Mindket

Heretic Grimoire

Newversion 1 · updated

Overview

Keeps a searchable archive of the reproduction records that Heretic-generated models publish, with CSV download. For researchers who track or rebuild such models.

Heretic Grimoire is a kit in the Research category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page, a zip of its files. Nothing in it runs by itself.

At a glance

You need
  • A web browser, which is all the online page needs
  • Python 3.13 and the packages in requirements.txt, for a local copy
  • For hosting on Hugging Face, the Gradio SDK, and attached persistent storage with read and write permission so the archive survives restarts
Runs in
  • A web browser, on the Space's page on Hugging Face
  • A local copy of the files, run on Python 3.13 under Windows, macOS or Linux
  • Your own Hugging Face Space built from the same files
Start with
pip install -U huggingface_hub
hf download heretic-org/Heretic-Grimoire --repo-type space --local-dir Heretic-Grimoire
cd Heretic-Grimoire

Platform

WindowsOperating system
macOSOperating system
LinuxOperating system

About

What it is

Heretic Grimoire is a Gradio app that collects, stores and lets you browse the reproduce.json files of public models made with Heretic. It runs the Heretic collector twice a day on a fixed schedule and indexes new records it finds on Hugging Face. The saved records stay available even if a model's source repository is later deleted.

Who it suits

It suits researchers and model users who track Heretic-generated models, and anyone who wants a permanent copy of the records needed to rebuild one.

IndustryWho uses itHow they use itExample
AI researchResearchersCompare reproducibility data across many modelsFilter the table and export it as CSV
Model hostingCommunity maintainersKeep a record of models that could disappearCheck that an archived record still exists
SoftwareDevelopersRun their own archive with the collectorStart app.py on their own machine

If you do not work with Heretic models, you can skip it.

Where to use it

  • A web browser, on the Space's page on Hugging Face
  • A local copy of the files, run on Python 3.13 under Windows, macOS or Linux
  • Your own Hugging Face Space built from the same files
Use caseWhoWhat they get back
Search the archiveResearchersA filtered table of reproducibility records
Export the dataAnalystsThe full archive as a CSV file
Host your own archiveDevelopersA local or hosted copy that updates twice a day

How to set it up

  • A web browser, which is all the online page needs
  • Python 3.13 and the packages in requirements.txt, for a local copy
  • For hosting on Hugging Face, the Gradio SDK, and attached persistent storage with read and write permission so the archive survives restarts
  1. Open the Space's page on Hugging Face to use it online. Nothing needs installing.
  2. For a local copy, install the Hugging Face hub package and download the Space's files.
  3. Move into the downloaded folder.
  4. Install Gradio 6.19.0, then the packages from requirements.txt.
  5. Start the app with python app.py, which creates your own local archive.
pip install -U huggingface_hub
hf download heretic-org/Heretic-Grimoire --repo-type space --local-dir Heretic-Grimoire
cd Heretic-Grimoire
pip install gradio==6.19.0
pip install -r requirements.txt
python app.py

How to use it

Open the page to see a searchable table with the model, its base model, creation date, Heretic version, KLD, refusal counts, trials, accelerator and a JSON file. Three actions are available.

ToolWhat it does
Search filtersNarrow the table to the records you want
CSV downloadSaves the full archive as a CSV file
RefreshReloads the visible table and the collector logs

Example

Filter the table down to one base model, then download the archive as CSV. You get the saved records in a spreadsheet-friendly file, even for models whose source repository is gone.

For newcomers and experts

Open the Space page and look at the table first. Each row is one saved reproduction record, and the filters help you find the model you care about.

You can read the collector's log in the page, though every fresh collection run empties it first. The archive keeps earlier records, so deleting a model's repository does not remove its saved entry. When you host your own copy, the persistent storage is what keeps the archive across restarts.

  1. The collector runs twice a day
  2. It finds new reproduction records on Hugging Face
  3. The records are added to the archive
  4. The table shows them for search and CSV export

Good to know

The app only archives public records and does not run or download models.

  • License: AGPL-3
  • Cost to run: none named, apart from persistent storage if you host it on Hugging Face
  • Data it reaches: public Heretic model records on Hugging Face, and the archive stored by the app

README and files

README

Heretic Grimoire

Keeps a searchable archive of the reproduction records that Heretic-generated models publish, with CSV download. For researchers who track or rebuild such models.

Made by heretic-org on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: Heretic Grimoire is a Gradio app that collects, stores and lets you browse the reproduce.json files of public models made with Heretic. It runs the Heretic collector twice a day on a fixed schedule and indexes new records it finds on Hugging Face. The saved records stay available even if a model's source repository is later deleted.

