Lists (14)
Starred repositories
This repository contains the official implementation of "FastVLM: Efficient Vision Encoding for Vision Language Models" - CVPR 2025
Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning, achieving state-of-the-art performance on 38 out of 60 public benchmarks.
Thera: Aliasing-Free Arbitrary-Scale Super-Resolution with Neural Heat Fields
depyf is a tool to help you understand and adapt to PyTorch compiler torch.compile.
A fast structure from motion pipeline written in Pytorch.
A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.
The simplest, fastest repository for training/finetuning small-sized VLMs.
An animation engine implemented in rust, inspired by manim and JAnim
Constraint-based geometry sketcher for blender
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
Fast implementation of BERT inference directly on NVIDIA (CUDA, CUBLAS) and Intel MKL
Embedded Rust device crates for WCH's RISC-V and Cortex-M microcontrollers
[CVPR 2025 Oral] Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models
CUDA Python: Performance meets Productivity
Command and Conquer: Generals - Zero Hour
This package contains the original 2012 AlexNet code.
🚀 Efficient implementations of state-of-the-art linear attention models in Torch and Triton
3D Gaussian Splatting (3DGS) on fisheye cameras
Distributed Triton for Parallel Systems
A GUI Agent application based on UI-TARS(Vision-Language Model) that allows you to control your computer using natural language.
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.