Claude Mythos Preview
Claude Mythos Preview is behind the leaders in reasoning and coding. Too few results yet to rate agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, instruction following, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Agentic, Safety, Long Context, Math, Multimodal, Multilingual, Instruction Following or Factuality.
Price
No current price is tracked for this model. See the rate card
Evidence
25results on23benchmarks
- 25 vendor-reported
From 3 sources · latest Nov 14, 2026 · How verification works
Claude Mythos Preview benchmark results
25 results on 23 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
26.2% behind the leader2 of 6 ranked benchmarks measured
- GPQA Diamond94.60Nov 14, 2026GPQA
- 64.70Nov 14, 2026
28.3% behind the leader3 of 10 ranked benchmarks measured
- 93.90Nov 14, 2026
- 87.30Nov 14, 2026
- 77.80Nov 14, 2026
0 of 7 ranked benchmarks measured
- 79.60Nov 14, 2026
- 86.90Nov 14, 2026
- 85.40Jun 12, 2026
0 of 6 ranked benchmarks measured
- CharXiv (reasoning)93.20Nov 14, 2026CharXiv-R
- CharXiv (reasoning)86.20Jun 12, 2026CharXiv Reasoning (No tools)
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 92.70Nov 14, 2026
- 83.10Nov 14, 2026
- 82.00Nov 14, 2026
- 59.00Nov 14, 2026
- 95.50Jun 12, 2026
- GraphWalks — Parents task, 256K-token context subset99.90Jun 12, 2026GraphWalks Parents 256K subset
Show 9 more resultsHide 9 results
- 87.90Jun 12, 2026
- 92.50Jun 12, 2026
- 74.30Jun 12, 2026
- GraphWalks BFS 256K85.70Jun 12, 2026GraphWalks BFS 256K subset
- 61.10Jun 12, 2026
- 64.70Jun 12, 2026
- HLE (with tools)64.70Jun 12, 2026Humanity's Last Exam (With tools)
- Legal Agent Benchmark, Full Public Set13.40Jun 12, 2026Legal Agent Full Public Benchmark Set
- 43.00Jun 12, 2026
Claude Mythos Preview: common questions
Who makes Claude Mythos Preview?
Claude Mythos Preview is made by Anthropic.
What is Claude Mythos Preview good at?
Claude Mythos Preview is behind the leaders in reasoning and coding. Too few results yet to rate agentic tasks, safety, long context, math, multimodal tasks, multilingual tasks, instruction following, or factuality.
How many benchmarks has Claude Mythos Preview been tested on?
We track 25 results for Claude Mythos Preview on 23 benchmarks from 3 sources. The latest was recorded on Nov 14, 2026.
About this record
Where Claude Mythos Preview's numbers come from, and every name it appears under.
- Tracked since
- Sep 16, 2026
- Newest source mention
- Sep 16, 2026
Where the results come from
Verification: 25 scores · 0 independently verified · 25 vendor-reported. How these tiers are assigned
From 3 sources on 2 sites. www-cdn.anthropic.com supplies 13 of them. Bars are coloured by trust tier.
- www-cdn.anthropic.com13
- api.llm-stats.com12