricing

What is GPT-6.1 Sol Pro?

GPT-6.1 Sol Pro is GPT-6.1 Sol with OpenAI's pro reasoning mode switched on.

OpenAI documents pro as a setting on the same model ID, meant for difficult tasks that need more model work and can tolerate higher latency and token usage. Answers take longer and use more tokens than the same request on standard GPT-6.1 Sol. OpenAI hasn't published benchmark results for pro mode, so the official results on the GPT-6.1 Sol page are the closest reference.

Provider
OpenAI
Context window
1.05M tokens
Input
Text, Image
Reasoning
Low to max
Max output
128K tokens
Knowledge cutoff
Apr 30, 2026

How GPT-6.1 Sol Pro performs

On LMArena

People compare two anonymous answers and vote for the better one. Ranked here against the other chat models you can pick in Super Agent.

LMArena has no separate entry for GPT-6.1 Sol Pro. It runs the same model as GPT-6.1 Sol in a deeper reasoning mode, and that page shows the base model's arena results.

Source: LMArena, leaderboard published . Scores move as new votes come in.

What you can do with GPT-6.1 Sol Pro

  • Same model as GPT-6.1 Sol, run in OpenAI's pro reasoning mode

  • More model work on every request, at every effort level

  • Keeps GPT-6.1 Sol's 1.05M-token context and image input

  • Effort from low to max, as on the base model

How to prompt GPT-6.1 Sol Pro

  • Ask for one well-defined result per message. Pro mode works best on a single hard problem.
  • Give it every constraint and reference at once, since each round trip takes a while.
  • Try the question on GPT-6.1 Sol first and switch to Pro only if the answer falls short.

Where GPT-6.1 Sol Pro fits best

  • One hard question where a wrong answer is costly

  • Planning a complex multi-asset campaign before generating anything

  • Checking a long script or brief for logic gaps

  • Problems GPT-6.1 Sol got wrong on the first try

How GPT-6.1 Sol Pro compares

GPT-6.1 Sol Pro compared with other chat models on Flashloop
ModelLMArena textLMArena codingContext windowReasoning
GPT-6.1 Sol ProNoNo1.05M tokensLow to max
GPT-6.1 Sol1480#271537#141.05M tokensLow to max
Claude Sonnet 5.51478#291528#261M tokensLow to max
GPT-6 Luna1442#911507#651.05M tokensOff to max
GPT-6 Luna ProNoNo1.05M tokensOff to max

Good to know before you start

  • Slower than GPT-6.1 Sol on the same request
  • Uses more tokens per answer, which OpenAI notes also raises cost
  • OpenAI publishes no pro-mode benchmarks, so the gain is hard to measure up front

Start chatting with GPT-6.1 Sol Pro

Chat with GPT-6.1 Sol Pro in Super Agent

  1. Select GPT-6.1 Sol Pro

    Open Super Agent on Flashloop and choose GPT-6.1 Sol Pro in the model picker.

  2. Enter your prompt

    Ask for what you need: an answer, a plan, or the images and videos you want the agent to make.

  3. Send and keep going

    Send the message. GPT-6.1 Sol Pro answers in Super Agent and runs any generations you asked for.

Questions about GPT-6.1 Sol Pro

Is GPT-6.1 Sol Pro a different model from GPT-6.1 Sol?

No. OpenAI serves the same model with its reasoning mode set to pro, which makes it work longer before answering.

When should I use GPT-6.1 Sol Pro?

For difficult questions where you can wait longer for the answer. OpenAI keeps standard mode as the default and offers pro for tasks that need more model work.

Why is GPT-6.1 Sol Pro slower?

Pro mode does more work per request, so it uses more tokens and takes longer than the same request on GPT-6.1 Sol.

Are there official benchmarks for GPT-6.1 Sol Pro?

No. OpenAI hasn't published results for pro mode. Its launch results cover GPT-6.1 Sol across its standard effort levels, and you can find them on the GPT-6.1 Sol page.

Which reasoning levels does GPT-6.1 Sol Pro support in Super Agent?

Low, medium, high, extra high and max. Thinking is always on, and Super Agent starts it at low.

More chat models on Flashloop

Browse all models