v0.8.16

ruvnet/ruflov0.8.16Jun 20, 2026by EricLBuehler

AI Summary

Optimizes distributed training by enabling NCCL only for world sizes greater than 1 and fixes launch/index math for large MoE expert stacks.

Key Highlights

Enables NCCL only for world size >= 2 to optimize single-GPU setups
Fixes AFQ launch/index math for large MoE expert stacks
Prevents potential index calculation errors in large models

New Features

Distributed training optimization (NCCL)
MoE expert stack math fixes

Full Release Notes

## What's Changed
* feat(distributed): nccl on only for world size >= 2 by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/2257
* fix(afq): fix afq launch/index math for large MoE expert stacks by @EricLBuehler in https://github.com/EricLBuehler/mistral.rs/pull/2258


**Full Changelog**: https://github.com/EricLBuehler/mistral.rs/compare/v0.8.15...v0.8.16