Run c56989a5

Name: gmod-bench run c56989a5-0aa1-4fb5-be76-721322569d56
Creator: gmod-bench
License: https://github.com/Iydah/gmod-bench/blob/main/LICENSE

Jul 10, 2026, 22:54

OpenCode · 17 models · repeat ×1

c56989a5-0aa1-4fb5-be76-721322569d56

Fixtures

554

Attempts

Models

12m 31s

Duration

OpenCode×1 repeat

Downloads

Audit pack

Results

17 models

#	Adapter	Model	Think	Pass %	Cover %	Score	Outcomes	Time
1	OpenCode	deepseek-v4-flash-free	Low	73.6%	93.0%	73.7%	39/6/8/4	19.5s
#1 deepseek-v4-flash-free Low OpenCode Pass rate73.6% 39 pass · 53 scored Coverage93.0% 53 / 57 evidence-bearing Verified score73.7% 95% interval — · 1 run(s) pass@k73.6% 39/57 fixtures ever solved Outcome mix Pass39 Partial6 Wrong8 Invalid output4 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 Attempts57 Scored53 Passed39 Fixtures solved39/57 Run facts Avg time 19.5s Repeat Single-shot Evidence share 53/57 (93.0%) Verified runs 1 95% interval — Raw label opencode/deepseek-v4-flash-free@low Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 39/57 passed in linked run perf-seq-tablesperf-string-table-concatspawn-preventive-rate-limittransfer-lifecycle-cleanupdarkrp-shop-npcnet-bounded-chunk-transfernet-compact-targeted-updatenet-per-player-budgetperf-plain-lua-vs-fnperf-table-hasvaluesqlite-batched-transactionsqlite-typed-writefile-data-root-confinementnet-bounded-decompressionnet-secure-entity-actionperf-find-ents-nearperf-looking-atrealm-shared-authorityents-iteratorents-iterator-readonlyhook-addhook-object-lifecyclehudpaint-hot-pathisvalidnet-receiversnetworkvar-entity-stateperf-angle-zeroperf-config-varperf-darkrpvarperf-disttosqrperf-ents-iteratorperf-find-players-nearperf-for-vs-whileperf-hook-onceperf-hudpaint-cacheperf-local-colorperf-local-vs-globalperf-localplayer-cacheperf-meta-vs-argperf-msg-vs-printperf-pairs-ipairs-forperf-player-cache-hooksperf-ply-table-indexperf-setdrawcolor-splitperf-sqlite-vs-mysqlperf-steamid-cacheperf-surface-drawrectperf-surface-textperf-table-count-hashperf-table-emptyperf-table-literalperf-table-randomperf-x-times-xplayer-iteratorprediction-one-shot-effectspatial-maintained-settimer-delayed-validity
2	OpenCode	deepseek-v4-flash-free	High	66.7%	94.7%	70.2%	36/8/10/3	20.6s
#2 deepseek-v4-flash-free High OpenCode Pass rate66.7% 36 pass · 54 scored Coverage94.7% 54 / 57 evidence-bearing Verified score70.2% 95% interval — · 1 run(s) pass@k66.7% 36/57 fixtures ever solved Outcome mix Pass36 Partial8 Wrong10 Invalid output3 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 Attempts57 Scored54 Passed36 Fixtures solved36/57 Run facts Avg time 20.6s Repeat Single-shot Evidence share 54/57 (94.7%) Verified runs 1 95% interval — Raw label opencode/deepseek-v4-flash-free@high Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 36/57 passed in linked run perf-for-vs-whileperf-seq-tablesperf-string-table-concatdarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnetworkvar-entity-stateperf-angle-zeroperf-darkrpvarperf-ply-table-indexspatial-maintained-setsqlite-batched-transactionsqlite-typed-writefile-data-root-confinementnet-per-player-budgetnet-secure-entity-actionperf-find-ents-nearperf-looking-atrealm-shared-authorityspawn-preventive-rate-limittransfer-lifecycle-cleanupents-iteratorents-iterator-readonlyhook-addhook-object-lifecyclehudpaint-hot-pathisvalidnet-compact-targeted-updatenet-receiversperf-config-varperf-disttosqrperf-ents-iteratorperf-find-players-nearperf-hook-onceperf-hudpaint-cacheperf-local-colorperf-local-vs-globalperf-localplayer-cacheperf-meta-vs-argperf-msg-vs-printperf-pairs-ipairs-forperf-plain-lua-vs-fnperf-player-cache-hooksperf-setdrawcolor-splitperf-sqlite-vs-mysqlperf-steamid-cacheperf-surface-drawrectperf-surface-textperf-table-count-hashperf-table-emptyperf-table-hasvalueperf-table-literalperf-table-randomperf-x-times-xplayer-iteratorprediction-one-shot-effecttimer-delayed-validity
