github
GitHub APIExtracts: repo, release, stars delta · metadata or bounded excerpt
Retention: D1 event history with canonical source link and deduplication metadata.
Access basis: Public API; token recommended for production rate limits.
Last ingested: 2026-08-25 · run status: success with data
#### MoE * **Quantile Balancing router:** adds auxiliary-loss-free load balancing using per-expert routing biases derived from quantile estimates ([\#5349](https://github.com/NVIDIA/Megatron-LM/pull/5349)). * **Fused shared-expert MLP:** enables shared experts through group
NVDAgithub:NVIDIA/Megatron-LM# Patch release v5.15.1 This patch most notably solves a few issues with DFlash and MTP candidate generators, as well as an issue where images could sometimes not be processed on accelerator if using Lanczos filter. It contains the following commits: - Fix DFlash candida
github:huggingface/transformers