Open source

Libraries, tools and repos: releases you depend on and projects gaining stars.

Verify using the pinned commit hash (recommended):

Converting dense models into Mixture-of-Experts

For the past few weeks I've been trying out converting existing dense models to sparse Mixture-of-Experts models, with no pretraining from scratch.

GitHub: huggingface/trl4d ago

huggingface/trl v1.15.0

SFT, DPO, KTO, GRPO, RLOO and Distillation now score tokens with a fused LM head: a Triton kernel projects the hidden states through the LM head in tiles and reduces to per-token log-probs and entropy directly, so the [batch, seq, vocab] logits tensor is never built.

Code

GitHub: unslothai/unsloth4d ago

unslothai/unsloth v0.1.905-beta: Sandboxing is here!

We're introducing Windows, Mac and Linux sandboxing in Unsloth!

Code

NVIDIA Technical Blog5d ago

Faster Scientific Image Analysis with NVIDIA cuPhoton

Observatories and telescopes, lasers and X-ray light sources, and other high-throughput instruments generate image data faster than CPU-bound pipelines can...Observatories and telescopes, lasers and X-ray light sources, and other high-throughput instruments generate image data faster than CPU-bound pipelines can process it to support timely decisions.

GitHub: vllm-project/vllm7d ago

vllm-project/vllm v0.31.0

Fast restart: the new vllm preload CLI launches the weight-cache daemon that keeps post-quantized weights resident in GPU memory across engine restarts (#56680), now with data parallelism (#57386), MTP draft models (#57312), a /health endpoint (#58552) and a readiness wait (#58370).

Code

Hugging Face trending models7d ago