v1.3.3

OlivierDehaene released this 15 Dec 00:22

· 653 commits to main since this release

What's Changed

fix gptq params loading
improve decode latency for long sequences two fold
feat: add more latency metrics in forward by @OlivierDehaene in #1346
fix: max_past default value must be -1, not 0 by @OlivierDehaene in #1348

Full Changelog: v1.3.2...v1.3.3

Contributors

OlivierDehaene

Assets 2