Read original ↗

repoGitHubTrust 82 · PrimaryPublished 2d agoLive · 2d ago

NVIDIA/kvpress

LLM KV cache compression made easy

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

PossiblePossibly related (embedding) · 67%DKV: Open-source KV-cache compression framework for local LLM inference (CLI + technical report) →
PossiblePossibly related (embedding) · 53%Reducing High-Bandwidth Memory Bottlenecks in JAX-Based LLM Training with Host Offloading - NVIDIA Developer →
PossiblePossibly related (embedding) · 52%I mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset) →

Covers

newsDKV: Open-source KV-cache compression framework for local LLM inference (CLI + technical report)newsReducing High-Bandwidth Memory Bottlenecks in JAX-Based LLM Training with Host Offloading - NVIDIA Developer newsI mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset)

Related across the graph

newsReducing High-Bandwidth Memory Bottlenecks in JAX-Based LLM Training with Host Offloading - NVIDIA Developer newsI mapped which local LLMs actually fit each RAM tier, 8 to 128GB (open dataset)newsDKV: Open-source KV-cache compression framework for local LLM inference (CLI + technical report)

Topics

inference kv-cache kv-cache-compression large-language-models llm long-context python pytorch transformers