Blog post · 2024
Raising the bar on SWE-bench Verified with Claude 3.5 Sonnet
| URL | https://www.anthropic.com/research/swe-bench-sonnet |
|---|---|
| Topics | ReplicHAI / Agent Evals / 02 Software engineering & coding |
| Tags | agent-eval |
Abstract
Measures: Agent scaffold design for SWE-bench. Metric: % resolved.
BibTeX
@misc{anthropic_raising_2024,
title = {Raising the bar on {SWE}-bench {Verified} with {Claude} 3.5 {Sonnet}},
url = {https://www.anthropic.com/research/swe-bench-sonnet},
abstract = {Measures: Agent scaffold design for SWE-bench. Metric: \% resolved.},
author = {{Anthropic}},
month = oct,
year = {2024},
keywords = {agent-eval},
}