Search results

Investigate GAIA null entries in output.jsonl

Status: Open.

#291 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 9, 2026

Persist conversation events to disk during runs

Status: Open.

#284 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 8, 2026

Apptainer for Swebench

Status: Open.

#283 In OpenHands/benchmarks;

· lilys012 opened on Jan 8, 2026

Run 50-image SWE-bench eval to debug runtime instability

Status: Open.

#282 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 8, 2026

Add Datadog metrics permissions to API key for runtime debugging

Status: Open.

#281 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 8, 2026

Refine runtime failure classification for resource backoff

Status: Open.

#277 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 7, 2026

Fix malformed cache tags (tag_latest truncated) in SWE-bench build outputs

Status: Open.

#274 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 7, 2026

Parallelize SWT-bench eval for 400+ instances

Status: Open.

#257 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 6, 2026

Track full GAIA 2023_all test sweep (301 instances)

Status: Open.

#256 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 6, 2026

Track Commit0 lite test sweep (default 16 repos)

Status: Open.

#255 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 6, 2026

Clarify cost_report summary (total_cost vs only_main_output_cost)

Status: Open.

#249 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 5, 2026

Track 500-image SWE-bench eval (eval-20699625809-gpt-5-mini)

Status: Open.

#239 In OpenHands/benchmarks;

· simonrosenberg opened on Jan 4, 2026