Le Chonk goes local: open release ships FP8, NVFP4 and GGUF quants
pcuenq · x · 2026-10-29
Hugging Face's pcuenq points to the blog post with full details of the open release, plus local deployment builds: FP8, NVFP4 and GGUF quantizations — jokingly ideal for a '512 GB M5 Studio Ultra'. Developers wanting to run the open model locally can grab the weights via the linked blog post.
Related event: Unsloth and Hugging Face Bring Le Chonk Quantized for Local Deployment(3 posts)→
More from Infra
- Modders unlock hidden AI performance on Nvidia RTX 20/30/40 GPUs, nearly doubling some benchmarks — Resident-Fall4351 · 2026-10-29
- Microsoft discloses Azure's actual revenue for the first time: $33.2 billion — Resident-Fall4351 · 2026-10-29
- Cloudflare wants to charge AI agents per request, eyeing a ~$200B machine-payment prize — LexSokolin · 2026-10-29
- Qwen3-8B on a rented A6000 measured at just 10 tokens/s — huangyun_122 · 2026-10-29
- LM Studio adds oMLX engine to run MLX models natively on Apple Silicon — victormustar · 2026-10-29
- Decommissioned supercomputer with 9,600 A100 GPUs listed on eBay for $59.7M, shipping $50 — siri_1110 · 2026-10-29