[ICLR 2026] Official implementation of JavisDiT and JavisDiT++ series.

A unified multimodal model toolkit

[CVPR 2026] Official codes of "Monet: Reasoning in Latent Visual Space Beyond Image and Language"

We propose an efficient flow-based multimodal generation model with bidirectional flows.

Open-source native multimodal pretraining — without catastrophic forgetting.

SenseNova-U series: Native Unified Paradigm with NEO-unify from the First Principles

[ICLR 2026] This is an early exploration to introduce Interleaving Reasoning to Text-to-image Generation field and achieve the SoTA benchmark performance. It also significantly improves the quality…

[CVPR 2026] Official repo for "EVATok: Adaptive Length Video Tokenization for Efficient Visual Autoregressive Generation"

LLaDA2.0-Uni: Understanding and Generation the World.

MonkeyOCRv2 Vision Encoder — A Document-Native Visual Backbone

Official implementation of DeltaV, a unified multimodal model for interleaved reasoning with visual state updates.

🔨AI 方向好用的科研工具

(NeurIPS 2025) Vision Foundation Models as Effective Visual Tokenizers for Autoregressive Image Generation

Reverse Chain-of-Thought Problem Generation for Geometric Reasoning in Large Multimodal Models

Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks

Spatial Aptitude Training for Multimodal Langauge Models

MSTAR: Box-free Multi-query Scene Text Retrieval with Attention Recycling

Multimodal OCR: Parse Anything from Documents

An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

Official electron build of draw.io

Open-source unified multimodal model

Towards Efficient Multimodal Large Language Models: A Survey on Token Compression

[ICLR26] ThinkOmni: Lifting Textual Reasoning to Omni-modal Scenarios via Guidance Decoding

[ICLR 2026] The official repository for paper "ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning"

OCR in the Era of Large Language Models

Deformable ConvNets V2 (DCNv2) in PyTorch

[ICCV 2023] Official implementation of the paper "DFA3D: 3D Deformable Attention For 2D-to-3D Feature Lifting"

Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence