Feluda

license

GPL-3.0(OSI copyleft)

source

public

core features withheld

no — all nine analysis operators ship in operators/ under the one GPL-3.0 license. Tattle offers no hosted edition and no paid tier

Feluda is GPL-3.0 licensed with the full source public, and every operator the engine ships lives in the same repository under that license. Tattle runs no hosted edition, pricing page or enterprise tier, and the README's one mention of a paid alternative, Google's Cloud Vision API, is something Feluda helps users avoid rather than a feature it withholds.

• https://api.github.com/repos/tattle-made/feluda recorded 2026-08-12

  Repo metadata: license spdx_id GPL-3.0, default branch main, 95 stars, not archived, under the tattle-made org.

• https://api.github.com/repos/tattle-made/feluda/contents/operators recorded 2026-08-12

  Every operator the README advertises is published here: classify_video_zero_shot, cluster_embeddings, detect_lewd_images, detect_text_in_image, dimension_reduction, image_vec_rep, vid_vec_rep, video_hash. Nothing is held back for a paid edition, and the root tree carries a single LICENSE.md with no ee/ or enterprise/ path beside it.

• https://github.com/tattle-made/feluda recorded 2026-06-22

  GPL-3.0 license, public source under tattle-made org

Feluda distributes no package and publishes no download, install or customer figure, leaving GitHub stars as the only available adoption signal. A star is not a use, so the reading stays directional.

• https://api.github.com/repos/tattle-made/feluda recorded 2026-08-12

  stargazers_count = 95 for tattle-made/feluda

Flexible modular engine focused on multimodal content analysis.

• https://github.com/tattle-made/feluda recorded 2026-08-13

  README still describes the modular operator engine - embeddings, hash search, image text extraction, entity extraction

• https://pypi.org/pypi/feluda/json recorded 2026-09-29

  The package README lists what the engine enables: "Support for hash-based search using pHash." and "Entity extraction from text and images and indexing into the engine." alongside vector embeddings and text extraction from images.

• https://tattle-made.github.io/feluda/ recorded 2026-09-29

  The documentation home says "Feluda is a configurable engine for analysing multi-lingual and multi-modal content" and lists "Easy to extend with custom operators" among its key features, covering text, images, audio and video.

• https://tattle-made.github.io/feluda/api/operators.html recorded 2026-09-29

  The operators reference lists eight operators: "ImageVecRep-Resnet Operator", "DetectLewdImages Operator", "DetectTextInImage Operator", "VidVecRep Operator", "VideoHash Operator", "Classify Video Zero Shot Operator", "Cluster Embeddings Operator" and "Dimension Reduction Operator".