RepoRadar
Leave a Tip
Item detail

alexzhang13/rlm

RLM is an MIT-licensed plug-and-play inference library for Recursive Language Models, a research-to-product release that lets a long-context LLM decompose a huge input (think 10M-token context) into smaller sub-queries it can recursively call itself or sub-models to answer. 4.8k stars, Python, paper at arxiv.org/abs/2512.24601, supports multiple sandboxes (local Python, Modal, Docker) and ships th

Open source →Back to radar
Score8.5
Popularity85.0
Risknone
TierGold
Score breakdown
Usefulness8.0
Novelty9.0
Momentum8.0
Maturity8.4
Open-source/build8.4
Evidence7.2
Workflow potential10.0
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for researchers and engineers who need a working, production-shaped RLM implementation rather than a paper-only sketch: pip-install rlm, point it at a sandbox backend, and let it recursively decompose context-heavy tasks without writing the recursion plumbing yourself.

Who should use it

researchers replicating the RLM paper results against their own long-context corporaengineers building document-QA / codebase-QA products that need to exceed the 1-10M token context limitAI infra teams evaluating recursive inference patterns versus simple RAGML practitioners who want a paper-grade library to fork and adapt for their own agentstudents learning how recursive LLM calls actually work end-to-end

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

No inherent user-impacting risk is flagged from the captured evidence.

Evidence links

Closest alternatives / related signals

recursive-language-modelslong-contextinferencepaper-implementationresearchmitpythonarxiv-2512