量子化 | Local Model Watch

ExLlamaV3 v1.4.9 Released: Faster CPU Offloading

September 10, 2026Engines and Tools

September 20, 2026

ExLlamaV3 v1.4.9 is released with faster CPU offloading, Intel AVX512BW support, and new model architectures.

September 22, 2026 : Companies and Funding

Hugging Face adds direct execution suppo ...

September 22, 2026 : New Models

Explore MiMo-V2.6-Flash-RL-GGUF, a GGUF ...

September 22, 2026 : Engines and Tools

vLLM v0.30.0 is released, featuring Fast ...

September 22, 2026 : New Models

Download the GGUF version of MiMo-V2.6-D ...

September 22, 2026 : Companies and Funding

NVIDIA Dynamo-Triton 26.07 introduces Te ...