ALL BRIEFS
RANKED · 4 THIS DAYReflection Deploys Beam Model on CoreWeave Infrastructure
Reflection's open-weight Beam model is now available on CoreWeave, spanning training with Weights & Biases Models, supervised fine-tuning in partnership with Axolotl, and Day 0 API access via Serverless Inference on CoreWeave Forge.
Zephon Replaces TorchTitan Data Loader, Cuts GPU-Count Score Variance from 0.82 to 0.05 Points
PyTorch's Zephon data-loading system now integrates with TorchTitan, replacing its default data loader with a pipeline that tokenizes, packs, shuffles, and mixes data inside the training job rather than requiring offline preprocessing.
Spotify Taps Nvidia TensorRT-LLM for Recommendation Inference
Spotify disclosed a collaboration with Nvidia to integrate TensorRT-LLM into its generative recommendation inference pipeline.