Run 70b164dc
Jul 10, 2026, 16:08
AntiGravity · 2 models · repeat ×5
70b164dc-6ef6-4a75-93d6-33918b6faa41
Downloads
Audit pack
Results
2 models
| # | Adapter | Model | Think | Pass % | Cover % | Score | Outcomes | Time |
|---|---|---|---|---|---|---|---|---|
| 1 | AntiGravity | Gemini 3.1 Pro | Low | 86.8% | 95.8% | — | 158/18/6/8 | 11.4s |
#1 Gemini 3.1 ProLowPass rate86.8% 158 pass · 182 scored Coverage95.8% 182 / 190 evidence-bearing Verified score— 95% interval — · 1 run(s) pass@k91.9% 34/38 fixtures ever solved Outcome mix
Invalid output means the model returned content that broke the answer contract, so it scores zero. Harness failures returned no answer and are excluded from score. Attempt funnel
Run facts
Latest run fixtures34/38 passed in linked run · fails firstperf-string-table-concat0/5perf-darkrpvar1/5perf-find-ents-near1/5perf-steamid-cache1/5perf-localplayer-cache0/5perf-looking-at0/5player-iterator0/5ents-iterator5/5hook-add5/5isvalid5/5net-receivers5/5perf-angle-zero5/5perf-config-var5/5perf-disttosqr5/5perf-ents-iterator5/5perf-find-players-near5/5perf-for-vs-while5/5perf-hook-once5/5perf-hudpaint-cache5/5perf-local-color5/5perf-local-vs-global5/5perf-meta-vs-arg5/5perf-msg-vs-print5/5perf-pairs-ipairs-for5/5perf-plain-lua-vs-fn5/5perf-player-cache-hooks5/5perf-ply-table-index5/5perf-seq-tables5/5perf-setdrawcolor-split5/5perf-sqlite-vs-mysql5/5perf-surface-drawrect5/5perf-surface-text5/5perf-table-count-hash5/5perf-table-empty5/5perf-table-hasvalue5/5perf-table-literal5/5perf-table-random5/5perf-x-times-x5/5 | ||||||||
| 2 | AntiGravity | Gemini 3.1 Pro | High | 88.6% | 60.0% | — | 101/9/4/76 | 11.3s |
#2 Gemini 3.1 ProHighPass rate88.6% 101 pass · 114 scored Coverage60.0% 114 / 190 evidence-bearing Verified score— 95% interval — · 1 run(s) pass@k91.7% 22/38 fixtures ever solved Outcome mix
Invalid output means the model returned content that broke the answer contract, so it scores zero. Harness failures returned no answer and are excluded from score. Attempt funnel
Run facts
Latest run fixtures22/38 passed in linked run · fails firstperf-seq-tables0/5perf-setdrawcolor-split0/5perf-sqlite-vs-mysql0/5perf-steamid-cache0/5perf-string-table-concat0/5perf-surface-drawrect0/5perf-surface-text0/5perf-table-count-hash0/5perf-table-empty0/5perf-table-hasvalue0/5perf-table-literal0/5perf-table-random0/5perf-x-times-x0/5player-iterator0/5perf-darkrpvar4/5perf-localplayer-cache1/5perf-pairs-ipairs-for3/5perf-ply-table-index3/5perf-find-ents-near0/5perf-looking-at0/5ents-iterator5/5hook-add5/5isvalid5/5net-receivers5/5perf-angle-zero5/5perf-config-var5/5perf-disttosqr5/5perf-ents-iterator5/5perf-find-players-near5/5perf-for-vs-while5/5perf-hook-once5/5perf-hudpaint-cache5/5perf-local-color5/5perf-local-vs-global5/5perf-meta-vs-arg5/5perf-msg-vs-print5/5perf-plain-lua-vs-fn5/5perf-player-cache-hooks5/5 | ||||||||
Verified score aggregates compatible runs. Outcomes = pass / partial / wrong / invalid output; harness failures are excluded. Click a row for evidence and the linked run's fixtures.
Fixtures
38
gmod.ents-iterator.v1gmod.hook-add.v1gmod.isvalid.v1gmod.net-receivers.v1gmod.perf.angle-zero.v1gmod.perf.config-var.v1gmod.perf.darkrpvar.v1gmod.perf.disttosqr.v1gmod.perf.ents-iterator.v1gmod.perf.find-ents-near.v1gmod.perf.find-players-near.v1gmod.perf.for-vs-while.v1gmod.perf.hook-once.v1gmod.perf.hudpaint-cache.v1gmod.perf.local-color.v1gmod.perf.local-vs-global.v1gmod.perf.localplayer-cache.v1gmod.perf.looking-at.v1gmod.perf.meta-vs-arg.v1gmod.perf.msg-vs-print.v1gmod.perf.pairs-ipairs-for.v1gmod.perf.plain-lua-vs-fn.v1gmod.perf.player-cache-hooks.v1gmod.perf.ply-table-index.v1gmod.perf.seq-tables.v1gmod.perf.setdrawcolor-split.v1gmod.perf.sqlite-vs-mysql.v1gmod.perf.steamid-cache.v1gmod.perf.string-table-concat.v1gmod.perf.surface-drawrect.v1gmod.perf.surface-text.v1gmod.perf.table-count-hash.v1gmod.perf.table-empty.v1gmod.perf.table-hasvalue.v1gmod.perf.table-literal.v1gmod.perf.table-random.v1gmod.perf.x-times-x.v1gmod.player-iterator.v1