The fastest repo in history to surpass 100K stars ⭐. Better Harness Tools that make real things done. Built in Rust using oh-my-codex.

Rust crate with Python bindings for the HAT splitting rule

[NeurIPS 2025] MergeBench: A Benchmark for Merging Domain-Specialized LLMs

Official implementation of "OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging".

Source codes for the paper "Expert Merging: Model Merging with Unsupervised Expert Alignment and Importance-Guided Layer Chunking"

Paper2Demo: Turn any academic PDF into a stunning demo video

LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs

🦞+🔬: NanoResearch: The Autonomous AI Research Assistant

Official Codebase for "Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights"

FusionBench: A Comprehensive Benchmark/Toolkit of Deep Model Fusion

My learning notes for ML SYS.

Official Repo for WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching

HY-WU (Part I): An Extensible Functional Neural Memory Framework and An Instantiation in Text-Guided Image Editing

Webpage: https://chuny9743.github.io/AI4WaterEnv_Webpage/

SparAlloc: A Simple and Modular Framework for Decoupled Sparsity Allocation in Layerwise Pruning for LLM

[CPAL 2026 oral] Offical implementation of "ROSE: Reordered SparseGPT for More Accurate One-Shot Large Language Models Pruning”

[arXiv 2026] dVoting: Fast Voting for dLLMs

Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities. ACM Computing Surveys, 2026.

[arXiv 2024] Is Oracle Pruning the True Oracle?

[CVPR 2026] ReasonMap: Towards Fine-Grained Visual Reasoning from Transit Maps

[TMLR 2025] Efficient Reasoning Models: A Survey

[ICLR 2026] RewardMap: Tackling Sparse Rewards in Fine-grained Visual Reasoning via Multi-Stage Reinforcement Learning

[ICLR 2026] MergeMix: A Unified Augmentation Paradigm for Visual and Multi-Modal Understanding

Tools for merging pretrained large language models.

An official repository for GPTailor

CUDA Templates and Python DSLs for High-Performance Linear Algebra