llava

Answer visual questions and chat about images using vision-language models.

3|Updated Feb 21, 2026

Search Tags:#multimodal #conversational AI #instruction tuning #vision-language #VQA #image chat

Ready to run

npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill llava-ihatesea69

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: llava
Source: https://github.com/ihatesea69/HieuNghi-AI-Skills/tree/main/airesearch_skills/18-multimodal/llava
Command: npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill llava-ihatesea69

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires transformers, torch, pillow, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables AI to understand and interact with visual information, bridging the gap between text and images for richer AI applications.

Core Features & Use Cases

Visual Question Answering (VQA): Ask questions about the content of an image.
Conversational Image Chat: Engage in multi-turn dialogues about an image.
Instruction Following: Guide the AI to perform tasks based on visual context.
Use Case: Upload a photo of a room and ask "What color is the sofa?" or "Suggest a rug that would match the existing furniture."

Quick Start

Use the llava skill to describe the attached image 'photo.jpg'.

Frequently Asked Questions about llava

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I do visual question answering with an image using a language model?▼

Visual question answering integrates vision encoders with large language models to analyze image content. You can ask questions about an uploaded photo, such as identifying object colors or suggesting matching items based on visual context.

Can I have a multi-turn image chat to analyze a photo?▼

Yes, multi-turn image chat is supported for conversational image understanding. This allows you to engage in continuous dialogues about an image, asking follow-up questions to perform detailed image analysis and visual instruction following.

Do I need torch and transformers to run multimodal conversational AI?▼

Yes, you need torch, transformers, and pillow libraries to run multimodal conversational AI. These dependencies provide the core functionality for integrating vision encoders with large language models for visual instruction following.

What is the best way to follow visual instructions for detailed image analysis?▼

The best way to follow visual instructions is using a vision-language model that bridges text and images. This approach enables detailed image analysis by guiding the AI to perform specific tasks based on the provided visual context.

Does conversational image understanding work with standard photo formats?▼

Conversational image understanding works with standard photo formats processable by the pillow library. The AI can analyze these images to answer questions, suggest matches for existing furniture, or describe visual content.

When should I not use a vision-language model for image chat?▼

You should not use a vision-language model for image chat if your environment lacks torch and transformers support. The core functionality requires these specific dependencies to integrate vision encoders with large language models effectively.