Home Model series Gemini Flash Gemini 3.8 Flash Gemini 3.8 Flash · BioMysteryBench · Human-difficult subset · Scores and configurations Gemini 3.8 Flash · BioMysteryBench · Human-difficult subset · Scores and configurations Gemini 3.8 Flash record for BioMysteryBench · Human-difficult subset. 1 score record for this release, including 1 exact value (may include historical records). Includes 1 source location.
Scores and evaluation conditions Gemini 3.8 Flash · BioMysteryBench scores and configurations Metric and score Configuration and evaluation conditions Sources Accuracy 56.5% Reported 56.5% · Unit % Published Unconfirmed
Gemini 3.8 Flash default sampling Settings Gemini API gemini-3.8-flash; default sampling; pass@1 except where noted. Unknowns Exact sampling parameters are not published. Tools Tools allowed Reasoning budget not recorded Framework not recorded Protocol Self-computed with Linux terminal, bioinformatics tools, Python and R; internet restricted to approved domains. Conditions BioMysteryBench human-difficult subset; pass@1. View this score in rankings
Benchmark versions and metrics Aliases BioMysteryBench human-difficult
Accuracy · % · Higher is better