Inferentie, implementatie en runtime
NVIDIA/Model-Optimizer
A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
★ 3,3K⑂ 513Python
Apache-2.0Q86
Robotica & Edge AI
NVIDIA/warp
Compiles typed Python functions into compute kernels with geometry, physics and differentiable programming capabilities for simulation and machine learning.
★ 7,1K⑂ 621Python
Apache-2.0Q92
Fine-tuning, training en data
NVIDIA/Megatron-LM
Ongoing research training transformer models at scale
★ 17,2K⑂ 4,3KPython
NOASSERTIONQ86
Agents en multi-agents
openai/openai-agents-js
A lightweight, powerful framework for multi-agent workflows and voice agents
★ 3,5K⑂ 871TypeScript
MITQ98
Data-engineering en pipelines
NVIDIA/aistore
AIStore: scalable storage for AI applications
★ 1,9K⑂ 277Go
MITQ94
Robotica & Edge AI
google-deepmind/mujoco
A multijoint contact dynamics engine maintained by Google DeepMind, providing model description, physics stepping, an interactive viewer, and a Python interface. It is suitable for developing robotics, control, and reinforcement learning environments.
★ 15,1K⑂ 1,7KC++
Apache-2.0Q92
Fine-tuning, training en data
hydra-ecosystem/hydra
A configuration framework for Python applications that manages experiments through configuration composition, command-line overrides, and multirun execution. It is suitable for organizing data, model, and training parameters into reusable configuration structures.
★ 10,6K⑂ 977Python
MITQ92
Multimodale AI
Tencent-Hunyuan/UniRL
UniRL is a Framework for Unified Multimodal Model Reinforcement Learning
★ 854⑂ 56Python
NOASSERTIONQ94
LLM en foundationmodellen
PaddlePaddle/PaddleFormers
PaddleFormers is an easy-to-use library of pre-trained large language model zoo based on PaddlePaddle.
★ 13K⑂ 2,2KPython
Apache-2.0Q81
AI-programmeer- en ontwikkelaarstools
xai-org/grok-build
SpaceXAI's coding agent harness and TUI. Fullscreen, mouse interactive, extensible.
★ 22,6K⑂ 4,3KRust
Apache-2.0Q79
Fine-tuning, training en data
PaddlePaddle/Paddle
PArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (『飞桨』核心框架,深度学习&机器学习高性能单机、分布式训练和跨平台部署)
★ 24K⑂ 6KC++
Apache-2.0Q88
AI-programmeer- en ontwikkelaarstools
anthropics/claude-code-action
Integrates Claude Code with GitHub Actions for comment-driven assistance, pull-request reviews and prompted maintenance tasks across supported providers.
★ 8,9K⑂ 2,1KTypeScript
MITQ84