Models and checkpoints

OmniParser v2 model bundle

A screen-parsing weight bundle containing icon-region detectors and an icon captioner for converting screenshots into agent-readable interface elements.

AI-assisted editorial review against a primary source. This is a catalog annotation, not an independent reproduction or endorsement.

Original source ↗Read as Markdown Record JSON

What is this resource?

Resource type: Models and checkpoints
Environment: unspecified
Topics: screen-understanding, grounding
Record publication date: Not recorded
Evidence retrieved: 2026-10-16T06:36:04Z
Resource ID: url:e21ca7c859544dbf5fb6

What licenses are recorded?

Code, training data, weights, and documents can have different terms. A blank declaration means unknown, not unrestricted use.

Code: Not recorded
Data: Not recorded
Document: Not recorded
Weights: Mixed by component: icon_detect = AGPL-3.0; icon_detect_v3 = MIT; icon_caption = MIT (Publisher per-folder model license declarations and retained LICENSE files; this is a component map, not one SPDX license for the full bundle.)

License review notes

Weights: The top-level Hub metadata says MIT, but the same card and folder licenses distinguish the original detector from the new detector and captioner. Preserve one model-bundle identity with a mixed license declaration. Source 1 Source 2 Source 3 Source 4

Recorded specifications

disabled: false
download url: https://huggingface.co/microsoft/OmniParser-v2.0/tree/main
gated: false
hub id: microsoft/OmniParser-v2.0

Explore related topics

Screen understanding GUI grounding

Scope and limitations

The manifest contains icon_detect/model.pt, icon_detect_v3/model.pt and icon_caption/model.safetensors. Licenses differ by folder: original YOLOv8 detector AGPL-3.0; YOLOv9-E detector and Florence-2 captioner MIT. The card’s top-level MIT metadata must not cover the entire bundle. Components remain one resource; no binaries were executed.

Primary evidence and verification

Source retrieval is separate from experiment reproduction. Use the original publication or publisher page when citing claims about the resource.

editorial primary review
Retrieved 2026-10-16T06:36:04Z. Supports: name, kind, summary, environments, topics, scope, relations, details.
Content fingerprint
f4ebaafcc0ab0b9e9cc55cb587678e5c83a5bbd64469fb2cc52bc63843c5cc6d
editorial primary review
Retrieved 2026-10-17T04:03:41Z. Supports: identity, release_status, details, licenses.weights.
Content fingerprint
1ea8af0cfd9cd716de8af8b78b2451470e1e1f22233a7286c5f40e1f496dc4be
editorial primary review
Retrieved 2026-10-17T04:03:41Z. Supports: summary, licenses.weights.
Content fingerprint
6ae154d65bfc2755703440be02acf63a37b741a1101b7c42b666da93a4ed2310
editorial primary review
Retrieved 2026-10-17T04:09:13Z. Supports: licenses.weights.
Content fingerprint
4df3c306dddaaf4baffdff5ca820cc679ac8cd6dc263c6a74517783e42fa7a3b
editorial primary review
Retrieved 2026-10-17T04:06:57Z. Supports: licenses.weights.
Content fingerprint
646f8936b8ddcd14e13e578ff6857e368780b0d1a4f6066bee89211923a373e2
editorial primary review
Retrieved 2026-10-17T04:09:13Z. Supports: licenses.weights.
Content fingerprint
d9a1b1e30d633d5732ea18e3cba9538d293ebc53e1a9e4e96ab739e0c5c4f1cb

Suggest a correction

Open a source-backed correction. Include this record ID and the original source. Read our editorial policy and coverage boundaries.