Evo 2 genomic model now runs on vLLM: 7B fits 1-megabase context on a single H100
anshulkundaje · x · 2026-10-30
Brian Hie's team at Arc Institute announced that the Evo 2 genomic foundation model is now compatible with vLLM, with the open-source evo2flash repo offering a drop-in replacement for evo2.Evo2.
- Runs on a much wider range of GPUs, including RTX 3090, L4, and A100 (available on Google Colab)
- Evo 2 7B can use the full 1-megabase context on a single H100
- fp8 input projections run as fp8 GEMMs on sm89+ GPUs (L4/L40S/H100/H200); older cards need a fallback path, and 40B requires an H200 or fp8 weights on an H100
- The goal is making genomic in-context learning accessible across diverse biological systems
More from Infra
- llmman: run Claude Code, Codex and other agents on any local model with one command — weskambale · 2026-10-30
- How Instagram cut feed load time by 50% and Discord stores 100B+ messages — _jaydeepkarale · 2026-10-30
- YC-backed VirajAero launches turbine that powers datacenters with 40% more power from same fuel — Scobleizer · 2026-10-30
- Proof of Agent rolls out 15-minute checkpoints anchored to Solana and Base, capping validation costs — adamamcbride · 2026-10-30
- Pooling $8K for a shared local LLM rig takes 3.3 years to beat a $20/mo ChatGPT sub — Any-Blacksmith-2054 · 2026-10-30
- 10 coding agents tested on local Qwen 3.8 27B: Pi wins on efficiency at 9.2k tokens per task — latent_lurker8587 · 2026-10-30