Claude Opus 4.8
Made by Anthropic, released 28 May 2026. Epoch AI lists its access as “API access”. Appears solid at coding, knowledge, maths and writing among the models listed here.
In short
What it does well
- Coding: Strong (31st of 77 listed models)
- Knowledge: Strong (18th of 43 listed models)
- Maths: Strong (14th of 43 listed models)
- Writing: Strong (26th of 77 listed models)
Where it falls short
The four main ratings show no weakness. That is not evidence it has none.
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
- StrongHow often people preferred its answers to writing prompts. 1,455 on Arena's scale · Source for Claude Opus 4.8 writing score (lmarena)
- Coding
- StrongHow often people preferred its answers to coding questions. 1,491 on Arena's scale · Source for Claude Opus 4.8 coding score (lmarena)
- Maths
- StrongHow many very hard maths problems it solved. 80.0% · Source for Claude Opus 4.8 maths score (epoch)
- Knowledge
- StrongHow many short factual questions it answered correctly. 53.0% · Source for Claude Opus 4.8 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
- Long prompts: 17th of 77 on long, detailed prompts, against 35th of 77 overall. Source for Long prompts
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 | 22nd of 77 | 1,503 95% range 1,495 to 1,511 | 7,061 | Source for Expert-level prompts |
| Hard prompts Kind of request | 30th of 77 | 1,482 95% range 1,478 to 1,487 | 42,464 | Source for Hard prompts |
| Long prompts Kind of request | 17th of 77 | 1,484 95% range 1,479 to 1,489 | 30,147 | Source for Long prompts |
| Longer conversations Kind of request | 25th of 77 | 1,476 95% range 1,469 to 1,482 | 11,189 | Source for Longer conversations |
| Chinese Language | 40th of 76 | 1,506 95% range 1,496 to 1,516 | 4,342 | Source for Chinese |
| English prompts Language | 35th of 77 | 1,469 95% range 1,464 to 1,474 | 28,015 | Source for English prompts |
| French Language | 30th of 72 | 1,480 95% range 1,466 to 1,494 | 2,185 | Source for French |
| German Language | 22nd of 55 | 1,471 95% range 1,452 to 1,490 | 1,082 | Source for German |
| Japanese Language | 27th of 54 | 1,442 95% range 1,420 to 1,464 | 899 | Source for Japanese |
| Korean Language | 24th of 54 | 1,431 95% range 1,411 to 1,450 | 1,107 | Source for Korean |
| Non-English prompts Language | 33rd of 77 | 1,450 95% range 1,446 to 1,455 | 35,972 | Source for Non-English prompts |
| Russian Language | 24th of 77 | 1,475 95% range 1,467 to 1,483 | 7,106 | Source for Russian |
| Spanish Language | 30th of 70 | 1,465 95% range 1,450 to 1,479 | 1,858 | Source for Spanish |
| Entertainment, sport and media Topic | 33rd of 77 | 1,441 95% range 1,434 to 1,447 | 16,141 | Source for Entertainment, sport and media |
| Law and government Topic | 29th of 76 | 1,475 95% range 1,466 to 1,484 | 5,423 | Source for Law and government |
| Medicine and health Topic | 37th of 75 | 1,464 95% range 1,455 to 1,474 | 4,762 | Source for Medicine and health |
| Science Topic | 33rd of 77 | 1,477 95% range 1,471 to 1,484 | 10,434 | Source for Science |
| Software and IT Topic | 30th of 77 | 1,488 95% range 1,483 to 1,493 | 25,108 | 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 Strong on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Strong 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 Strong on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: rated Strong 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)
- 28 May 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 Opus family (USD ($ as shown; the page does not state a currency code); checked 28 Oct 2026)
- Free ($0): Opus not available
- Pro ($20 per month billed monthly, or $17 per month billed annually ($200 up front)): Opus available
- Max (From $100 per month): Opus 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) |
1,461 Arena score 95% range 1,457 to 1,465, 63,994 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,455 Arena score 95% range 1,448 to 1,461, 12,718 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,491 Arena score 95% range 1,485 to 1,497, 17,403 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,485 Arena score 95% range 1,474 to 1,496, 2,970 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,477 Arena score 95% range 1,472 to 1,482, 22,619 votes |
Top tier | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,458 Arena score 95% range 1,452 to 1,465, 12,742 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · text/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,556 Arena score 95% range 1,550 to 1,562, 18,408 votes |
Strong | effort: High, bradley-terry rating | Claude Opus 4.8 (High) lmarena · webdev/latest |
28 Oct 2026 | Source for Claude Opus 4.8 Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
80.0% share of problems solved standard error ±2.4 points |
Strong | effort: max, best score across scorers | claude-opus-4-8_max epoch · frontiermath_tiers_1_3_v2.csv |
10 Jun 2026 | Source for Claude Opus 4.8 FrontierMath Tiers 1-3 (v2) score (epoch) · Run details for FrontierMath Tiers 1-3 (v2) |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
98.3% share of problems solved standard error ±1.4 points |
Strong | effort: max, best score across scorers | claude-opus-4-8_max epoch · otis_mock_aime_2024_2025.csv |
7 Jun 2026 | Source for Claude Opus 4.8 OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
91.0% share of questions correct standard error ±1.9 points |
Strong | effort: max, best score across scorers | claude-opus-4-8_max epoch · gpqa_diamond.csv |
7 Jun 2026 | Source for Claude Opus 4.8 GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
53.0% share of questions correct standard error ±1.6 points |
Strong | effort: max, best score across scorers | claude-opus-4-8_max epoch · simpleqa_verified.csv |
27 Aug 2026 | Source for Claude Opus 4.8 SimpleQA Verified score (epoch) · Run details for SimpleQA Verified |
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
46.5% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: max, tool: claude-code, mean@5 | claude-opus-4-8_unknown epoch · frontiercode_external.csv |
not provided | Source for Claude Opus 4.8 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.