SWE-bench Live

What it measures

Continuously updated contamination-free stream of newly closed GitHub issues created after model training cutoffs.

Reading the measurement

Look beyond a plausible code sample to the executable artifact. Tests, repository context, language coverage and time limits shape the result. Separate first-attempt correctness from success after debugging, retries or agent assistance.

Published evaluation

Resolved tasks; SWE-bench Live · Lite comparison. Model configurations and missing observations follow the original publication.

Where this fits in an evaluation suite

The agent wrapper is part of the system. Pin tools, dependencies, timeouts, tests, and the repository revision. Compare contest coding separately from repository maintenance.

Start with a capability hypothesis, then choose a second test with a different failure mode. Keep the checkpoint, reasoning effort, agent tools, sampling policy, and scoring procedure fixed when measuring a training intervention. Record failures and missing outputs in the denominator.

A reproducible comparison

Resolved tasks; SWE-bench Live · Lite comparison. Model configurations and missing observations follow the original publication.

Publish the exact run configuration, item coverage, score units, and available uncertainty. If a source supplies only an aggregate or a single run, preserve that limit. A new model name or a rounded score alone does not establish a capability improvement.