Lists (16)
Stars
slime is a LLM post-training framework aiming at scaling RL.
Muon: An optimizer for hidden layers in neural networks
The official repository of Dynamic-SUPERB.
Turn detection for full-duplex dialogue communication
TEN VAD: low-latency high-performance Voice Activity Detector
Production-ready platform for agentic workflow development.
JamendoMaxCaps is a large-scale dataset of 362,000 instrumental creative commons tracks
A song aesthetic evaluation toolkit trained on SongEval.
Benchmark data and code for MMAR: A Challenging Benchmark for Deep Reasoning in Speech, Audio, Music, and Their Mix
Dynamic Memory Management for Serving LLMs without PagedAttention
MiMo: Unlocking the Reasoning Potential of Language Model – From Pretraining to Posttraining
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
💩State-of-the-art shitcode principles your project should follow to call it a proper shitcode
Easily create large video dataset from video urls
Kimi-Audio, an open-source audio foundation model excelling in audio understanding, generation, and conversation
A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training
Distributed Compiler Based on Triton for Parallel Systems
VeOmni: Scaling any Modality Model Training to any Accelerators with PyTorch native Training Framework
ByteCheckpoint: An Unified Checkpointing Library for LFMs
Understanding R1-Zero-Like Training: A Critical Perspective
🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.