Developer Tool / LLM Serving / vLLM
An out-of-tree vLLM plugin that compresses KV cache with STAR-KV to fit more concurrent requests and longer contexts on the same GPU memory budget.
The page clearly describes an AI infrastructure tool with a concrete use case: improving vLLM serving capacity through low-rank KV-cache compression. It includes install steps, a quick start, and measured results showing concurrency and context-length gains on real GPU hardware, which is enough for a useful public listing.
The crawl only shows one repository page, so a visitor still cannot quickly assess maintenance status, release packaging, compatibility beyond vLLM 0.10.1.1, or whether the private reference checkpoint is available to others.
Last evaluated 19 days ago. Current rank #1038. Down 3 spots in the rankings.
A Rust Blu-ray disc analyzer that reimplements BDInfo for BDMV folders and ISO images, with CLI, desktop, and browser/WASM front ends.
WebLLM is a high-performance in-browser LLM inference engine that runs locally in the browser with WebGPU acceleration. It exposes an OpenAI-compatible API, supports streaming and JSON mode, and includes examples for building chat apps and browser extensions.
Molt is an open protocol for letting an AI agent make bounded purchases through disposable, single-use payment credentials. The repo includes a clear quickstart, MCP integration, a threat model, and a working test-mode demo.
Self-hosted observability and control plane for production and coding AI agents, with live timelines, fleet-wide feeds, token budgets, MCP allow/block rules, and support for Claude Code plus a Python sensor.
