BEND 2.0 benchmark: AlphaGenome tops DNA design evals, multi-objective design remains hard
Andreas Bjerregaard · hf · 2026-11-04
BEND 2.0, released on Hugging Face, benchmarks DNA language models on mutation-effect prediction and sequence design. AlphaGenome leads with a combined score of 136.0 (embeddings readout), well ahead of Nucleotide Transformer, Evo2 20B, and DNABERT-2; linear readouts and post-training improve results, but context-specific and multi-objective design tasks remain far from experimental reference performance.
More from Research
- CellType claims 27B biology model beats all 12 frontier models at predicting cell behavior — alifcoder · 2026-11-05
- No-Prerequisite Information Geometry Tutorial Released With Video and PDF Notes — FrnkNlsn · 2026-11-05
- structRFM: structure-guided RNA foundation model trained on 21M pairs lands in Nature — EricTopol · 2026-11-05
- Yandex open-sources Yambda-5B, a 4.79B-interaction music recommendation dataset — teortaxesTex · 2026-11-05
- Adversarial speculative decoding boosts long-horizon agent success with 0.06% params — kastnerkyle · 2026-11-05
- Nature paper unveils AutoDNA, an AI-native lab where humans only maintain the hardware — EricTopol · 2026-11-05