BioMysteryBench (overall, v11 release) (BioMysteryBench)
We show this table for reference; we do not rank on it.
Agentic investigation of biological datasets, reported as an overall score on the official v11 release.
Benchmark score on BioMysteryBench — October 29, 2026
We compile the BioMysteryBench rows from provider self-reports. Kimi K3.1 Pro leads the table at 80.6%. We do not use these results to rank models overall.
1 modelAgenticCurrentDisplay onlyUpdated October 29, 2026
Benchmark score table (1 model)
| Rank | Model / configuration | Score | Parameters (B) | Open / closed |
|---|---|---|---|---|
| 1 | Kimi K3.1 ProMoonshot AI | 80.6% | Not reported | Pending |
About BioMysteryBench
Year
2026
Tasks
Biological data investigations
Format
Overall task score
Difficulty
Expert scientific analysis
Moonshot draws the dataset from the official v11 release and uses Kimi K3 as the judge model; harnesses differ by model. BenchLM keeps these provider-run overall scores separate from the human-solvable and human-difficult splits and from the Vals runs.
Freshness and provenance
Version
BioMysteryBench 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.
Questions
What does BioMysteryBench measure?
Agentic investigation of biological datasets, reported as an overall score on the official v11 release.
Which model scores highest on BioMysteryBench?
Kimi K3.1 Pro by Moonshot AI currently leads with a score of 80.6% on BioMysteryBench.
How many models are evaluated on BioMysteryBench?
1 AI models have published results on BioMysteryBench in the BenchLM catalog.
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 6,100+ subscribers.
One email each week. Unsubscribe anytime.