Claude Sonnet 5.5 vs Claude Haiku 5.5.

Claude Sonnet 5.5 scores higher, 55 against 36, but Claude Haiku 5.5 is more than 24 times cheaper per task.

Anthropic
Data from 13 November 2026.

Which is better?

Claude Sonnet 5.5 scores higher on Artificial Analysis's Intelligence Index, 55 against 36. It is 6th of the 58 models it scores, and Claude Haiku 5.5 13th.

Claude Sonnet 5.5 wins all four coding and reasoning tests.

Per task, Claude Haiku 5.5 is more than 24 times cheaper and faster too, $0.41 against $10.02 and 14.7 minutes against 34.1 minutes.

Per token, Anthropic charges $2 per million input tokens and $10 per million output tokens for Claude Sonnet 5.5, and $0.10 and $0.50 for Claude Haiku 5.5. Both read up to 1 million tokens at once.

  • Choose Claude Sonnet 5.5 if

    • you want the higher score, 55 against 36
    • you want it for coding, it wins FrontierCode 2.0 and Terminal-⁠Bench 4.0
    • you want it for reasoning, it wins HLE Diamond and CritPt v2
  • Choose Claude Haiku 5.5 if

    • the cost of each task matters, $0.41 against $10.02
    • you want it to take less time, 14.7 minutes against 34.1 per task

At a glance

FigureClaude Sonnet 5.5AnthropicClaude Haiku 5.5Anthropic
ScoreArtificial Analysis's index55Better6th of 583613th of 58
Cost per task$10.0248th of 51$0.41Better15th of 51
Time per task34.1 min41st of 5014.7 minBetter26th of 50
Input pricePer million tokens$2$0.10Better
Output pricePer million tokens$10$0.50Better
Context1 million1 million
WeightsClosedClosed
Effort measuredEach one's highestMaxMax
Released28 Sept 20267 Oct 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 page. The table of the 58.

Their tests

Coding

FrontierCode 2.0 Changes to real open-⁠source projects.

  • Claude Sonnet 5.541.7%
  • Claude Haiku 5.534.9%

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

  • Claude Sonnet 5.563.6%
  • Claude Haiku 5.532.8%

Reasoning

HLE Diamond Expert-⁠level questions.

  • Claude Sonnet 5.553.3%
  • Claude Haiku 5.529.6%

CritPt v2, hard subset Physics research problems.

  • Claude Sonnet 5.559.0%
  • Claude Haiku 5.543.5%

Questions.

  • Which one is cheaper?

    Per task, Claude Haiku 5.5, with $0.41 against $10.02 on the tasks of Artificial Analysis's index. Per token, Claude Haiku 5.5, with $0.10 and $0.50 per million input and output tokens against $2 and $10.

  • Which one reads more text at once?

    Both read up to 1 million tokens at once.

  • Can they run on your own servers?

    No. Neither has its weights published, so they are used through Anthropic 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.