v0.1.19
EricLBuehler/mistral.rsv0.1.19Jun 15, 2024by EricLBuehler
AI Summary
Focuses on memory optimization during loading and code cleanup.
Key Highlights
- Lowered memory spike when loading with ISQ on CUDA
- Removed multiple tracing initializations
- Improved Auto dtype determination
- Added reboot functionality
Full Release Notes
## What's Changed * Format readme by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/427 * Remove multiple tracing initializations and init outside of mistralrs-core by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/428 * Run clippy by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/429 * adding reboot functionality by @gregszumel in https://github.com/EricLBuehler/mistral.rs/pull/378 * Lower memory spike when loading with ISQ on CUDA by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/433 * Fix failing docs workflow by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/435 * Remove unused line in dockerignore by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/436 * Improve `Auto` dtype determination by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/438 * Bump version to 0.1.19 by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/439 **Full Changelog**: https://github.com/EricLBuehler/mistral.rs/compare/v0.1.18...v0.1.19