Inherent unveils Faraday, a 27B-parameter AI scientist that outperforms Claude Opus 4.8 and GPT-5.5 in paper replication
YesThisIsLion · x · 2026-08-15
Inherent Labs introduces Faraday, a 27B-parameter AI scientist trained via long-horizon RL, outperforming Claude Opus 4.8 and GPT-5.5 on replicating research papers. Their approach involves having an agent reproduce masked figures in popular ML papers by running scaled-down experiments with careful controls. They also use GRPO to train a 27B model as a controller for frontier coding agents, enabling skill adaptation. They believe reproducing existing science is the first step to creative discovery.
More from coding & agent
- How do you debug bad agent runs when you already have tracing? — Sensitive-Parsnip-12 · 2026-10-16
- Where does pre-task context live for agents? Devs debate workflows beyond handoff — shishen2026 · 2026-10-16
- No-instruction autonomous agent wakes up, sets own goals, and tries hacking by default — sethlazar · 2026-10-16
- CODESTRUCT reframes codebases as AST action spaces to fix brittle LLM code agents — burkov · 2026-10-16
- Would you install another SDK for agent debugging on top of your tracing stack? — Sensitive-Parsnip-12 · 2026-10-16
- Law student builds all-in-one AI study platform LexLab Studio with ChatGPT and Muse — Acrobatic_Rhubarb598 · 2026-10-16