Stars
Radial Attention Official Implementation
[ICML 2025] SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity
The official implementation of "Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers" (arXiv 2025)
RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning
This is the official Python version of Angles Don’t Lie: Unlocking Training-Efficient RL Through the Model’s Own Signals.
Lightweight coding agent that runs in your terminal
Official Repository: A Comprehensive Benchmark for Logical Reasoning in MLLMs
This is Official PyTorch implementation for 2025-ICML-CoreMatching: Co-adaptive Sparse Inference Framework for Comprehensive Acceleration of Vision Language Model
JaxPP is a library for JAX that enables flexible MPMD pipeline parallelism for large-scale LLM training
Scalable toolkit for efficient model reinforcement
When Agent Becomes the Scientist – Building Closed-Loop System from Hypothesis to Verification
Official Repository of OmniCaptioner
Official code implementation for 2025 ICLR accepted paper "Dobi-SVD : Differentiable SVD for LLM Compression and Some New Perspectives"
(ACL-2025 main conference) SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing
DeepEP: an efficient expert-parallel communication library
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
FlashMLA: Efficient MLA decoding kernels
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train Qwen3, Llama 4, DeepSeek-R1, Gemma 3, TTS 2x faster with 70% less VRAM.
[NeurIPS 2024] Simple and Effective Masked Diffusion Language Model
Ramulator 2.0 is a modern, modular, extensible, and fast cycle-accurate DRAM simulator. It provides support for agile implementation and evaluation of new memory system designs (e.g., new DRAM stan…
(ACL-2025 main conference) Dolphin: Moving Towards Closed-loop Auto-research through Thinking, Practice, and Feedback
A WebUI for Side-by-Side Comparison of Media (Images/Videos) Across Multiple Folders
A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.
VILA is a family of state-of-the-art vision language models (VLMs) for diverse multimodal AI tasks across the edge, data center, and cloud.
A generative world for general-purpose robotics & embodied AI learning.
[ICML 2024 Best Paper] Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution (https://arxiv.org/abs/2310.16834)