Benchmarks

OSWorld-Human

Human reference trajectories and scoring code for measuring the action efficiency of computer-use agents on OSWorld tasks.

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: Benchmarks
Environment: desktop
Topics: evaluation, efficiency, workflow
Record publication date: Not recorded
Evidence retrieved: 2026-10-17T03:59:54Z
Resource ID: github:wuklab/osworld-human

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: Not recorded

Evaluation protocol

Focus: action efficiency relative to humans
Setting: offline
Observations: OSWorld result trajectories, human reference actions
Actions / outputs: efficiency scores
Evaluation: The scorer consumes OSWorld results and a specified maximum step budget; the README distinguishes single-action and grouped-action WES+ and WES-.

Scale and counting units

Not recorded

Reference solutions are published for evaluation and expressly discouraged for training. No task count is inferred from file count.

Compare source-reviewed profiles

Explore related topics

Evaluation Efficiency Workflows

Scope and limitations

The repository publishes per-application reference JSON files and score.py. It explicitly states that these evaluation solutions should not be used for training. No root code or data-license declaration was found; no license is inherited from OSWorld, and the score script was not executed.

Related resources

documented by: OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents

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 2026-10-17T03:59:54Z. Supports: name, kind, summary, environments, topics, scope, relations, details.
Content fingerprint
35c2f4dbed9d9bfc57f4c5a3f2fd70d313d6b463403b482c2fd4066d27b02cc1
editorial primary review
Retrieved 2026-10-17T03:59:54Z. Supports: identity.
Content fingerprint
21c86601da6e77a3326fbcd143e3c5d9a033cd59e6e13973250966dff6f51ae9
editorial primary review
Retrieved 2026-10-17T03:59:55Z. Supports: release_status.
Content fingerprint
c7b958d7a075adf5d282be08b68ad2b090568da342d4cb4b501ae808e54d0053

Suggest a correction

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