All repos — GH Trending Explorer

We propose Prism, a dynamic sparse attention framework for natively training joint video-audio generation models at 2K.

Adaptive Block Sparse Attention for MiniMax-H3

[CVPR 2026] Towards Real-Time Diffusion-Based Streaming Video Super-Resolution — An efficient one-step diffusion framework for streaming VSR with locality-constrained sparse attention and a tiny conditional decoder.

dynamicslab / pysindy

A package for the sparse identification of nonlinear dynamical systems from data

decoderesearch / SAELens

Training Sparse Autoencoders on Language Models

vdev-ctrl / Adaptive-Learnable-Hierarchical-Routing-

This repo is a alternative solution for standard self-attention. It uses tree routing with learnable functions for cheaper self-attention while not compromising accuracy.

intel / neural-compressor

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

castorini / pyserini

Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

microsoft / MInference

[NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention, which reduces inference latency by up to 10x for pre-filling on an A100 while maintaining accuracy.

Evianis / travel-photo-abstraction

A source-available Codex skill for distilling photographs into sparse editorial abstractions.

HKUSTDial / flash-sparse-attention

Trainable fast and memory-efficient sparse attention

pydata / sparse

Sparse multi-dimensional arrays for the PyData ecosystem

facebookresearch / mvdust3r

Open source impl of **MV-DUSt3R+ Single-Stage Scene Reconstruction from Sparse Views In 2 Seconds** from Meta Reality Labs. Project page https://mv-dust3rp.github.io/

thu-ml / SLA

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

veda-sparse / Veda-on-ComfyUI

Veda Sparse Attention for ComfyUI

Saganaki22 / ComfyUI-sol-attn

NVIDIA Sol-Attn for ComfyUI / Triton kernel on SM89 - SM121, with zero-copy MiniMax H3 nodes: memory-efficient attention, scheduled tau with graph preview, and feed-forward chunking. Measured 1.14–1.44× vs SageAttention and −37% MLP peak VRAM on H3

sepiablue-ai / ComfyUI-MiniMax-H3-W4A4-VSA

MiniMax H3 Ref2VA acceleration for ComfyUI on an RTX 4070 12GB: 4-bit FC1 weights + Streaming VSA sparse attention. 1024x1792 video with audio in about 3 min 17 s, about 3.4x faster than dense attention.

chenkangjie1123 / VGGT-Diff

Official implementation of the paper "VGGT-Diff: Visual Geometry Meets Diffusion for Sparse-View Novel View Synthesis."

InternLM / ResOPD

ResOPD: Tail Residualization for Sparse On-Policy Distillation. Core implementation built on verl.

Reederey87 / glm53-flash-exl3-2x-dgx-spark

GLM-5.3-Flash EXL3 (320B MoE) on 2x NVIDIA DGX Spark — production serving kit, 1M context, 97%+ multi-session prefix caching, DFlash2 spec decode

veda-sparse / Miowtion

Sparse acceleration and LoRA fine-tuning of MiniMax-H3 on resource-constrained machines.

HaoyiZhu / Piecewise-Taylor-Attention

Piecewise-Taylor Attention

dodododddo / mcsparse

MC-Sparse: Deconstructing and Closing the Dense–Sparse Attention Gap in Diffusion Transformers.

tardellirs / gemma3moe-builder

Sparse-upcycle a dense Gemma-3-270m into a Mixture-of-Experts (gemma3moe) and serve it on llama.cpp (CPU). Upcycle, train, inspect routing, GGUF.

sepiablue-ai / ComfyUI-MiniMax-H3-TaoMate-C3

TaoMate-H3 3-step acceleration package for ComfyUI on an RTX 4070 12GB: 4-bit (W4A4) FFN weights, 10% sparse attention and extra INT8 weight caching, with Windows setup and sample workflows.

YuuZhen / restrained-hand-drawn-illustration

11:00 Create playful, restrained hand-drawn illustrations from photos or prompts, featuring natural open contours, tactile variable-width lines, pale backgrounds, sparse muted color, JPG delivery, and optional GIF motion.

YilunKuang / lpworldmodel

Official Code for LpWM: A Case for Sparse Representations in World Models

extropic-ai / sparse-transformers

Sparse Transformers in JAX

fraunhoferhhi / RIPEpp

[LIMIT@ECCV 2026] RIPE, minus the negatives, plus a trainable matcher - full sparse matching pipelines learned from positive image pairs alone.

FemtoRhythm / pyramis-l3d

Pyramis-L3D: hierarchical sparse attention research prototype (code only; weights on Hugging Face)

LAMDA-Model-Reuse / SaveRouter

Sparse-supervision LLM routing that accounts for upfront feedback acquisition and deployment payback.

cupy / cupy

NumPy & SciPy for GPU