Labsco
firecrawl logo

llava

✓ Official★ 11

by firecrawl · part of firecrawl/ai-research-skills

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language…

🔥🔥🔥FreeQuick setup
#scraping#git
🧩 One of 7 skills in the firecrawl/ai-research-skills package — works on its own, and pairs well with its siblings.

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language…

Inspect the full instructions your agent will receiveExpand

This is the exact playbook injected into your agent when the skill activates — shown here so you can audit it before installing. You don't need to read it to use the skill.

name: llava description: Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis. version: 1.0.0 author: Orchestra Research license: MIT tags: [LLaVA, Vision-Language, Multimodal, Visual Question Answering, Image Chat, CLIP, Vicuna, Conversational AI, Instruction Tuning, VQA] dependencies: [transformers, torch, pillow]

LLaVA - Large Language and Vision Assistant

Open-source vision-language model for conversational image understanding.

When to use LLaVA

Use when:

  • Building vision-language chatbots
  • Visual question answering (VQA)
  • Image description and captioning
  • Multi-turn image conversations
  • Visual instruction following
  • Document understanding with images

Metrics:

  • 23,000+ GitHub stars
  • GPT-4V level capabilities (targeted)
  • Apache 2.0 License
  • Multiple model sizes (7B-34B params)

Use alternatives instead:

  • GPT-4V: Highest quality, API-based
  • CLIP: Simple zero-shot classification
  • BLIP-2: Better for captioning only
  • Flamingo: Research, not open-source

Available models

ModelParametersVRAMQuality
LLaVA-v1.5-7B7B~14 GBGood
LLaVA-v1.5-13B13B~28 GBBetter
LLaVA-v1.6-34B34B~70 GBBest
Copy & paste — that's it
# Load different models
model_7b = "liuhaotian/llava-v1.5-7b"
model_13b = "liuhaotian/llava-v1.5-13b"
model_34b = "liuhaotian/llava-v1.6-34b"

# 4-bit quantization for lower VRAM
load_4bit = True  # Reduces VRAM by ~4×

Web UI (Gradio)

Copy & paste — that's it
# Launch Gradio interface
python -m llava.serve.gradio_web_server \
    --model-path liuhaotian/llava-v1.5-7b \
    --load-4bit  # Optional: reduce VRAM

# Access at http://localhost:7860

Multi-turn conversations

Copy & paste — that's it
# Initialize conversation
conv = conv_templates["llava_v1"].copy()

# Turn 1
conv.append_message(conv.roles[0], DEFAULT_IMAGE_TOKEN + "\nWhat is in this image?")
conv.append_message(conv.roles[1], None)
response1 = generate(conv, model, image)  # "A dog playing in a park"

# Turn 2
conv.messages[-1][1] = response1  # Add previous response
conv.append_message(conv.roles[0], "What breed is the dog?")
conv.append_message(conv.roles[1], None)
response2 = generate(conv, model, image)  # "Golden Retriever"

# Turn 3
conv.messages[-1][1] = response2
conv.append_message(conv.roles[0], "What time of day is it?")
conv.append_message(conv.roles[1], None)
response3 = generate(conv, model, image)

Common tasks

Image captioning

Copy & paste — that's it
question = "Describe this image in detail."
response = ask(model, image, question)

Visual question answering

Copy & paste — that's it
question = "How many people are in the image?"
response = ask(model, image, question)

Object detection (textual)

Copy & paste — that's it
question = "List all the objects you can see in this image."
response = ask(model, image, question)

Scene understanding

Copy & paste — that's it
question = "What is happening in this scene?"
response = ask(model, image, question)

Document understanding

Copy & paste — that's it
question = "What is the main topic of this document?"
response = ask(model, document_image, question)

Training custom model

Copy & paste — that's it
# Stage 1: Feature alignment (558K image-caption pairs)
bash scripts/v1_5/pretrain.sh

# Stage 2: Visual instruction tuning (150K instruction data)
bash scripts/v1_5/finetune.sh

Quantization (reduce VRAM)

Copy & paste — that's it
# 4-bit quantization
tokenizer, model, image_processor, context_len = load_pretrained_model(
    model_path="liuhaotian/llava-v1.5-13b",
    model_base=None,
    model_name=get_model_name_from_path("liuhaotian/llava-v1.5-13b"),
    load_4bit=True  # Reduces VRAM ~4×
)

# 8-bit quantization
load_8bit=True  # Reduces VRAM ~2×

Best practices

  1. Start with 7B model - Good quality, manageable VRAM
  2. Use 4-bit quantization - Reduces VRAM significantly
  3. GPU required - CPU inference extremely slow
  4. Clear prompts - Specific questions get better answers
  5. Multi-turn conversations - Maintain conversation context
  6. Temperature 0.2-0.7 - Balance creativity/consistency
  7. max_new_tokens 512-1024 - For detailed responses
  8. Batch processing - Process multiple images sequentially

Performance

ModelVRAM (FP16)VRAM (4-bit)Speed (tokens/s)
7B~14 GB~4 GB~20
13B~28 GB~8 GB~12
34B~70 GB~18 GB~5

On A100 GPU

Benchmarks

LLaVA achieves competitive scores on:

  • VQAv2: 78.5%
  • GQA: 62.0%
  • MM-Vet: 35.4%
  • MMBench: 64.3%

Integration with frameworks

LangChain

Copy & paste — that's it
from langchain.llms.base import LLM

class LLaVALLM(LLM):
    def _call(self, prompt, stop=None):
        # Custom LLaVA inference
        return response

llm = LLaVALLM()

Gradio App

Copy & paste — that's it
import gradio as gr

def chat(image, text, history):
    response = ask_llava(model, image, text)
    return response

demo = gr.ChatInterface(
    chat,
    additional_inputs=[gr.Image(type="pil")],
    title="LLaVA Chat"
)
demo.launch()

Resources