#13 · +1.5K · 1d ·GitHub
VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
Global
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input.
A high-throughput and memory-efficient inference and serving engine for LLMs
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.
OmniVoice is a state-of-the-art massively multilingual zero-shot text-to-speech (TTS) model supporting over 600 languages.
#237 · +184 · 1d ·GitHub
1-Flash-1017 (748B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or Windows. Engine, server, dashboard, OpenAI- and Anthropic-compatible API, quant tools.
Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
#307 · +137 · 1d ·GitHub
8B to 27B: a calibrated probability for every option in one pass, and a short thought only when unsure. cpp and MLX.
ggml / ops / maintainer PRs%20sort%3Aupdated-desc) / dev stats / lib llama API / llama-server REST API
#360 · +115 · 1d ·GitHub
log、/api/dh/status 附带 duix 容器状态与 stats 计数（submits / segmentsubmits / statuspolls，用于核对「任务到底提交了没有」）。
#444 · +91 · 1d ·GitHub
526 in M4 CPU. You can test our huggingface demo (half H200) to experience high-speed inference .
#457 · +88 · 1d ·GitHub
One day, frontier AI research used to be done by meat computers in between eating, sleeping, having other fun, and synchronizing once in a while using sound wave interconnect in the ritual of "group meeting". That era is long gone.
Clone any voice. Generate speech. Dictate into any app.
SGLang is a high-performance serving framework for large language models and multimodal models.
#514 · +74 · 1d ·GitHub
py starts the SGLang server automatically and sends concurrent requests for an image directory or PDF:
ASPIRE is a new type of continual learning: "training" is skill refinement instead of gradient descent.
DeepGEMM is a unified, high-performance tensor core kernel library that brings together the key computation primitives of modern large language models — GEMMs (FP8, FP4, BF16), fused MoE with overlapped communication (Mega MoE), MQA scoring for the lightning indexer…
#612 · +61 · 1d ·GitHub
Voice-to-text dictation app with local (Nvidia Parakeet/Whisper) and cloud models (BYOK). Privacy-first and available cross-platform.
Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own
Fine-tune an 8B model on a 4 GB laptop GPU. Layer streaming keeps the frozen base out of VRAM and feeds it to the GPU one decoder layer at a time. 32 GB.
PyTorch is a Python package that provides two high-level features: - Tensor computation (like NumPy) with strong GPU acceleration - Deep neural networks built on a tape-based autograd system
#710 · +51 · 1d ·GitHub
Give a face to your voice AI in 5 lines of code. Generate realtime avatar video connected to your voice AI locally
nanochat is the simplest experimental harness for training LLMs. It is designed to run on a single GPU node, the code is minimal/hackable, and it covers all major LLM stages including tokenization, pretraining, finetuning, evaluation, and inference.
VoxCPM is a tokenizer-free Text-to-Speech system that directly generates continuous speech representations via an end-to-end diffusion autoregressive architecture, bypassing discrete tokenization to achieve highly natural and expressive synthesis.
2-Exp models. This repository contains the following implementations:
pth） - 支持跨不同 GPU 数量恢复（自动调整 step） - 支持 wandb 训练记录连续性（自动恢复同一个 run）
#752 · +47 · 1d ·GitHub
Ultralytics creates cutting-edge, state-of-the-art (SOTA) YOLO models built on years of foundational research in computer vision and AI. Constantly updated for performance and flexibility, our models are fast, accurate, and easy to use.
WanGP by DeepBeepMeep : The best Open Source Generative Models Accessible to the GPU Poor
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels