xserv

Files

Gahow Wang 99b212e6c1 model/sampling: NaN-safe argmax + optional repetition penalty

Make argmax skip NaN logits (warn once) instead of panicking the engine
thread on a single NaN. Add sample_greedy_penalized() applying an
HF-style repetition penalty over recent ids on the greedy path, to break
greedy repetition loops on reasoning models without touching the forward
pass.

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

2026-05-31 00:56:27 +08:00

xserv-cuda

cuda: add cached_trim() to release pooled GPU buffers

2026-05-30 12:50:04 +08:00

xserv-distributed

distributed: NCCL P2P primitives (PpContext + send/recv)

2026-05-29 18:45:42 +08:00

xserv-kernels

kernels: flash attention with gpt-oss sinks + sliding window

2026-05-31 00:56:10 +08:00

xserv-model

model/sampling: NaN-safe argmax + optional repetition penalty