Benchmarks
The benchmark runs below link to pages containing each run’s detailed observations and a browser scoped to that run. Use the cross-run browser to explore all published runs, or see the aggregate overview and its per-benchmark metric details.
Published runs: 63. Selected best-case pairs: 358.
Benchmark runs
- alltasks-r15-gemini37-explicit-prescriptive-20261006 (120 comparison records)
- alltasks-r15-luna6-explicit-prescriptive-20261006 (630 comparison records)
- gemini37-quota-probe-20261006 (2 comparison records)
- gpt-6-luna-all-r10-c5-prescriptive-20260927 (0 comparison records)
- gpt-6-luna-all-r10-c5-prescriptive-20260927-retry (70 comparison records)
- gpt-6-luna-all-r10-c5-prescriptive-postfix-20260927 (70 comparison records)
- gpt-6-luna-all-r5-c10-descriptive-20260926 (35 comparison records)
- gpt-6-luna-all-r5-c10-none-20260926 (0 comparison records)
- gpt-6-luna-all-r5-c10-none-20260926-retry (35 comparison records)
- gpt-6-luna-all-r5-c10-prescriptive-20260926 (35 comparison records)
- hard-smoke-luna6-medium-20261011T105400Z (6 comparison records)
- hard-smoke-saluki-20261011T113200Z (6 comparison records)
- hard-smoke-saluki-20261011T203354Z (6 comparison records)
- hard-smoke-saluki-fixed-20261011T165800Z (6 comparison records)
- hard-smoke-saluki-rerun-20261011T191650Z (6 comparison records)
- luna-medium-edit-scale-distinct-r10-20260928T223304Z (60 comparison records)
- luna-medium-five-rename-multifile-r10-20260929T1134Z (10 comparison records)
- luna-medium-staged-followup-r10-20260929T054733Z (10 comparison records)
- luna-medium-staged-followup-smoke-20260929 (1 comparison records)
- luna-medium-staged-prewarm-r10-20260929 (10 comparison records)
- luna-medium-staged-prewarm-smoke-20260929 (1 comparison records)
- luna6-multiswe-go-hints-smoke-20261007-r1 (4 comparison records)
- luna6-multiswe-go-smoke-20261007-r3 (24 comparison records)
- opencode-qwen36-iq4-xs-hard-20261004-1635 (3 comparison records)
- opencode-qwen36-iq4-xs-post-bugfix-tail3-20261005-1057 (3 comparison records)
- opencode-qwen36-iq4-xs-post-schema-10x-20261005-0022 (40 comparison records)
- opencode-qwen36-iq4-xs-smoke-20261004 (4 comparison records)
- opencode-qwen36-iq4-xs-task004-20261004-1622 (1 comparison records)
- retry-20260925-agy-descriptive-task11-large-default-baseline-killed-escalated (1 comparison records)
- retry-20260925-agy-descriptive-task11-large-default-semedit-killed (1 comparison records)
- retry-20260925-agy-descriptive-task11-large-discover-baseline-quota (1 comparison records)
- retry-20260925-agy-descriptive-task11-large-discover-semedit-quota (1 comparison records)
- retry-20260925-agy-prescriptive-task11-large-default-baseline-quota (1 comparison records)
- retry-20260925-agy-prescriptive-task11-large-default-semedit-quota (1 comparison records)
- retry-20260925-agy-prescriptive-task11-large-discover-baseline-quota (1 comparison records)
- retry-20260925-agy-prescriptive-task11-large-discover-semedit-quota (1 comparison records)
- retry-20260925-agy-prescriptive-task11-small-discover-semedit-killed (1 comparison records)
- retry-20260927-agy-descriptive-task11-large-default-latest-b4be (0 comparison records)
- run-20260924-agy-descriptive (14 comparison records)
- run-20260924-agy-prescriptive (14 comparison records)
- run-20260924-codex-descriptive (35 comparison records)
- run-20260924-codex-prescriptive (35 comparison records)
- saluki-opencode-relay-20261011T095400Z (0 comparison records)
- saluki-opencode-smallest-20261011084912 (0 comparison records)
- saluki-opencode-terminal-runtime-20261011T110500Z (2 comparison records)
- swe-smoke-luna6-medium-20261011T105900Z (7 comparison records)
- swe-smoke-saluki-20261011T120900Z (2 comparison records)
- swe-smoke-saluki-20261011T203354Z (0 comparison records)
- swe-smoke-saluki-fixed-20261011T165800Z (0 comparison records)
- task-04-insert-public (3 comparison records)
- task-07-generate-template-main (6 comparison records)
- task-09-composite-refactor (3 comparison records)
- task-11-mixed-sink-api-migration (3 comparison records)
- task-12-strengthened-readwrite-20260929 (10 comparison records)
- task-12-strengthened-write-20260929 (10 comparison records)
- task-all-gpt-6-luna-medium-prescriptive-repeats10-20260928T164359Z (70 comparison records)
- task02-corrected-overlay-explicit-r15-luna6-20261006 (30 comparison records)
- task02-corrected-overlay-r15-luna6-20261006 (30 comparison records)
- task11-explicit-smoke-luna6-gemini37-20261005 (0 comparison records)
- task11-explicit-smoke-luna6-gemini37-retry-20261005 (4 comparison records)
- task11-gpt-6-luna-medium-mcp-matrix-20260927-retry (6 comparison records)
- task11-gpt-6-luna-medium-mcp-matrix-repeats10-20260927 (60 comparison records)
- task11-gpt-6-luna-medium-mcp-matrix-retry-failed-cell-20260927 (1 comparison records)