Claude Haiku 5.5
Made by Anthropic, released 7 Oct 2026. Epoch AI lists its access as “API access”. Appears solid at maths among the models listed here.
In short
What it does well
- Maths: Strong (16th of 43 listed models)
Where it falls short
- Coding: Behind the leaders (64th of 77 listed models)
- Knowledge: Behind the leaders (41st of 43 listed models)
- Writing: Behind the leaders (76th of 77 listed models)
What we don't know yet
- Fixing real software bugs: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- Behind the leadersHow often people preferred its answers to writing prompts. 1,336 on Arena's scale · Source for Claude Haiku 5.5 writing score (lmarena)
- Coding
- Behind the leadersHow often people preferred its answers to coding questions. 1,454 on Arena's scale · Source for Claude Haiku 5.5 coding score (lmarena)
- Maths
- StrongHow many very hard maths problems it solved. 75.1% · Source for Claude Haiku 5.5 maths score (epoch)
- Knowledge
- Behind the leadersHow many short factual questions it answered correctly. 23.8% · Source for Claude Haiku 5.5 knowledge score (epoch)
Evidence completeness: Broad evidence, 12 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
No category stands out as clearly stronger than its overall position.
Relatively weaker
No category stands out as clearly weaker than its overall position.
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 59th of 77 | 1,454 95% range 1,433 to 1,475 | 789 | Source for Expert-level prompts |
| Hard prompts Kind of request | 72nd of 77 | 1,421 95% range 1,412 to 1,430 | 4,893 | Source for Hard prompts |
| Long prompts Kind of request | 75th of 77 | 1,400 95% range 1,389 to 1,411 | 3,513 | Source for Long prompts |
| Longer conversations Kind of request | 74th of 77 | 1,398 95% range 1,380 to 1,415 | 1,219 | Source for Longer conversations |
| Chinese Language | 73rd of 76 | 1,438 95% range 1,416 to 1,461 | 679 | Source for Chinese |
| English prompts Language | 73rd of 77 | 1,411 95% range 1,399 to 1,422 | 3,089 | Source for English prompts |
| French Language | 66th of 72 | 1,430 95% range 1,396 to 1,464 | 334 | Source for French |
| Non-English prompts Language | 73rd of 77 | 1,387 95% range 1,378 to 1,397 | 4,798 | Source for Non-English prompts |
| Russian Language | 77th of 77 | 1,370 95% range 1,351 to 1,389 | 1,056 | Source for Russian |
| Spanish Language | 69th of 70 | 1,387 95% range 1,353 to 1,420 | 306 | Source for Spanish |
| Entertainment, sport and media Topic | 76th of 77 | 1,352 95% range 1,338 to 1,367 | 2,032 | Source for Entertainment, sport and media |
| Law and government Topic | 75th of 76 | 1,391 95% range 1,367 to 1,416 | 625 | Source for Law and government |
| Medicine and health Topic | 73rd of 75 | 1,372 95% range 1,346 to 1,399 | 534 | Source for Medicine and health |
| Science Topic | 74th of 77 | 1,407 95% range 1,390 to 1,424 | 1,243 | Source for Science |
| Software and IT Topic | 70th of 77 | 1,449 95% range 1,437 to 1,461 | 2,775 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
This model is not among the options we recommend for any of the five tasks. That reflects the rules and the models listed, not a verdict on the model.
Where an alternative might be more suitable
- Writing and creating content: rated Behind the leaders on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Behind the leaders on coding votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: rated Strong on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Business emails and everyday work: rated Behind the leaders on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: rated Behind the leaders on short factual questions. Compare it with Gemini 4 Argon, the best-supported choice for this task.
What it is
- Maker
- Anthropic
- Release date (Epoch AI)
- 7 Oct 2026
- Access, as listed by Epoch AI
- API access
- Licence, as listed by Arena
- Proprietary
- Maker's country (Epoch AI)
- United States of America
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
Epoch has no 'max' run of OTIS Mock AIME for this model; the 'xhigh' run is used and labelled.
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Proprietary. The model's files are not published, so it is generally available only through its maker or partners the maker chooses.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. Epoch AI lists access as “API access”. See models, apps and harnesses and how AI pricing works.
Price
Price not independently verified
What Anthropic's own page says about the Haiku family (USD ($ as shown; the page does not state a currency code); checked 28 Oct 2026)
- Free ($0): Haiku available
- Pro ($20 per month billed monthly, or $17 per month billed annually ($200 up front)): Haiku available
- Max (From $100 per month): Haiku available
The page marks plans against model families, not versions, so it does not say which version of a family a plan includes. Check that before relying on it.
See Claude plans and pricing (checked 28 Oct 2026). Page opened and read.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,401 Arena score 95% range 1,393 to 1,408, 7,890 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,336 Arena score 95% range 1,321 to 1,351, 1,793 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,454 Arena score 95% range 1,440 to 1,469, 1,813 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,434 Arena score 95% range 1,407 to 1,460, 435 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,415 Arena score 95% range 1,403 to 1,427, 2,758 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,405 Arena score 95% range 1,389 to 1,422, 1,340 votes |
Behind the leaders | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,589 Arena score 95% range 1,579 to 1,599, 4,758 votes |
Strong | effort: High, bradley-terry rating | Claude Haiku 5.5 (High) lmarena · webdev/latest |
28 Oct 2026 | Source for Claude Haiku 5.5 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
75.1% share of problems solved standard error ±2.6 points |
Strong | effort: max, best score across scorers | claude-haiku-5-5_max epoch · frontiermath_tiers_1_3_v2.csv |
7 Oct 2026 | Source for Claude Haiku 5.5 FrontierMath Tiers 1-3 (v2) score (epoch) No public run log in the data file. |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
98.9% share of problems solved standard error ±0.5 points |
Strong | effort: xhigh, best score across scorers | claude-haiku-5-5_xhigh epoch · otis_mock_aime_2024_2025.csv |
7 Oct 2026 | Source for Claude Haiku 5.5 OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
89.6% share of questions correct standard error ±2.0 points |
Middle | effort: max, best score across scorers | claude-haiku-5-5_max epoch · gpqa_diamond.csv |
9 Oct 2026 | Source for Claude Haiku 5.5 GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
23.8% share of questions correct standard error ±1.3 points |
Behind the leaders | effort: max, best score across scorers | claude-haiku-5-5_max epoch · simpleqa_verified.csv |
9 Oct 2026 | Source for Claude Haiku 5.5 SimpleQA Verified score (epoch) No public run log in the data file. |
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
46.4% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: max, tool: claude-code, mean@5 | claude-haiku-5-5_max epoch · frontiercode_external.csv |
not provided | Source for Claude Haiku 5.5 FrontierCode (reported by Cognition) score (epoch) · Run details for FrontierCode (reported by Cognition) |
Missing evidence
Not enough data for these tests:
- SWE-bench Verified (Epoch runs)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.