Skip to content
Computer vision

PyMuPDF Chart

extract tables from charts

2 min read

A vision-language model based on PaddleOCR-VL that takes a chart image and converts its data into a text table. The tool is designed to work alongside the PyMuPDF library for automated PDF document processing. The model requires an NVIDIA GPU with CUDA support and about 2 GB of VRAM for inference. It is a useful tool for developers and analysts who need to automate the digitization of charts from reports and documents.

Key features

  • Accurate data recognition from charts and output in a delimited row format
  • Construction of table rows with cells separated by a vertical bar
  • Based on the PaddleOCR-VL architecture adapted for the Hugging Face Transformers format
  • Apache License 2.0 for free use

What you can do

  • Integrate the model into a PDF processing pipeline using PyMuPDF
  • Automatically convert charts from financial and analytical reports into Markdown table format
  • Extract number series and headers from chart images

Example

from PIL import Image
import torch
from transformers import AutoModelForImageTextToText, AutoProcessor

repo_id = "artifex-software/pymupdf-chart"
processor = AutoProcessor.from_pretrained(repo_id)
model = AutoModelForImageTextToText.from_pretrained(
    repo_id, dtype=torch.bfloat16
).to("cuda").eval()

image = Image.open("chart.png").convert("RGB")
messages = [{"role": "user", "content": [
    {"type": "image", "image": image},
    {"type": "text", "text": "Chart Recognition:"},
]}]
inputs = processor.apply_chat_template(
    messages, add_generation_prompt=True, tokenize=True,
    return_dict=True, return_tensors="pt",
).to(model.device)
with torch.inference_mode():
    output = model.generate(**inputs, max_new_tokens=2048, do_sample=False)
print(processor.decode(
    output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True
).strip())
Read moreDetails
Hardware requirements: CUDA GPU, about 2 GB VRAM (CPU not supported).
Dependencies: Python 3.12, PyTorch, torchvision, Transformers 5.13 or newer.
Limitations: labels, values, and units may contain errors; long tables or complex multi-panel charts may be truncated at the 2048 token limit.
Discuss Telegram 0 4

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 1.7 GB · ready files2.4 GB
  • FP8 / INT8recommendedweights 0.8 GB1.4 GB
  • 4-bitweights 0.5 GB1.0 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · BF16 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # artifex-software/pymupdf-chart
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download artifex-software/pymupdf-chart --local-dir ~/models/pymupdf-chart
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to pymupdf-chart
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of pymupdf-chart (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

artifex-software/pymupdf-chart
Author
artifex-software
Category
Computer vision
Released
Posted

On NeyroLearn

Hugging Face

40

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments