Diks et al. (scBench team)

scBench-Long: Verifiable Long-Horizon Single-Cell Biology

21 long-horizon single-cell evaluations across five research areas, deterministically graded with trajectory rubrics. The live leaderboard now has Opus 5 at 41.3% with Pi and GPT-5.6 Sol at 38.1% with Pi.

single-celltranscriptomicslong-horizontabularcode
Status
curated
Saturation
unsaturated
Reproduction
not yet
Published results
4 models

Editor's note

21 long-horizon single-cell evaluations across five research areas, deterministically graded with trajectory rubrics. The live leaderboard now has Opus 5 at 41.3% with Pi and GPT-5.6 Sol at 38.1% with Pi.

Published results

ModelScoreVariantProvenance
Opus 541.3% pass ratePi harnessexternal
GPT-5.6 Sol38.1% pass ratePi harnessexternal
Claude Opus 4.825.4% pass rateClaude Code harnesspaper
Gemini 3.5 Flash22.2% pass ratePI harnesspaper

Sources

  • external benchmarks.bio — scBench-Long leaderboard
  • paper scBench-Long (arXiv:2606.26563)