Aguvis stage 2 trajectory data

The second-stage Aguvis collection of interface trajectories for learning action sequences, planning and reasoning after grounding pretraining.

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

What is this resource?

Resource type

Datasets

Environment

cross-platform

Topics

planning, data-generation

Record publication date

Not recorded

Evidence retrieved

2026-10-17T02:03:33Z

Resource ID

url:b5e2127e07f692d516b9

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

apache-2.0 (publisher model/dataset card metadata; third-party assets retain their own terms)

Document

Not recorded

Weights

Not recorded

Dataset profile

Supervision

GUI trajectory supervised fine-tuning, planning and reasoning with inner monologue

Modalities

GUI screenshots, natural-language instructions, action trajectories

Formats

JSON annotations, ZIP image packages, multipart TAR image packages

Splits

training

Scale and counting units

Stage 2 trajectory release, Hub revision bd6f1ed4, is distinct from Stage 1 grounding. The public manifest lists source-specific JSON levels and image ZIP/multipart TAR packages. The training recipe uses all rows for some inputs and random:33% for others; several l2 paths occur twice while the manifest also contains l3 files. Recipe rows/levels are neither independent datasets nor disjoint train/test splits. No example/action/trajectory count is inferred. Apache-2.0 is the card declaration, with separate underlying source terms.

Compare source-reviewed profiles

Recorded specifications

disabled

false

download url

https://huggingface.co/datasets/xlangai/aguvis-stage2/tree/main

gated

false

hub id

xlangai/aguvis-stage2

Explore related topics

Scope and limitations

The manifest contains trajectory annotations and archives associated with mobile, desktop and web sources. The card declares Apache-2.0; underlying source content may retain separate terms. This stage has a different training purpose and files from the grounding release. No archive or training sample was downloaded.

Related resources

• documented by: Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction

Primary evidence and verification

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

1. editorial primary review

   Retrieved 2026-10-17T02:03:33Z. Supports: name, kind, summary, environments, topics, scope, relations, details, details.dataset_profile.

   Content fingerprint

   550293f00905615e19e63ac896818f9875dad38fea0ffe196c6302a29719aa72
2. editorial primary review

   Retrieved 2026-10-17T02:03:33Z. Supports: identity, release_status, details, licenses.data, details.dataset_profile.

   Content fingerprint

   f030264a7a8c7485dc7cf15ca554d84450cc492d2ac90e5965810a0207ef370b
3. editorial primary review

   Retrieved 2026-10-16T06:36:00Z. Supports: environments, summary, details.dataset_profile.

   Content fingerprint

   2dd9f36b818c1b7b6bd772ba8575f6c0ebab6d9e0d307edcc2f196f3e4ba502e
4. editorial primary review

   Retrieved 2026-10-17T14:21:11Z. Supports: details.dataset_profile.

   Content fingerprint

   8c1b39e9d16067018ce845fd4e996f34e2b47346c342b2ba26b030f4a5338b66
5. huggingface card

   Retrieved 2026-10-17T02:03:33Z. Supports: licenses.data.

   Content fingerprint

   550293f00905615e19e63ac896818f9875dad38fea0ffe196c6302a29719aa72

Suggest a correction

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