Benchmark profile
Weapons of Mass Destruction Proxy — Biology (WMDP Bio)
The biological hazardous-knowledge split of the public WMDP benchmark.
Data verified 26 confirmed releases in the last 30 daysStart free briefBenchmark score on WMDP Bio — August 12, 2026
BenchLM mirrors the published score view for WMDP Bio. Grok 4.5 leads the public snapshot at 90.9% , followed by Grok 4.6 (90.0%). We do not use these results to rank models overall.
Grok 4.5
xAI
grok-4-5
Grok 4.6
xAI
grok-4-6
About WMDP Bio
Year
2024
Tasks
Hazardous biological-knowledge multiple-choice questions
Format
Accuracy
Difficulty
Biosecurity knowledge
WMDP releases its data and evaluation code publicly. BenchLM keeps model-card comparison rows display-only because protocol details can vary across providers.
BenchLM freshness & provenance
Version
WMDP Bio 2024
Refresh cadence
Annual
Staleness state
Refreshing
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 WMDP Bio measure?
The biological hazardous-knowledge split of the public WMDP benchmark.
Which model scores highest on WMDP Bio?
Grok 4.5 by xAI currently leads with a score of 90.9% on WMDP Bio.
How many models are evaluated on WMDP Bio?
2 AI models have been evaluated on WMDP Bio on BenchLM.
Compare Top Models on WMDP Bio
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.