TRL v1.15: new Triton kernel scales SFT/RL past 100k tokens on one GPU
_lewtun · x · 2026-10-11
Hugging Face's Lewis Tunstall urges TRL users to upgrade to v1.15: a new Triton kernel scales SFT/RL training to over 100k tokens on the same GPU, a significant boost for long-sequence training on single cards.
More from Infra
- AirLLM runs 70B models on a 4GB GPU via layer-wise inference, scaling to 405B on 8GB — JensHonack · 2026-10-11
- Speech Model Shrunk 13x to 153M Params by Looping 2 Shared Blocks — pbaylies · 2026-10-11
- Inference demand went vertical, yet is a flat line next to post-training/RL growth — zainhas · 2026-10-11
- Pat Gelsinger slams HBM as "a lousy memory" wasting four bits for every one it makes — SumitGup · 2026-10-11
- Qualcomm CEO predicts AI phone supercycle, smart glasses as top AI wearable — SuB8u · 2026-10-11
- Zero cold starts: Building and shipping MCP servers with WebAssembly, Spin and Akamai Functions — AI Engineer · 2026-10-11