Claude Sonnet 4.6
Made by Anthropic, released 17 Feb 2026. Epoch AI lists its access as “API access”. Appears strongest at coding among the models listed here.
In short
What it does well
- Coding: Top tier (17th of 77 listed models)
Where it falls short
- Knowledge: Behind the leaders (39th of 43 listed models)
- Writing: Middle (47th of 77 listed models)
What we don't know yet
- Hard maths problems: 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
- MiddleHow often people preferred its answers to writing prompts. 1,435 on Arena's scale · Source for Claude Sonnet 4.6 writing score (lmarena)
- Coding
- Top tierHow often people preferred its answers to coding questions. 1,503 on Arena's scale · Source for Claude Sonnet 4.6 coding score (lmarena)
- Maths
- Not enough dataNo result yet for hard maths problems.
- Knowledge
- Behind the leadersHow many short factual questions it answered correctly. 32.8% · Source for Claude Sonnet 4.6 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
- German: 49th of 55 on questions asked in German, against 36th of 77 overall. Source for German
- Korean: 41st of 54 on questions asked in Korean, against 36th of 77 overall. Source for Korean
- Russian: 57th of 77 on questions asked in Russian, against 36th of 77 overall. Source for Russian
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 25th of 77 | 1,500 95% range 1,492 to 1,508 | 7,392 | Source for Expert-level prompts |
| Hard prompts Kind of request | 26th of 77 | 1,484 95% range 1,480 to 1,489 | 46,219 | Source for Hard prompts |
| Long prompts Kind of request | 27th of 77 | 1,479 95% range 1,474 to 1,484 | 31,064 | Source for Long prompts |
| Longer conversations Kind of request | 37th of 77 | 1,463 95% range 1,457 to 1,470 | 12,685 | Source for Longer conversations |
| Chinese Language | 51st of 76 | 1,491 95% range 1,482 to 1,501 | 4,514 | Source for Chinese |
| English prompts Language | 32nd of 77 | 1,472 95% range 1,468 to 1,477 | 31,970 | Source for English prompts |
| French Language | 47th of 72 | 1,465 95% range 1,451 to 1,479 | 2,271 | Source for French |
| German Language | 49th of 55 | 1,428 95% range 1,410 to 1,445 | 1,220 | Source for German |
| Japanese Language | 36th of 54 | 1,420 95% range 1,398 to 1,442 | 802 | Source for Japanese |
| Korean Language | 41st of 54 | 1,410 95% range 1,391 to 1,428 | 1,150 | Source for Korean |
| Non-English prompts Language | 44th of 77 | 1,440 95% range 1,435 to 1,444 | 38,942 | Source for Non-English prompts |
| Russian Language | 57th of 77 | 1,440 95% range 1,432 to 1,447 | 7,987 | Source for Russian |
| Spanish Language | 32nd of 70 | 1,463 95% range 1,449 to 1,477 | 2,161 | Source for Spanish |
| Entertainment, sport and media Topic | 41st of 77 | 1,431 95% range 1,425 to 1,437 | 15,926 | Source for Entertainment, sport and media |
| Law and government Topic | 36th of 76 | 1,470 95% range 1,461 to 1,479 | 5,702 | Source for Law and government |
| Medicine and health Topic | 52nd of 75 | 1,450 95% range 1,441 to 1,459 | 5,438 | Source for Medicine and health |
| Science Topic | 35th of 77 | 1,475 95% range 1,468 to 1,481 | 11,654 | Source for Science |
| Software and IT Topic | 29th of 77 | 1,491 95% range 1,486 to 1,496 | 27,747 | 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
- Homework and maths: not ranked, because there is not enough data for the main test.
Where an alternative might be more suitable
- Writing and creating content: rated Middle on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: no result on hard maths problems, so we cannot say. See which models the evidence does support.
- Business emails and everyday work: rated Strong 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)
- 17 Feb 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
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 Sonnet family (USD ($ as shown; the page does not state a currency code); checked 28 Oct 2026)
- Free ($0): Sonnet available
- Pro ($20 per month billed monthly, or $17 per month billed annually ($200 up front)): Sonnet available
- Max (From $100 per month): Sonnet 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) |
1,458 Arena score 95% range 1,454 to 1,461, 70,916 votes |
Strong | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,435 Arena score 95% range 1,429 to 1,442, 12,351 votes |
Middle | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,503 Arena score 95% range 1,498 to 1,509, 19,723 votes |
Top tier | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,462 Arena score 95% range 1,452 to 1,472, 3,795 votes |
Middle | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,476 Arena score 95% range 1,471 to 1,481, 23,684 votes |
Strong | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,464 Arena score 95% range 1,458 to 1,470, 14,057 votes |
Strong | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · text/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,523 Arena score 95% range 1,517 to 1,528, 22,733 votes |
Middle | effort: not stated, bradley-terry rating | Claude Sonnet 4.6 lmarena · webdev/latest |
28 Oct 2026 | Source for Claude Sonnet 4.6 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) | Not enough data | |||||
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
71.1% share of problems solved standard error ±6.8 points |
Behind the leaders | effort: max, best score across scorers | claude-sonnet-4-6_max epoch · otis_mock_aime_2024_2025.csv |
6 Aug 2026 | Source for Claude Sonnet 4.6 OTIS Mock AIME 2024-2025 score (epoch) · Run details for OTIS Mock AIME 2024-2025 |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
78.8% share of questions correct standard error ±2.9 points |
Behind the leaders | effort: max, best score across scorers | claude-sonnet-4-6_max epoch · gpqa_diamond.csv |
6 Aug 2026 | Source for Claude Sonnet 4.6 GPQA Diamond score (epoch) · Run details for GPQA Diamond |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
32.8% share of questions correct standard error ±1.5 points |
Behind the leaders | effort: max, best score across scorers | claude-sonnet-4-6_max epoch · simpleqa_verified.csv |
27 Aug 2026 | Source for Claude Sonnet 4.6 SimpleQA Verified score (epoch) · Run details for SimpleQA Verified |
| SWE-bench Verified (Epoch runs) SWE-bench Verified, benchmark released 2024-08-13, as run by Epoch AI |
75.2% share of issues resolved standard error ±2.0 points |
Middle | effort: not stated, best score across scorers | claude-sonnet-4-6 epoch · swe_bench_verified.csv |
21 Feb 2026 | Source for Claude Sonnet 4.6 SWE-bench Verified (Epoch runs) score (epoch) · Run details for SWE-bench Verified (Epoch runs) |
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
24.3% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: max, tool: claude-code, mean@5 | claude-sonnet-4-6_unknown epoch · frontiercode_external.csv |
not provided | Source for Claude Sonnet 4.6 FrontierCode (reported by Cognition) score (epoch) · Run details for FrontierCode (reported by Cognition) |
Missing evidence
Not enough data for these tests:
- FrontierMath Tiers 1-3 (v2)
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.