3	OpenCode	deepseek-v4-flash-free	Medium	63.0%	94.7%	67.5%	34/9/11/3	19.5s
#3 deepseek-v4-flash-free Medium OpenCode Pass rate63.0% 34 pass · 54 scored Coverage94.7% 54 / 57 evidence-bearing Verified score67.5% 95% interval — · 1 run(s) pass@k63.0% 34/57 fixtures ever solved Outcome mix Pass34 Partial9 Wrong11 Invalid output3 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 Attempts57 Scored54 Passed34 Fixtures solved34/57 Run facts Avg time 19.5s Repeat Single-shot Evidence share 54/57 (94.7%) Verified runs 1 95% interval — Raw label opencode/deepseek-v4-flash-free@medium Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 34/57 passed in linked run net-per-player-budgetperf-find-ents-nearperf-string-table-concatdarkrp-shop-npcents-iterator-readonlynet-compact-targeted-updatenetworkvar-entity-stateperf-darkrpvarperf-plain-lua-vs-fnprediction-one-shot-effectspatial-maintained-setsqlite-batched-transactionsqlite-typed-writetransfer-lifecycle-cleanupfile-data-root-confinementnet-bounded-chunk-transfernet-bounded-decompressionnet-secure-entity-actionperf-angle-zeroperf-looking-atplayer-iteratorrealm-shared-authorityspawn-preventive-rate-limitents-iteratorhook-addhook-object-lifecyclehudpaint-hot-pathisvalidnet-receiversperf-config-varperf-disttosqrperf-ents-iteratorperf-find-players-nearperf-for-vs-whileperf-hook-onceperf-hudpaint-cacheperf-local-colorperf-local-vs-globalperf-localplayer-cacheperf-meta-vs-argperf-msg-vs-printperf-pairs-ipairs-forperf-player-cache-hooksperf-ply-table-indexperf-seq-tablesperf-setdrawcolor-splitperf-sqlite-vs-mysqlperf-steamid-cacheperf-surface-drawrectperf-surface-textperf-table-count-hashperf-table-emptyperf-table-hasvalueperf-table-literalperf-table-randomperf-x-times-xtimer-delayed-validity
4	OpenCode	deepseek-v4-flash-free	Max	61.1%	90.0%	62.5%	22/6/8/4	27.7s
#4 deepseek-v4-flash-free Max OpenCode Pass rate61.1% 22 pass · 36 scored Coverage90.0% 36 / 40 evidence-bearing Verified score62.5% 95% interval — · 1 run(s) pass@k61.1% 22/40 fixtures ever solved Outcome mix Pass22 Partial6 Wrong8 Invalid output4 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 Attempts40 Scored36 Passed22 Fixtures solved22/40 Run facts Avg time 27.7s Repeat Single-shot Evidence share 36/40 (90.0%) Verified runs 1 95% interval — Raw label opencode/deepseek-v4-flash-free@max Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 22/40 passed in linked run net-per-player-budgetperf-seq-tablesperf-string-table-concatspawn-preventive-rate-limitdarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-compact-targeted-updateperf-angle-zerorealm-shared-authoritysqlite-batched-transactiontransfer-lifecycle-cleanupfile-data-root-confinementnet-secure-entity-actionperf-find-ents-nearperf-looking-atspatial-maintained-setsqlite-typed-writeents-iteratorents-iterator-readonlyhook-addhook-object-lifecyclehudpaint-hot-pathisvalidnet-receiversnetworkvar-entity-stateperf-config-varperf-darkrpvarperf-disttosqrperf-ents-iteratorperf-find-players-nearperf-for-vs-whileperf-hook-onceperf-hudpaint-cacheperf-local-colorperf-local-vs-globalperf-localplayer-cacheperf-meta-vs-argprediction-one-shot-effecttimer-delayed-validity
5	OpenCode	big-pickle	—	56.9%	89.5%	57.0%	29/7/15/6	21.3s
