I Benchmarked Local LLMs on the Laptop I Have
konmam
20 points
2 comments
August 10, 2026
Related Discussions
Found 5 related stories in 93.5ms across 8,687 title embeddings via pgvector HNSW
- Why your local LLM feels dumber than it is felineflock · 265 pts · August 22, 2026 · 60% similar
- The LLM Critics Are Right. I Use LLMs Anyway JeremyTheo · 209 pts · July 16, 2026 · 58% similar
- WebLLM: high-performance in-browser LLM inference engine saikatsg · 103 pts · September 02, 2026 · 58% similar
- LLM Ass Bench fragmede · 147 pts · September 22, 2026 · 58% similar
- LLMs could control their host machines by exploiting inference engines zdw · 117 pts · August 24, 2026 · 56% similar
Discussion Highlights (2 comments)
dpoloncsak
This kinda goes along with my ancedotal findings "Local models are cool but still not quite there for consumer-grade hardware, but getting closer and closer" It just feels, at the moment, there's no task you'd want to throw at this over a frontier model, and while prices are subsidized you really can't compete at home
segmondy
From the article, "One guide this summer was literally titled “Open Weights You Can’t Run.”" I ran Kimi a few days ago at 1/2token per second. I only get to play with it during the weekend when i have time, but I'm certain I'll be able to get it up to 5tk/sec when I'm done in a month or two. So yeah, we can run them all locally. Folks might say it's not run if it's that slow, but feh! If you can run the best AI model locally at 1tk/sec, why won't you?