1.
2.
Clean up Claude 5's token vomit with a separate LLM
(news.ycombinator.com)
3.
Show HN: Distilling DeepSeek into GPT-OSS doesn't transfer censorship. Try it
(news.ycombinator.com)
4.
Running local models is good now
(news.ycombinator.com)
5.
Show HN: Find the best local LLM for your hardware, ranked by benchmarks
(news.ycombinator.com)
6.
Making LLM Training Faster with Unsloth and NVIDIA
(news.ycombinator.com)
7.
8.
Lemonade by AMD: a fast and open source local LLM server using GPU and NPU
(news.ycombinator.com)
9.