Huawei releases MACaQ: MXFP4 W4A4 PTQ compatible with llama.cpp and vLLM
huawei-csl · hf · 2026-11-11
MACaQ is a model-agnostic post-training quantization method from Huawei that pushes weights and activations down to MXFP4 W4A4.
- How it works: quantizes each linear layer independently using online activation transforms and Hessian-aware refinement, cutting cost while preserving accuracy.
- Scope: works across language and diffusion models; the announcement claims one quantizer covering text, image, audio, video, and 3D.
- Ecosystem: claims compatibility with llama.cpp, vLLM, and Diffusers.
More from Infra
- Amazon Prime 360 uses AI to rebuild NBA plays in 3D for live, any-angle replays — import_jmr · 2026-11-11
- Altman says OpenAI aims to replace Nvidia in the medium to long term — zephyr_z9 · 2026-11-11
- Chinese University Bought Nvidia H200 Chips Despite US Blacklist — pstAsiatech · 2026-11-11
- Baidu's ERNIE Evolve tops public bests on NanoGPT Speedrun, brings RSI to real inference optimization — xiaohu · 2026-11-11
- Baidu says RSI auto-optimized ERNIE inference, 6.24x decode speedup on Kunxin P800 — xiaohu · 2026-11-11
- Nvidia's supply commitments jump from $119B to $279B in one quarter, locking up 2027-28 memory — tengyanAI · 2026-11-11