Fully Open Framework for Democratized Multimodal Training

Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Official implementation for "TRIO: Token Reduction via Inference-Objective Guidance for Efficient Vision-Language Models" https://arxiv.org/pdf/2602.04657

omo/lazycodex: The coding agent for tokenmaxxers;the one and only agent harness for complex codebases. For your Codex, for your OpenCode

A simple yet powerful agent framework that delivers with open-source models

ERGO (Efficient Reasoning & Guided Observation) is a large vision-language model trained with reinforcement learning on efficiency objectives. [ICLR'26]

[ICLR'26] Traceable Evidence Enhanced Visual Grounded Reasoning: Evaluation and Methodology

Nano vLLM

[NeurIPS 2025] Official code for paper: Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs.

Official code for NeurIPS 2025 paper "GRIT: Teaching MLLMs to Think with Images"

Open-source unified multimodal model

This repository provides valuable reference for researchers in the field of multimodality, please start your exploratory travel in RL-based Reasoning MLLMs!

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

Solve Visual Understanding with Reinforced VLMs

Official repository of 'Visual-RFT: Visual Reinforcement Fine-Tuning' & 'Visual-ARFT: Visual Agentic Reinforcement Fine-Tuning'’

Witness the aha moment of VLM with less than $3.

A high-throughput and memory-efficient inference and serving engine for LLMs

A paper list of some recent works about Token Compress for Vit and VLM

A Compressed Stable Diffusion for Efficient Text-to-Image Generation [ECCV'24]

A 28× Compressed Wav2Lip for Efficient Talking Face Generation [ICCV'23 Demo] [MLSys'23 Workshop] [NVIDIA GTC'23]

Compressed LLMs for Efficient Text Generation [ICLR'24 Workshop]

The official NetsPresso Python package.

A library for training, compressing and deploying computer vision models (including ViT) with edge devices

Repository for 2023 AI City Challenge (Track1: Multi-Camera People Tracking)

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Polynomial Learning Rate Decay Scheduler for PyTorch