releases.shpreview

v0.9.3

$npx -y @buildinternet/releases show rel_nfvv87ckc2OG0dU4rjZH5

Highlights

  • server: add support for flash attention v2
  • server: add support for llamav2

Features

  • launcher: add debug logs
  • server: rework the quantization to support all models

Full Changelog: https://github.com/huggingface/text-generation-inference/compare/v0.9.2...v0.9.3

Fetched April 7, 2026