skills.rest
twjarviszyra-web
twjarviszyra-webCOMMUNITY
Explore all skills by this author →
GitHub

llava

Answer visual questions and generate captions from natural-language image prompts.

3|Updated Apr 4, 2026
Search Tags:#instruction tuning#multimodal ai#image understanding#visual question answering#llava#image chat
One-click install
npx skills add https://github.com/twjarviszyra-web/hermes-unbound --skill llava-twjarviszyra-web
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llava
Source: https://github.com/twjarviszyra-web/hermes-unbound/tree/main/optional-skills/mlops/llava
Command: npx skills add https://github.com/twjarviszyra-web/hermes-unbound --skill llava-twjarviszyra-web

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

LLaVA removes the friction of interpreting images by turning visual inputs into accurate, conversational answers, captions, and explanations.

Core Features & Use Cases

  • Visual Question Answering: Ask detailed questions about a single image and get context-aware responses.
  • Multi-turn Image Chat: Maintain an ongoing conversation about an image across several follow-up prompts.
  • Training and Fine-tuning: Support staged pretraining, instruction tuning, and LoRA-based customization for new datasets.
  • Use Case: A team can upload a product screenshot or document photo and quickly ask what is shown, what text appears, or what action should happen next.

Quick Start

Ask LLaVA to describe the attached image and answer one specific visual question in plain language.

Frequently Asked Questions about llava

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

FAQPage Schema
How do I perform visual question answering on an image using natural language?▼

Visual question answering is achieved by submitting an image alongside a natural-language prompt to receive context-aware responses. The system supports multi-turn image chat, allowing ongoing conversations about an image across several follow-up prompts.

Can I use LLaVA for multi-turn image chat and document analysis?▼

Yes, multi-turn image chat and document analysis are supported. You can upload a document photo or screenshot and ask what text appears, what is shown, or what action should happen next across multiple conversational follow-ups.

What frameworks do I need for visual instruction following and fine-tuning?▼

Visual instruction following and fine-tuning require PyTorch plus Transformers for inference. You also need a CLIP-based vision encoder and a Vicuna or LLaMA language model to process image and text inputs.

How does visual instruction tuning work for custom datasets?▼

Visual instruction tuning works by supporting staged pretraining and LoRA-based customization for new datasets. This process aligns the vision encoder and language model to follow specific natural-language instructions for image understanding tasks.

Related Skills

Explore More Education & Research Skills→
nelohenriqnelohenriq
1

llava

Merge a vision encoder with a language model for image-based conversation.

Community
Advanced
handsomelong922handsomelong922

implementing-llms-litgpt

Implement LitGPT workflows for LLM fine-tuning, pretraining, and deployment.

Community
Advanced
Wesley1600Wesley1600

vision

Extract text, classify objects, and analyze charts from images.

Community
Advanced
manashmandalmanashmandal

imggen

Generate images from text prompts using OpenAI image APIs and extract OCR text from images.

Community
Basic