ToolAI.io · GitHub Channel
A fact-based index of LLM, agent, MCP, RAG and AI development projects, with license, setup, download, screenshots and related resources for each entry.
Public repository data
llm-d/llm-d
llm-d is an open-source serving stack that adds distributed orchestration, routing, cache management, autoscaling, and batch processing around model servers such as vLLM and SGLang. It targets high-scale production inference on Kubernetes and modern hardware accelerators.
PaddlePaddle/FastDeploy
High-performance Inference and Deployment Toolkit for LLMs and VLMs based on PaddlePaddle
vllm-project/vllm-ascend
A community-maintained hardware plugin that enables vLLM to run model inference and serving workloads on supported Ascend NPUs.
dstackai/dstack
A vendor-agnostic control plane for provisioning and orchestrating training, inference, development, and agentic workloads across GPU clouds, Kubernetes, and on-premises infrastructure.
mozilla-ai/any-llm
any-llm is a Python SDK for communicating with multiple LLM providers through one interface. It supports switching providers and models with minimal code changes while using official provider SDKs when available.
theopenco/llmgateway
Route, manage, and analyze your LLM requests across multiple providers with a unified API interface.
xllm-ai/xllm
xLLM is a C++ inference framework for LLM, VLM, DiT and REC models. It targets high-throughput, low-latency deployment on several AI accelerator families and separates service-layer scheduling and availability from engine-layer computation.
deepspeedai/deepspeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
NVIDIA/TensorRT
NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT.
avifenesh/memra
A from-scratch LLM inference engine built in Rust and CUDA, specifically optimized for RTX 5090 (sm_120a) and H100 (sm_90a) architectures, delivering exactness-gated performance without relying on frameworks like ggml.
PaddlePaddle/Paddle-Lite
PaddlePaddle High Performance Deep Learning Inference Engine for Mobile and Edge (飞桨高性能深度学习端侧推理引擎）
ollama/ollama
A tool for running and managing large language models locally.
Page 2 / 4 · 46 projects
