[SIGMOD'27] Easy Data Preparation with latest LLMs-based Operators and Pipelines.
Project Vyasa is a local-first research execution framework for DGX Spark that helps researchers, journal authors, and domain experts turn unstructured documents into defensible, evidence-bound manuscripts for high-stakes, long-running inquiry. It keeps humans in control of judgment while AI handles extracting, validating, and governing evidence.
A simple LLM gateway for routing requests across multiple LLM servers.
MLA-aware prefix caching for SGLang — exploit latent compression in RadixAttention (DeepSeek based model)
Production-style async serving for image-editing/generation diffusion models via two interchangeable backends — vLLM Omni and SGLang — with FastAPI gateways, Celery + Redis job queues, Docker Compose, and a React frontend.
To associate your repository with the sglang-bankend topic, visit your repo's landing page and select "manage topics."