Tencent's Hyra 2.0 Generates Protein Sequences 57x Faster Than ProteinMPNN
YvesMulkers · x · 2026-11-04
According to 7wData, Tencent's Hyra 2.0 runs protein sequence generation 57x faster than ProteinMPNN. It is also used to train the next Hunyuan model, with each version bootstrapping the next — raising the question of who audits that loop.
More from Research
- KierstonBench simulates Allentown, PA to test LLMs on crime investigation—only frontier models hold up — Andrew Semenov · 2026-11-04
- Liber-0 hits 36% zero-shot success across 60 robot manipulation tasks without task-specific demos — heyshrutimishra · 2026-11-04
- TACL bans papers submitted to ARR in past 9 months, citing lottery-rejected submissions — EhudReiter · 2026-11-04
- 387 papers distill one question: when is a better model the answer in AI for science? — bravo_abad · 2026-11-04
- Noetra engineer's first project: building an image encoder from scratch for multimodal LLMs — HirokatuKataoka · 2026-11-04
- Japanese team trains image encoder from scratch to match SigLIP2, with better CJK performance — HirokatuKataoka · 2026-11-04