Open Source
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace. Discover projects across categories like LLM, Vision, Audio, and more.
625 projects
Shanghai AI Laboratory (InternLM)
Shanghai AI Lab's Apache 2.0 LLM toolkit centered on the C++ TurboMind engine. v0.13.0 adds TurboQuant KV cache, Qwen3.5 MoE on Blackwell, Anthropic-compatible endpoints, and prefill-starvation fixes. Supports NVIDIA, Ascend, ROCm, Cambricon, and Apple Maca.
JD.com
JD.com's Apache 2.0 inference engine for LLMs, VLMs, DiT, and recommendation models, optimized for Huawei Ascend, Cambricon MLU, MThreads MUSA, and Nvidia CUDA. Production-validated across JD Retail's customer service and recommendation workloads.
antirez
Salvatore Sanfilippo's MIT-licensed local inference engine for DeepSeek V4 Flash, in pure C with Metal, CUDA, and ROCm backends. Asymmetric 2-bit quantization fits the model in 96 GB. 11.7k stars in under three weeks.
Presenton
Open-source self-hostable AI presentation generator under Apache 2.0. Brings-your-own-key support for OpenAI, Gemini, Claude, Bedrock, Ollama. Editable PPTX/PDF output and Electron desktop apps. 6.4k stars.
Lum1104
MIT-licensed plugin that converts any codebase into an interactive knowledge graph with Tree-sitter parsing plus LLM semantic analysis. Works with Claude Code, Cursor, Codex, OpenCode, and Gemini CLI. 21k+ stars.
NVIDIA Labs
NVIDIA Labs open-source infrastructure for long autoregressive video generation. Apache 2.0, NVFP4 W4A4 quantization, balanced sequence parallelism, multi-shot training, 1.3B-5B models reaching 45.7 FPS quantized.
Nari Labs
1.6B-parameter open dialogue TTS from Nari Labs generating two-speaker English conversation with non-verbal cues (laughs, sighs) in a single pass. Apache 2.0, runs on a single RTX 4090 in 4.4 GB.
k2-fsa
Apache-2.0 zero-shot multilingual TTS with the broadest language coverage in open source — 600+ languages, voice cloning, voice design, and 40x real-time inference on NVIDIA, Apple Silicon, and Intel Arc.
Microsoft
Microsoft's MIT-licensed open frontier voice AI: 1.5B long-form TTS up to 90 minutes with 4 speakers, 0.5B streaming TTS at 300 ms latency, and 7B ASR for 60-minute single-pass transcription. 47k+ stars.
Screenpipe (YC S26)
Y Combinator S26 open-source local AI memory daemon that continuously transcribes audio via local Whisper, OCRs the screen, and indexes everything into a searchable SQLite store. MIT, 18k+ stars.
TypeWhisper
Open-source on-device speech-to-text for macOS supporting WhisperKit, Parakeet TDT v3, Apple SpeechAnalyzer, and local Gemma 4 post-processing. System-wide dictation with per-app workflows. GPL-3.0.
Zackriya Solutions
Privacy-first AI meeting assistant in Rust with 4x faster Parakeet/Whisper live transcription, Sortformer speaker diarization, and Ollama-powered local summaries. MIT, 12k+ stars.