v0.8.22

ggml-org/llama.cppv0.8.22Jun 22, 2026by EricLBuehler

AI Summary

This release from the mistral.rs repository improves CUDA matrix handling, error reporting, and release configuration.

Key Highlights

  • Improve CUDA release matrix
  • Better errors
  • Explicit cutile handling

New Features

  • Improve CUDA release matrix
  • Better errors
  • Explicit cutile handling

Full Release Notes

## What's Changed
* feat(cuda): improve release matrix, better errors, explicit cutile handling by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/2274


**Full Changelog**: https://github.com/EricLBuehler/mistral.rs/compare/v0.8.21...v0.8.22