llava

Integrate LLaVA-compatible vision-language models for image chats and VQA.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/nadicodeai/argo-agent --skill llava-nadicodeai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: llava
Source: https://github.com/nadicodeai/argo-agent/tree/main/optional-skills/mlops/llava
Command: npx skills add https://github.com/nadicodeai/argo-agent --skill llava-nadicodeai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Vision-language multimodal AI integration enabling conversational image understanding and VQA tasks across apps.

Core Features & Use Cases

  • Vision-language multimodal chat with image inputs and text outputs.
  • Visual Question Answering on images and documents.
  • Image instruction following and multi-turn conversations for complex tasks.

Quick Start

Load a pretrained LLaVA-compatible model and start a multimodal image chat by providing an image.

Frequently Asked Questions about llava

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

FAQPage Schema
How do I add visual question answering to my image chat application?

Visual question answering is enabled by loading a pretrained LLaVA-compatible model with a CLIP vision encoder, allowing your application to process image inputs and generate text responses for multi-turn image chats.

What is needed to start a multimodal image chat with a vision-language model?

To start a multimodal image chat, you need a pretrained LLaVA-compatible model and a CLIP vision encoder to process image inputs and generate text outputs for visual instruction following.

Can I use this for multi-turn image conversations and instruction following?

Yes, the vision-language integration supports multi-turn image chats and image-guided instruction following, enabling complex conversational tasks across experiments and products.

Does this vision-language integration work with existing pretrained models?

It requires pretrained LLaVA-compatible models paired with a CLIP vision encoder, supporting image understanding and visual question answering tasks via provided SKILL.md frontmatter dependencies and references.

What are the limitations of using a CLIP vision encoder for image understanding?

Image understanding is constrained by the CLIP vision encoder and LLaVA-compatible model dependencies, requiring proper frontmatter setup to execute visual question answering and instruction-following tasks effectively.