#5 big-pickle OpenCode Pass rate56.9% 29 pass · 51 scored Coverage89.5% 51 / 57 evidence-bearing Verified score57.0% 95% interval — · 1 run(s) pass@k56.9% 29/57 fixtures ever solved Outcome mix Pass29 Partial7 Wrong15 Invalid output6 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 Attempts57 Scored51 Passed29 Fixtures solved29/57 Run facts Avg time 21.3s Repeat Single-shot Evidence share 51/57 (89.5%) Verified runs 1 95% interval — Raw label opencode/big-pickle Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 29/57 passed in linked run perf-find-ents-nearperf-seq-tablesperf-string-table-concatperf-surface-textperf-table-hasvaluetransfer-lifecycle-cleanupdarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-compact-targeted-updatenetworkvar-entity-stateperf-angle-zeroperf-config-varperf-darkrpvarperf-disttosqrperf-plain-lua-vs-fnperf-ply-table-indexrealm-shared-authorityspatial-maintained-setspawn-preventive-rate-limitsqlite-batched-transactionfile-data-root-confinementnet-per-player-budgetnet-secure-entity-actionperf-ents-iteratorperf-find-players-nearperf-looking-atsqlite-typed-writeents-iteratorents-iterator-readonlyhook-addhook-object-lifecyclehudpaint-hot-pathisvalidnet-receiversperf-for-vs-whileperf-hook-onceperf-hudpaint-cacheperf-local-colorperf-local-vs-globalperf-localplayer-cacheperf-meta-vs-argperf-msg-vs-printperf-pairs-ipairs-forperf-player-cache-hooksperf-setdrawcolor-splitperf-sqlite-vs-mysqlperf-steamid-cacheperf-surface-drawrectperf-table-count-hashperf-table-emptyperf-table-literalperf-table-randomperf-x-times-xplayer-iteratorprediction-one-shot-effecttimer-delayed-validity
6	OpenCode	hy3-free	Low	33.3%	91.3%	50.0%	7/9/5/2	28.6s
#6 hy3-free Low OpenCode Pass rate33.3% 7 pass · 21 scored Coverage91.3% 21 / 23 evidence-bearing Verified score50.0% 95% interval — · 1 run(s) pass@k33.3% 7/23 fixtures ever solved Outcome mix Pass7 Partial9 Wrong5 Invalid output2 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 Attempts23 Scored21 Passed7 Fixtures solved7/23 Run facts Avg time 28.6s Repeat Single-shot Evidence share 21/23 (91.3%) Verified runs 1 95% interval — Raw label opencode/hy3-free@low Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 7/23 passed in linked run perf-find-ents-nearperf-string-table-concatnet-compact-targeted-updateprediction-one-shot-effectrealm-shared-authorityspatial-maintained-setsqlite-batched-transactiondarkrp-shop-npcfile-data-root-confinementnet-bounded-chunk-transfernet-bounded-decompressionnet-per-player-budgetnet-secure-entity-actionspawn-preventive-rate-limitsqlite-typed-writetransfer-lifecycle-cleanupents-iteratorents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnetworkvar-entity-stateperf-seq-tablestimer-delayed-validity
7	OpenCode	hy3-free	Medium	33.3%	85.7%	47.6%	6/8/4/3	25.3s
#7 hy3-free Medium OpenCode Pass rate33.3% 6 pass · 18 scored Coverage85.7% 18 / 21 evidence-bearing Verified score47.6% 95% interval — · 1 run(s) pass@k33.3% 6/21 fixtures ever solved Outcome mix Pass6 Partial8 Wrong4 Invalid output3 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 Attempts21 Scored18 Passed6 Fixtures solved6/21 Run facts Avg time 25.3s Repeat Single-shot Evidence share 18/21 (85.7%) Verified runs 1 95% interval — Raw label opencode/hy3-free@medium Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 6/21 passed in linked run net-compact-targeted-updateperf-string-table-concatspatial-maintained-setdarkrp-shop-npcperf-plain-lua-vs-fnspawn-preventive-rate-limitsqlite-batched-transactionfile-data-root-confinementnet-bounded-chunk-transfernet-per-player-budgetnet-secure-entity-actionprediction-one-shot-effectrealm-shared-authoritysqlite-typed-writetransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnet-bounded-decompressionnetworkvar-entity-statetimer-delayed-validity
8	OpenCode	hy3-free	High	26.3%	95.0%	47.5%	5/9/5/1	24.2s
