Benchmark profile
BioUseBench severity-5 refusal rate (BioUseBench refusal)
The refusal rate on severity-5 biological-use prompts reported in the Grok 4.6 model card.
Data verified 26 confirmed releases in the last 30 daysStart free briefBenchmark score on BioUseBench refusal — August 12, 2026
BenchLM mirrors the published score view for BioUseBench refusal. Grok 4.6 leads the public snapshot at 90.7%. We do not use these results to rank models overall.
Benchmark score table (1 model)
ScoreAbout BioUseBench refusal
Year
2026
Tasks
Severity-5 biological-use prompts
Format
Refusal rate
Difficulty
High-severity bio safety
No public BioUseBench artifact or result page was found. The single exact card result remains display-only.
BenchLM freshness & provenance
Version
BioUseBench refusal 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
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.
FAQ
What does BioUseBench refusal measure?
The refusal rate on severity-5 biological-use prompts reported in the Grok 4.6 model card.
Which model scores highest on BioUseBench refusal?
Grok 4.6 by xAI currently leads with a score of 90.7% on BioUseBench refusal.
How many models are evaluated on BioUseBench refusal?
1 AI models have been evaluated on BioUseBench refusal on BenchLM.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 2,000+ readers.
One email each week. Unsubscribe anytime.