benchmark evidence
Remote Labor Index
Scale Labs benchmark for economically valuable end-to-end remote work automation.
winner on Remote Labor Index
direct benchmark result, not a broad vertical composite | source row dated 2025-10-27
scored on 2025-10-27 · stale source data (215d)
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | Anthropic: Claude Sonnet 4.5 | 48.4 | model-only independent_benchmark | 2025-10-27 |
| 2 | OpenAI: GPT-5 | 38.9 | model-only independent_benchmark | 2025-10-27 |
what this result means
Scale Labs benchmark for economically valuable end-to-end remote work automation.
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on Remote Labor Index. Broad task pages require independent corroboration before naming a general winner.
source record
category: agentic
metric: accuracy
matched models: 2
latest source date: 2025-10-27
direction: higher is better