TinyFish open-sourced a cookbook of 30+ working web agent recipes, from happy hour scrapers to competitor intelligence dashboards, all built on its browser automation APIs.
The vLLM project ships llm-compressor, a Transformers-native toolkit that turns Hugging Face models into vLLM-ready quantized checkpoints with one API call.
Biohub's open-source ESMFold2 beats AlphaFold3 on antibody-antigen prediction, adds inference-time compute scaling, and predicts proteins with DNA, RNA, and ligands.
Unsloth published GGUF quants of Qwen3.5-4B with multi-token prediction baked in, delivering roughly 1.5x to 2x faster local generation through llama.cpp.
OpenMVS ships v2.4.0 with automatic region-of-interest detection, a rewritten viewer, VGGSFM support, and improved dense reconstruction from sparse SfM output.
LangChain re-open sourced its revamped documentation chatbot, ditching vector embeddings for direct API access and showing how a production Q&A agent handling nearly 2T tokens weekly is built.
A CLI tool that converts document collections into browsable knowledge graphs, extracting entities and relations via LLM with human-approved deduplication.