Discover
Every project in the index. Underrated first: good projects that haven’t been noticed yet.
- Hitoshi Togasaki740
Karukan
Japanese input method with neural kana-kanji conversion
1.3k
- 2.7k
TokenSpeed
LLM serving engine tuned for agentic workloads on datacenter GPUs
- Antoine2.3k
forge
Reliability layer that makes local LLM tool calls succeed more often
- 3.4k
SIE
Self-hosted inference server running many open models behind one API
AI models and runtimesAgent frameworks and toolsServers and infrastructure
3.4k
- 3.5k
LiteRT
On-device runtime for running ML and generative AI models on edge hardware
- 5.6k
Weave Router
Prompt-level model router with an OpenAI-compatible endpoint
6.2k
6.7k
7.2k
- 7.3k
vllm-omni
Serving engine for multimodal, diffusion, speech, and action models
7.4k
- 7.6k
ODS
Installer that sets up a complete private local AI stack on your own machine
- andy6.8k
whichllm
Command-line tool that ranks local LLMs that will run on your hardware
9k
- 10.3k
DeepEP
GPU communication kernels for expert-parallel MoE training and inference
- 12.1k
LMCache
KV cache layer that stores and reuses LLM attention state across engines
- 13.1k
FlashMLA
Optimized attention kernels for DeepSeek model inference on GPUs and NPUs
- 13.5k
Speech To Speech
Local voice agent pipeline chaining VAD, speech recognition, LLM and TTS
- 5.4k
NVIDIA Model Optimizer
Python library for quantizing, pruning, and distilling models for inference
- 19.6k
KTransformers
CPU-GPU hybrid inference and fine-tuning for large MoE language models
- 24.1k
Paddle
Deep learning framework for training and deploying neural networks at scale
- 37.3k
MediaPipe
On-device ML framework and ready-made tasks for vision, audio, and text