Claude Mythos Preview

Proprietary

Benchmarks

About Claude Mythos Preview

Claude Mythos Preview is an AI model developed by Anthropic, released in 2026, made available as a proprietary (API-only) model. Claude Mythos Preview is Anthropic's most capable model to date, made available only through "Project Glasswing" to roug.

On this page you'll find Claude Mythos Preview's full specifications, including 2 benchmark results. Review provider pricing and benchmark scores below, or compare Claude Mythos Preview head-to-head with other AI models.

→ More models by Anthropic → AI benchmarks leaderboard

Links

Announcement ↗

Community

Would you run Claude Mythos Preview again?

No votes yet — be the first.

Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Claude Mythos Preview benchmark scores▶

Compare on the benchmark table →

Benchmark	Category	Score	Variant
SWE-bench Pro	code	69.8%	preview
CWE-Bench-Java	cybersecurity	71.5%

More models like Claude Mythos Preview

Frequently asked questions

What is Claude Mythos Preview? ▶

Claude Mythos Preview is Anthropic's most capable model to date, made available only through "Project Glasswing" to roug It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Claude Mythos Preview? ▶

Claude Mythos Preview was developed by Anthropic and released in 2026.

Is Claude Mythos Preview open source or proprietary? ▶

Claude Mythos Preview is a proprietary model. It is accessed through an API rather than by downloading the weights.

How much does Claude Mythos Preview cost? ▶

Pricing for Claude Mythos Preview depends on the provider. See the providers table on this page for the latest API rates.

How does Claude Mythos Preview perform on benchmarks? ▶

Claude Mythos Preview is benchmarked across 2 evaluations on GenAIList, including SWE-bench Pro (69.8%). See the full benchmark table below and compare it with other models.