[CVPR 2026] Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking

Official implementation of paper "Vocoder is not all you need".

Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

Interactive World Model papers organized by core research challenges.

Dataflow-Oriented Reinforcement Learning for (Multi-)Agentic LLMs

Fine-tune Gemma 4 and 3n with audio, images and text on Apple Silicon, using PyTorch and Metal Performance Shaders.

End-to-end speech recognition large model: 31 languages, dialects, accents, lyrics, hotwords, timestamps, speaker diarization. Trained on tens of millions of hours.

The most accurate natural language detection library for Python, suitable for short text and mixed-language text

MOSS-TTS-Nano is an open-source multilingual tiny speech generation model from MOSI.AI and the OpenMOSS team. With only 0.1B parameters, it is designed for realtime speech generation, can run direc…

🌋LavaSR: Fast Speech restoration and enhancement

Minimalist ML framework for Rust

A TurboQuant inference server

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

High-Quality Voice Cloning TTS for 600+ Languages

Expose Antigravity as OpenAI & Anthropic compatible API (base_url + key)

A plug-and-play compiler that delivers free-lunch optimizations for both inference and training.

Triton kernel fusion for Qwen3-TTS 1.7B inference acceleration — RMSNorm, SwiGLU, M-RoPE, Norm+Residual

Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.

AI agents running research on single-GPU nanochat training automatically

Plug-and-play streaming semantic VAD for real-time full-duplex spoken dialogue systems.

Sparse Transition Matrix-Accelerated Trie Index for Constrained Decoding (https://arxiv.org/abs/2602.22647)

Open source voice dictation technology

LiteRT-LM is Google's production-ready, high-performance, open-source inference framework for deploying Large Language Models on edge devices.

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

Secure, Fast, and Extensible Sandbox runtime for AI agents.

onnxruntime-extensions: A specialized pre- and post- processing library for ONNX Runtime

A framework for efficient model inference with omni-modality models