Benchmark profile
Virology Capabilities Test (VCT)
A 322-question multimodal benchmark of practical virology troubleshooting knowledge.
Data verified 26 confirmed releases in the last 30 daysStart free briefBenchmark score on VCT — August 12, 2026
BenchLM mirrors the published score view for VCT. Grok 4.6 leads the public snapshot at 67.4% , followed by Grok 4.5 (65.5%). We do not use these results to rank models overall.
Grok 4.6
xAI
grok-4-6
Grok 4.5
xAI
grok-4-5
About VCT
Year
2025
Tasks
322 multimodal virology questions
Format
Accuracy
Difficulty
Expert virology troubleshooting
The paper and benchmark description are public. BenchLM stores xAI's exact model-card results as display-only capability evidence.
BenchLM freshness & provenance
Version
VCT 2025
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 VCT measure?
A 322-question multimodal benchmark of practical virology troubleshooting knowledge.
Which model scores highest on VCT?
Grok 4.6 by xAI currently leads with a score of 67.4% on VCT.
How many models are evaluated on VCT?
2 AI models have been evaluated on VCT on BenchLM.
Compare Top Models on VCT
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.