Show HN: An MCP proxy that sends 97% fewer tool-definition tokens per request

Hacker News Show2026-11-04 03:250 views

Points: 1

# Comments: 0

Running massive 30B+ Mixture-of-Experts (MoE) models locally usually requires dual RTX 3090/4090s (48GB VRAM) or costly

Pretend the topic is the most boring thing it could possibly be, and you're still super pumped. Should I consult some so

https://github.com/liyu1981/my-videostation my-videostation re-creates the experience of synology videostation in open s

Article URL: https://hecer.github.io/llvm-vulkan/ Comments URL: https://news.ycombinator.com/item?id=50359400 Points: 1

Article URL: https://github.com/argeuthiesen/claudinho Comments URL: https://news.ycombinator.com/item?id=50359514 Point

A few weeks ago I wrote hundreds of hypotheses for crypto trading strategies, re-screened every bot from my last post (

Related posts