27B Model Compressed to 3.8GB
victormustar · x · 2026-07-15
It was shared that Bonsai 27B utilizes 1-bit quantization to shrink its size from 54GB to 3.8GB, reportedly retaining 90% of its intelligence.
The post also notes that, powered by custom WebGPU kernels written using Fable 5 and GPT 5.6 Sol, the model can now run locally directly in the browser.
Related event: Bonsai 27B: The first 27B model that runs on phones(15 posts)→
More from Infra
- Engram's random reads don't suit SSDs; CPU-memory over NVLink could serve all 72 GPUs — bookwormengr · 2026-09-11
- 80% of the DIY LLM inference hype posters have already quit — it's brutally hard systems work — abhijithneil · 2026-09-11
- Hugging Face's Ultra Scale Playbook: a free book on training LLMs on GPU clusters — mdancho84 · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- LLM Serving Metrics Thread: Why TPOT and Uptime Make or Break User Experience — abhijithneil · 2026-09-11
- PlanetScale launches sharded Postgres: 768 servers acting as one, 1PB scale — dhruv2038 · 2026-09-11