Hacker News users discuss hardware for running local LLMs
Original title: Ask HN: What hardware are you using to run local LLMs?
Community shares Mac Studios, RTX 3090 systems, and GPU clusters for models like DeepSeek-V4-Flash
A Hacker News thread asks what hardware users are using for local LLMs. Responses mention Mac Studios with 512GB RAM, RTX 3090 setups, old Dell T5820 servers with V100 GPUs, and clusters of NVIDIA GPUs. Some note high power draw in clusters and prefer single devices for cost. No exploits or new releases reported.
- Mac Studio M1 Ultra with 128GB RAM
- RTX 3090 with 24GB VRAM
- Dell T5820 with 2x V100 GPUs
- ASUS GX10 cluster with tensor parallelism
- Products
- Mac StudioRTX 3090V100
- AI models
- DeepSeek-V4Qwen3.8
From the original article
8 min read at news.ycombinator.com[](https://news. ycombinator. com/)Hacker Newsnew past comments ask show jobs submitlogin [](https://news.
By leafletsigh · news.ycombinator.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.