Weight compression on the GPU: researcher proves his own instinct wrong
realDanFu · x · 2026-10-30
Dan Fu shares a project that defied his own first instinct — "that sounds cool, but there's no way it will actually work." He assumed compressing weights on disk was fine but GPU decompression overhead would kill it. The idea draws on old CPU compression techniques, and turned out to be feasible.
More from Infra
- Rogers Expands Starlink Satellite Roaming to Japan, Australia and New Zealand — XFreeze · 2026-10-30
- Corey Quinn flags macOS update silently restoring the removed eMudhra root cert — kingslyj · 2026-10-30
- NVIDIA shares guide on maximizing memory efficiency and cutting costs on Jetson — NVIDIAAI · 2026-10-30
- Rebuild NVIDIA's speaker-captions demo with one transformers serve command — yb2698 · 2026-10-30
- Cap moves transcription to its own GPU farm, up to 11x faster — JosephJacks_ · 2026-10-30
- Prime Intellect makes DeepSeek's MegaMoE trainable, up to 55% faster MoE training — samsja19 · 2026-10-30