Research / AI Memory / Long-Context
A Hugging Face paper page describing a memory layer for LLMs that stores KV state to encrypted NVMe and reloads it later without recomputing, with results from 50M-token experiments on Gemma models.
The page clearly presents an AI research result with a concrete system, named package, model setup, and benchmark claims. It includes enough visible detail to understand the approach, the scale tested, and the reported speed and energy gains, making it useful for a public AI directory entry.
The snapshot does not show the paper PDF or full implementation details, so a visitor cannot tell exactly how to install or reproduce the package beyond the brief summary and linked repo/wiki references.
Last evaluated 6 days ago. Current rank #2433. Holding steady in the rankings.
Primus is an autonomous AI researcher that hypothesizes, reads papers, writes code, runs experiments on compute, and drafts research papers. The page shows example tasks, published-paper claims, waitlist access, and positioning for ML research workflows.
An interactive dashboard that analyzes New York Times coverage since 2000 using the NYT Archive API, with views for reporters, beats, sections, subjects, geography, obituaries, and corrections.
An open benchmark and leaderboard for AI CAD agents, with 308 prompts across 20 categories and layered scoring for geometry, engineering, manufacturability, and cognition.
A research article from Applied Compute on how agentic, tool-using workloads differ from traditional LLM benchmarks, with production observations, workload profiles, and an open-source harness for replaying traces.
