Pure-MLX speech synthesis, voice cloning, dialogue, sound-effects, and ASR for Apple Silicon: Fish S2 Pro, VibeVoice, LongCat, MOSS, Step-Audio, Cohere ASR.
-
Updated
Sep 3, 2026 - Python
Pure-MLX speech synthesis, voice cloning, dialogue, sound-effects, and ASR for Apple Silicon: Fish S2 Pro, VibeVoice, LongCat, MOSS, Step-Audio, Cohere ASR.
LTX-2.3 22B text-to-video and image-to-video generation with synchronized audio on Apple Silicon — 8-bit/4-bit MLX inference plus on-device LoRA fine-tuning.
MLX-native computer vision for Apple Silicon: open-vocabulary grounding, object detection, depth and camera geometry, segmentation, and video object tracking.
Pure MLX inference for MiniMax Music 3 on Apple silicon.
Adds on-device Apple MLX inference to any app already using AIChatKit. Models are downloaded from Hugging Face Hub on first use and cached locally. Runs on Metal GPU and Apple Neural Engine — no network calls during inference.
Run LLMs on Apple Silicon with MLX — solo on one Mac or clustered over Thunderbolt. Terminal chat client, model-cache manager, and a built-in coding agent.
Enhancing User Privacy by Local Deployment of LLMs, Final Project of EAI 2024 Fall
Turn long-form speculative fiction into explorable 3D planets, scientific world models, and locally generated alien artwork.
Open-source, local-first AI meeting assistant (macOS HUD). Real-time Silero VAD + Apple-MLX Whisper on-device transcription, fastembed embeddings + LanceDB vector memory, RAG context cues in a transparent always-on-top window. Privacy-first and memory-augmented: no cloud, no API keys.
Apple MLX + Qwen3 local AI voice mood diary for iPhone
Non-Euclidean Latent Space Code Reasoning via Joint Embedding Predictive Architectures on AST Graphs
go-mlx native inference, training and research tooling.
Run and benchmark local LLMs on Apple Silicon with MLX. Compare latency and throughput, reuse prompt prefixes, and verify every speed change.
Graph-based Protein-Peptide Interaction Predictor. Reranking candidates using ESM-2 embeddings and Graph Neural Networks (GNN) on PROPEDIA data.
A native macOS AI assistant that runs large language models fully on-device on Apple Silicon via Apple MLX (mlx-swift-lm). No server, no account, no cloud — your conversations and files never leave your Mac.
Apple MLX-only synthetic market-data generator: vectorized GBM + GARCH/regime roadmap, reproducibility contract, stylized-facts validation, themed OHLCV inspection. No NumPy.
To associate your repository with the apple-mlx topic, visit your repo's landing page and select "manage topics."