Who it suits: It suits researchers and model users who track Heretic-generated models, and anyone who wants a permanent copy of the records needed to rebuild one.

Industry | Who uses it | How they use it | Example
AI research | Researchers | Compare reproducibility data across many models | Filter the table and export it as CSV
Model hosting | Community maintainers | Keep a record of models that could disappear | Check that an archived record still exists
Software | Developers | Run their own archive with the collector | Start app.py on their own machine

If you do not work with Heretic models, you can skip it.

Where to use it:

  • A web browser, on the Space's page on Hugging Face
  • A local copy of the files, run on Python 3.13 under Windows, macOS or Linux
  • Your own Hugging Face Space built from the same files
Use case | Who | What they get back
Search the archive | Researchers | A filtered table of reproducibility records
Export the data | Analysts | The full archive as a CSV file
Host your own archive | Developers | A local or hosted copy that updates twice a day

How to set it up:

  • A web browser, which is all the online page needs
  • Python 3.13 and the packages in requirements.txt, for a local copy
  • For hosting on Hugging Face, the Gradio SDK, and attached persistent storage with read and write permission so the archive survives restarts
  1. Open the Space's page on Hugging Face to use it online. Nothing needs installing.
  2. For a local copy, install the Hugging Face hub package and download the Space's files.
  3. Move into the downloaded folder.
  4. Install Gradio 6.19.0, then the packages from requirements.txt.
  5. Start the app with python app.py, which creates your own local archive.
pip install -U huggingface_hub
hf download heretic-org/Heretic-Grimoire --repo-type space --local-dir Heretic-Grimoire
cd Heretic-Grimoire
pip install gradio==6.19.0
pip install -r requirements.txt
python app.py

How to use it: Open the page to see a searchable table with the model, its base model, creation date, Heretic version, KLD, refusal counts, trials, accelerator and a JSON file. Three actions are available.

Tool | What it does
Search filters | Narrow the table to the records you want
CSV download | Saves the full archive as a CSV file
Refresh | Reloads the visible table and the collector logs

Example: Filter the table down to one base model, then download the archive as CSV. You get the saved records in a spreadsheet-friendly file, even for models whose source repository is gone.

For newcomers and experts: Open the Space page and look at the table first. Each row is one saved reproduction record, and the filters help you find the model you care about.

You can read the collector's log in the page, though every fresh collection run empties it first. The archive keeps earlier records, so deleting a model's repository does not remove its saved entry. When you host your own copy, the persistent storage is what keeps the archive across restarts.

The collector runs twice a day
It finds new reproduction records on Hugging Face
The records are added to the archive
The table shows them for search and CSV export

Good to know: The app only archives public records and does not run or download models.

  • License: AGPL-3
  • Cost to run: none named, apart from persistent storage if you host it on Hugging Face
  • Data it reaches: public Heretic model records on Hugging Face, and the archive stored by the app

On Hugging Face, 13 November 2026

  • Likes: 41
  • Built with: gradio
  • License: agpl-3.0

Links

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.

Files
README.md  5 KB

1 file, 5 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-12

Compare with similar

Listed Hugging Face Spaces in the same category that share the most tags with this one.

ProductLikesBuilt withLicenseLast change
Heretic Grimoire This page41gradioagpl-3.0
Open VLM LeaderboardShares gradio1,034gradioapache-2.0
Face SwapShares gradio808gradiocc-by-nc-sa-4.0
TTS Spaces ArenaShares gradio493gradiozlib
Depth Anything 3Shares gradio490gradiocc-by-nc-4.0
BLIPShares gradio431gradiobsd-3-clause
Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13.

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

anaksor-ocr

Sokheng · Kit

Reads Khmer and English text from documents, receipts and photos of signs, using a detector for text lines and a recogniser for each line. Suits developers who…

New

Qwen2.5-7B-Instruct as GGUF files in eight quantization sizes, to chat with locally through llama.cpp. For people who want a capable 7.6B model on their own co…

New

ACE-Step 1.5 XL SFT

ACE-Step · Kit

Generates music from text with a 4B diffusion decoder, 50 steps and guidance control, aiming for higher audio quality than the base. For creators with a GPU of…

New

Similar in Research

See all in Research

Merlin-Cancer-Super

AbdomenAtlas · Kit

Detects and outlines tumors in 9 organs on CT scans, trained with the R-Super method that learns from radiology reports. A research model for medical imaging g…

New

EarthJev

dBenf · Kit

Answers typed questions about satellite image patches from all 12 Sentinel-2 bands, returning a probability for each option without writing text. A Qwen3.5-0.8…

New

getting-over-it-rl-jax

zimmerk4 · Kit

Plays the full game Getting Over It from the bottom to the end with a small trained policy network that learned inside a JAX physics simulator. For reinforceme…

New