Global

VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.

AirLLM dramatically reduces inference memory usage, letting 70B large language models run on a single 4GB GPU card — without quantization, distillation, or pruning. 8-Flash-Next (125B) on 6GB, and DeepSeek-V3 (671B) on 12GB. 8-Flash-Next (125B) under 6GB.

Clone any voice. Generate speech. Dictate into any app.

8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. cpp and MLX.

Local UI to run and train LLMs and diffusion models. 8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.

Find out which open-source Large Language Models (LLMs) your hardware can comfortably run. llmfit inspects your CPU, system RAM, GPU(s), VRAM, and accelerator configuration to recommend models across popular quantizations.

Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own

AssetFurnace: the most complete open-source image-to-3D pipeline (Oct 2026). Prompt or picture → textured, rigged, animated game character, locally on Apple Silicon or NVIDIA. 2, Hunyuan3D, SF3D, auto-rig and 21 moves.

exo connects all your devices into an AI cluster. Not only does exo enable running models larger than would fit on a single device, but with day-0 support for RDMA over Thunderbolt, makes models run faster as you add more devices.

LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar

MLX is an array framework for machine learning on Apple silicon, brought to you by Apple machine learning research.

0) OpenAI- and Anthropic-compatible LLM inference server for Apple Silicon, built on MLX, focused on reliable tool calling for coding agents. Up to 4× faster than Apple's MLX (mlx-lm) on the same weights.

Native LLM inference server for Apple Silicon. OpenAI + Anthropic API compatible. No Python.

app — the end-signal probe, public pages, and the live steered-model lab.

Run Bonsai 2 27B locally on Mac (Metal), Linux/Windows (CUDA, Vulkan, ROCm), or CPU.

An open-source model family for long-form speech, dialogue synthesis, voice design, sound effects, and real-time streaming TTS

Strands decider is one of a new class of decision models, or "system one" models. Unlike an LLM, which can generate arbitrary text, a decision model picks between sets of options and rates things on a scale.

> Why not just run a simple Whisper model on every audio batch? Whisper is designed for complete utterances, not real-time chunks. Processing small segments loses context, cuts off words mid-syllable, and produces poor transcription.

The GIF is an original-speed render of a real local Snake run. Every move calls Laya; the visible cycle safety layer can correct unsafe proposals. The latency figures above are the separate one-question API benchmark, not the frame time of the three-question Snake loop.

> Independent research project. SemIf was formerly called OpenJev. It is not affiliated with or endorsed by TypeSafe.

Wispr Flow for your lips. Hold a key, silently mouth what you want to say, let go, and the text shows up at your cursor in whatever app you're in. No microphone and no sound, just your webcam.

Eight SDKs, one C++ core. A capability registry routes every call to the best engine on the device, and the Console manages your fleet from above.

DFlash is a lightweight block diffusion model designed for speculative decoding. It enables efficient and high-quality parallel drafting.

Nimble takes some text and a schema, and makes typed decisions about the text. The schema is the list of questions to answer. Each question is either a choice from a list that you give or a true or false question.

> 🚀 TurboQuant KV cache compression is now in vLLM (PR #38479, merged April 2026): --kv-cache-dtype turboquantk8v4 and friends, with fused Triton store/decode kernels. The PR discussion drew on the asymmetric K/V findings from this repo.

Local-first healthcare AI: clinical NER and HIPAA PII de-identification on hardware you control. 2,200+ medical models, 35 model-backed PII languages, and Python, MLX, Android and browser runtimes. 0 SDK.

8 is the large language model series developed by Qwen team, Alibaba Group.

What it is, and what it isn't. These are very small models. They make extremely fast, well-calibrated judgement calls between options, and they slot easily into your local code.

Run a 105 GB AI model on a Mac that can't hold it. 8-Flash-Next (125B mixture of experts) from your SSD and caches the busiest experts in memory, so it runs on Macs with 16 to 64 GB. One native Swift binary on MLX and Metal, no Python, offline.

Companies like OpenAI built "Super AI" that threatens human independence. We crave individuality: AI that amplifies, not erases, YOU.