Synthszr Charts — die großen AI-Marken im Wettkampf ums Podium
synthszr charts
huawei

Huawei · od November 2026 (Paper arXiv:2611.10649 am 7. November 2026; GitHub-Veröffentlichung um den 10. November 2026) · 2× · naposledy 13. 11. 2026

45
Momentum

MACaQ (Model Agnostic Calibrated Quantization) is an open-source quantization tool from Huawei Computing Systems Lab (huawei-csl) for LLMs, MoE models, and diffusion models for image, video, and audio. It quantizes each linear layer independently to 4-bit (MXFP4 W4A4) or produces 2–4-bit GGUF files. It is released under Apache 2.0 on GitHub and integrates with llama.cpp, vLLM, Diffusers, and PyTorch.

Vývoj momenta
17.08.15.11.

Vlastnosti

Throughput/LatencyOnly 2.3% end-to-end overhead vs. pure MXFP4 (Qwen3-32B, vLLM)
LicenseApache License 2.0
PlatformPython/PyTorch; GPU, Ascend CANN kernels via PRs
PriceFree, open source (Apache 2.0)
Protocol Compatibilityllama.cpp (GGUF), vLLM (W4A4/weight-only), Diffusers, PyTorch
Release DateNovember 2026 (paper: Nov 7, 2026)
Supported Models/ProvidersQwen3 (4B, 27B, 30B, 35B, 80B), OLMoE-1B-7B, FLUX.2-klein-4B, Wan2.2-TI2V, Stable Audio 1.0

Další produkty v této kategorii: LLM inference a serving

Zdroje (2)

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.