Gahow Wang f26db882e5 Merge t16-grad-accum into main ...

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

# Conflicts:
#	README.md
#	docs/evolution.md

2026-06-18 00:37:11 +08:00

..

bin

Merge t16-grad-accum into main

2026-06-18 00:37:11 +08:00

ddp.rs

train+ddp: micro-batch gradient accumulation (--accum-steps)

2026-06-17 23:45:33 +08:00

ffi.rs

dist: nccl ffi + comm bootstrap

2026-06-15 17:14:56 +08:00

lib.rs

Revert "dist: coalesce grads into buckets for all-reduce (KI-5)"

2026-06-16 09:39:38 +08:00