LongMemEval Run Browser
Browse LME benchmark runs, inspect per-task scores, and (for the active run) view live graph projections.
{{ run.label }}
{% if run.is_active %} ACTIVE {% elif run.live_graph_source_run_id %} LIVE SOURCE {% else %} ARTIFACT {% endif %}Benchmark catalog is not configured.
Set MENHIR_BENCH_RESULTS_ROOT to point at the archolith-bench results directory.
No benchmark runs found.
The results directory does not contain any run data.