Skip to main content
BenchLM
Data

Cybench

We show this table for reference; we do not rank on it.

Data verified

A cybersecurity benchmark of professional Capture the Flag tasks for measuring autonomous cyber agent capability and risk.

Benchmark score on Cybench — October 17, 2026

We compile the Cybench rows from provider self-reports. Mistral Large 4 leads the table at 93.0%, followed by Muse Spark 1.1 (92.9%). We do not use these results to rank models overall.

2 modelsAgenticCurrentDisplay onlyUpdated October 17, 2026

Benchmark score table (2 models)

Benchmark score results for Cybench
RankModel / configurationScoreParameters (B)Open / closed
1Mistral Large 4Mistral
93.0%
Not reportedPending
2Muse Spark 1.1Meta
92.9%
Not reportedClosed

About Cybench

Year

2025

Tasks

40 professional CTF tasks

Format

Cybersecurity agent task completion

Difficulty

Professional cybersecurity

Cybench packages recent professional CTF tasks with execution environments and subtasks. BenchLM tracks provider-reported Cybench rows as display-only security-agent capability signals.

Freshness and provenance

Version

Cybench 2025

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Questions

What does Cybench measure?

A cybersecurity benchmark of professional Capture the Flag tasks for measuring autonomous cyber agent capability and risk.

Which model scores highest on Cybench?

Mistral Large 4 by Mistral currently leads with a score of 93.0% on Cybench.

How many models are evaluated on Cybench?

2 AI models have published results on Cybench in the BenchLM catalog.

Compare top models on Cybench

Last updated: October 17, 2026 · BenchLM version Cybench 2025

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 5,600+ readers.

One email each week. Unsubscribe anytime.