DarkLight1337/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
PhD Candidate @ HKUST
A high-throughput and memory-efficient inference and serving engine for LLMs
A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM
A personal FoundryVTT module containing homebrew items.
Minkowski Engine is an auto-diff neural network library for high-dimensional sparse tensors
Command-line tool for checking Python compatibility
LiDAR point cloud ground filtering / segmentation (bare earth extraction) method based on cloth simulation
A framework for few-shot evaluation of language models.
LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via Hybrid Architecture
LlamaIndex is a data framework for your LLM applications
A specification that python filesystems should adhere to.
Command-line tool for organizing images
Push/pull templates to/from a wiki using AutoWikiBrowser.