Claude Sonnet 5.5 vs GPT-⁠6.1 Sol.

They have the same score, 55, but GPT-⁠6.1 Sol is almost 8 times cheaper per task.

Anthropic and OpenAI
Data from 13 November 2026.

Which is better?

They have the same score on Artificial Analysis's Intelligence Index, 55. Going by the decimals, GPT-⁠6.1 Sol is 5th of the 58 models it scores, and Claude Sonnet 5.5 6th.

On the four coding and reasoning tests, Claude Sonnet 5.5 wins FrontierCode 2.0 and Terminal-⁠Bench 4.0, and GPT-⁠6.1 Sol wins HLE Diamond and CritPt v2.

Per task, GPT-⁠6.1 Sol is almost 8 times cheaper and faster too, $1.27 against $10.02 and 23.3 minutes against 34.1 minutes.

Per token they cost the same, $2 per million input tokens and $10 per million output tokens, so the difference per task comes from the tokens each one uses. They read about the same amount at once, 1 million and 1.05 million tokens.

  • Choose Claude Sonnet 5.5 if

    • you want it for coding, it wins FrontierCode 2.0 and Terminal-⁠Bench 4.0
  • Choose GPT-⁠6.1 Sol if

    • the cost of each task matters, $1.27 against $10.02
    • you want it for reasoning, it wins HLE Diamond and CritPt v2
    • you want it to take less time, 23.3 minutes against 34.1 per task

At a glance

FigureClaude Sonnet 5.5AnthropicGPT-⁠6.1 SolOpenAI
ScoreArtificial Analysis's index556th of 58555th of 58
Cost per task$10.0248th of 51$1.27Better29th of 51
Time per task34.1 min41st of 5023.3 minBetter32nd of 50
Input pricePer million tokens$2$2
Output pricePer million tokens$10$10
Context1 million1.05 million
WeightsClosedClosed
Effort measuredEach one's highestMaxMax
Released28 Sept 202629 Sept 2026

The score, the cost and the time per task and the tests from Artificial Analysis, read on 13 November 2026 with each model at its highest effort. The price, the context and the release date from Anthropic's and OpenAI's pages. The table of the 58.

Their tests

Coding

FrontierCode 2.0 Changes to real open-⁠source projects.

  • Claude Sonnet 5.541.7%
  • GPT-⁠6.1 Sol40.6%

Terminal-⁠Bench 4.0 Complex work in the terminal, on its own.

  • Claude Sonnet 5.563.6%
  • GPT-⁠6.1 Sol56.1%

Reasoning

HLE Diamond Expert-⁠level questions.

  • Claude Sonnet 5.553.3%
  • GPT-⁠6.1 Sol58.8%

CritPt v2, hard subset Physics research problems.

  • Claude Sonnet 5.559.0%
  • GPT-⁠6.1 Sol69.5%

Questions.

  • Which one is cheaper?

    Per task, GPT-⁠6.1 Sol, with $1.27 against $10.02 on the tasks of Artificial Analysis's index. Per token they cost the same, $2 and $10 per million input and output tokens. The cost per task says more than the price per million tokens, because for the same work some models use many more tokens than others.

  • Which one reads more text at once?

    They read about the same amount at once, 1 million and 1.05 million tokens.

  • Can they run on your own servers?

    No. Neither has its weights published, so they are used through Anthropic and OpenAI or whoever offers them.

  • Which one for your company's software?

    It depends on the job. We cover it in “Which AI Model to Use to Build Your Company's Software”.

Let's talk.

Tell us what you need and one of the partners who write the code gets back to you.