Vlm

PixelRAG is a visual retrieval-augmented generation system that renders web pages and documents as pixel-based screenshots for search and retrieval,...

AngelSlim is a specialized model compression toolkit from Tencent for large language models, vision-language models, and audio models, offering...

PaperBanana is an open-source framework that automates the generation of publication-quality academic diagrams, plots, and research visualizations...

TokenSpeed is a specialized LLM inference engine optimized for agentic workloads, combining TensorRT-LLM-level performance with vLLM-level usability....

React Native ExecuTorch enables on-device AI model inference in React Native applications using Meta's ExecuTorch framework, supporting computer...

NVIDIA VSS is a reference architecture suite for building GPU-accelerated vision agents and video analytics applications that combine VLMs, LLMs, and...

EvalScope is a comprehensive evaluation framework for large language models, vision-language models, and other AI systems, offering built-in...

This is a specialized research and penetration-testing toolkit for reverse-engineering ChatGPT subscription payment flows (Plus/Team/Pro via Stripe,...

GenieX is an on-device inference runtime for running large language models and vision models locally on Qualcomm Snapdragon devices (NPU, GPU, CPU)....

SGLang is a high-performance inference serving framework optimized for large language models and multimodal models, with support for speculative...

UI-TARS-desktop is a desktop application providing GUI agents powered by the UI-TARS vision model, enabling local and remote computer automation as...

A curated collection of 100+ original architecture diagrams and visual maps explaining LLM, reinforcement learning, and vision-language model...

DeepCamera is an open-source AI camera platform that enables local video analysis using Vision Language Models (Qwen, DeepSeek, LLaVA) and object...

Awesome-Jailbreak-on-LLMs is a curated collection of research papers, code, datasets, and evaluations focused on jailbreak attacks and defenses for...

RunAnywhere is a production-ready SDK for running AI models (LLMs, vision, speech) entirely on-device across iOS, Android, Flutter, React Native, and...

LMMs-Eval is a unified evaluation toolkit for multimodal large language models across text, image, video, and audio tasks. It serves researchers and...

OSWorld is a NeurIPS 2024 benchmark for evaluating multimodal AI agents on open-ended real computer tasks, supporting multiple virtualization...

Awesome-LM-SSP is a curated reading list and resource collection focused on the trustworthiness of large language models, vision-language models, and...

StarVector is a multimodal vision-language model specialized for generating Scalable Vector Graphics (SVG) code from images and text inputs. It...

UltraRAG is a low-code framework for building retrieval-augmented generation (RAG) pipelines that integrates with model context protocol (MCP) and...

A collection of 61+ Jupyter notebooks providing tutorials on state-of-the-art computer vision models including YOLOv11, SAM 2, Florence-2, PaliGemma,...

Hugging Face Transformers is the de facto standard Python library for accessing, fine-tuning, and deploying state-of-the-art pretrained models across...

A curated repository of learning resources for anomaly detection and outlier detection, including academic papers, books, courses, toolboxes, and...

VLM-R1 is a specialized research framework for training vision-language models using reinforcement learning (R1-style approach and GRPO), built on...

ComfyUI LLM Party is a specialized node framework that integrates large language models, voice synthesis, OCR, and multimodal capabilities into...

Skywork-R1V is a series of advanced multimodal reasoning models that combine vision and language understanding with chain-of-thought capabilities,...

PromptEnhancer is a prompt-rewriting tool that transforms user prompts into clearer, structured versions to improve text-to-image and image-to-image...

ERNIE 4.5 is a family of large-scale multimodal models (text and vision) developed by Baidu, comprising 10 variants with up to 424B total parameters...