Tag
vLLM v0.21.0rc1 is a pre-release update for the high-performance LLM inference and serving library, featuring optimizations for throughput, quantization, and hardware support.
The author shares benchmark results for memweave, a Python library for agent memory, achieving 98% Recall@5 on LongMemEval-S using only local embeddings without LLM calls. The post details the methodology and compares performance against mempalace, highlighting stable retrieval across different question types.
ExLlamaV3 has released a series of major updates including Gemma 4 support, improved caching efficiency, and the new DFlash technology for significantly faster inference speeds across various model categories.
A Python library for calculating ephemerides and spectral data, hosted on PyPI.
nature-skills is a scientific visualization tool based on matplotlib. It features built-in Nature-style templates, supports various chart types, and outputs editable SVG format, aiming to streamline the figure creation process for scientific papers.
A Python library that scrapes websites while bypassing bot protections like Cloudflare and Akamai by rotating fingerprints, mimicking browser headers, and automatically handling CAPTCHAs. It uses Headless Chromium, Playwright, proxy rotation, and is fully open-source.
OpenAI releases version 2.35.0 of the openai-python library, featuring updated documentation for the new Responses API and examples using the GPT-5.2 model. The update also highlights workload identity authentication methods for secure cloud environments.
OpenAI Python library v2.32.0 release - a Python API client for accessing OpenAI REST APIs with type definitions, sync/async support, and new workload identity authentication options.
Sentence Transformers v5.4 introduces support for multimodal embedding and reranking, allowing users to encode and compare text, images, audio, and video using a unified API.
OpenAI Python library v2.30.0 release providing convenient access to OpenAI REST API with type definitions, sync/async clients, and support for workload identity authentication across Kubernetes, Azure, and Google Cloud Platform.
OpenAI releases GABRIEL, an open-source toolkit that uses GPT to convert unstructured qualitative data (text, images) into quantitative measurements for social scientists and economists. The tool enables researchers to analyze large-scale qualitative datasets more efficiently by automating repetitive labeling tasks while preserving the richness of human data.
OpenAI open-sources mujoco-py, a high-performance Python library for robotic simulation using the MuJoCo engine, featuring ~40x speedup with headless GPU rendering and VR interaction support.