Dataset
Data sourced from the official Workspace-Bench dataset and official Workspace-Bench-Lite dataset. This page is aligned to the latest public full and lite task metadata and shows only real static data.
Worker Profile Distribution
Distribution of the full benchmark task metadata from the latest official HuggingFace full split, with Lite statistics shown in the static bundle below.
| # | Worker profile | Tasks | Share |
|---|---|---|---|
| Loading... | |||
Collaboration Types
Each task is tagged with one or more collaboration types. A single task may require multiple abilities.
| # | Collaboration type | Tasks tagged |
|---|---|---|
| Loading... | ||
Rubric Evaluation Types
Rubric labels from the official metadata normalized into the three public reporting groups used on this site.
| Evaluation type | Count | Share |
|---|---|---|
| Loading... | ||
Task Browser
Browse local static task snapshots derived from the latest official full split.
| ID | Worker profile | Task description | Rubrics | Files |
|---|---|---|---|---|
| Loading... | ||||
API Playground
This module is temporarily marked and will be reopened after the public dataset interaction contract is finalized.
Coming soon
The site currently exposes only audited static benchmark statistics. Interactive API testing, custom task insertion, and session-side dataset mutation are disabled for now.
All statistics on this page are computed from the official
Workspace-Bench/Workspace-Bench
and
Workspace-Bench/Workspace-Bench-Lite.
The latest releases expose official full and lite splits with fields such as absolute_id, persona, task, task_diff, output_files, tested_capabilities, rubrics, rubric_types, file_dep_graph, and data_manifest.