#8 hy3-free High OpenCode Pass rate26.3% 5 pass · 19 scored Coverage95.0% 19 / 20 evidence-bearing Verified score47.5% 95% interval — · 1 run(s) pass@k26.3% 5/20 fixtures ever solved Outcome mix Pass5 Partial9 Wrong5 Invalid output1 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 Attempts20 Scored19 Passed5 Fixtures solved5/20 Run facts Avg time 24.2s Repeat Single-shot Evidence share 19/20 (95.0%) Verified runs 1 95% interval — Raw label opencode/hy3-free@high Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 5/20 passed in linked run perf-string-table-concatdarkrp-shop-npcnet-compact-targeted-updatespatial-maintained-setspawn-preventive-rate-limitsqlite-batched-transactionfile-data-root-confinementnet-bounded-chunk-transfernet-bounded-decompressionnet-per-player-budgetnet-secure-entity-actionprediction-one-shot-effectrealm-shared-authoritysqlite-typed-writetransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnetworkvar-entity-statetimer-delayed-validity
9	OpenCode	nemotron-3-ultra-free	Low	36.8%	86.4%	43.2%	7/5/7/3	22.9s
#9 nemotron-3-ultra-free Low OpenCode Pass rate36.8% 7 pass · 19 scored Coverage86.4% 19 / 22 evidence-bearing Verified score43.2% 95% interval — · 1 run(s) pass@k36.8% 7/22 fixtures ever solved Outcome mix Pass7 Partial5 Wrong7 Invalid output3 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 Attempts22 Scored19 Passed7 Fixtures solved7/22 Run facts Avg time 22.9s Repeat Single-shot Evidence share 19/22 (86.4%) Verified runs 1 95% interval — Raw label opencode/nemotron-3-ultra-free@low Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 7/22 passed in linked run perf-seq-tablesperf-string-table-concatspatial-maintained-setdarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-per-player-budgetprediction-one-shot-effectsqlite-batched-transactionsqlite-typed-writefile-data-root-confinementnet-secure-entity-actionrealm-shared-authorityspawn-preventive-rate-limittransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnet-compact-targeted-updatenetworkvar-entity-stateperf-hook-oncetimer-delayed-validity
10	OpenCode	nemotron-3-ultra-free	High	41.2%	77.3%	43.2%	7/5/5/5	16.2s
#10 nemotron-3-ultra-free High OpenCode Pass rate41.2% 7 pass · 17 scored Coverage77.3% 17 / 22 evidence-bearing Verified score43.2% 95% interval — · 1 run(s) pass@k41.2% 7/22 fixtures ever solved Outcome mix Pass7 Partial5 Wrong5 Invalid output5 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 Attempts22 Scored17 Passed7 Fixtures solved7/22 Run facts Avg time 16.2s Repeat Single-shot Evidence share 17/22 (77.3%) Verified runs 1 95% interval — Raw label opencode/nemotron-3-ultra-free@high Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 7/22 passed in linked run darkrp-shop-npcperf-seq-tablesperf-string-table-concatspatial-maintained-setspawn-preventive-rate-limitnet-bounded-chunk-transfernet-bounded-decompressionprediction-one-shot-effectsqlite-batched-transactionsqlite-typed-writefile-data-root-confinementnet-per-player-budgetnet-secure-entity-actionrealm-shared-authoritytransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnet-compact-targeted-updatenetworkvar-entity-stateperf-for-vs-whiletimer-delayed-validity
11	OpenCode	mimo-v2.5-free	Low	30.0%	90.9%	40.9%	6/6/8/2	24.6s
#11 mimo-v2.5-free Low OpenCode Pass rate30.0% 6 pass · 20 scored Coverage90.9% 20 / 22 evidence-bearing Verified score40.9% 95% interval — · 1 run(s) pass@k30.0% 6/22 fixtures ever solved Outcome mix Pass6 Partial6 Wrong8 Invalid output2 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 Attempts22 Scored20 Passed6 Fixtures solved6/22 Run facts Avg time 24.6s Repeat Single-shot Evidence share 20/22 (90.9%) Verified runs 1 95% interval — Raw label opencode/mimo-v2.5-free@low Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 6/22 passed in linked run perf-string-table-concatsqlite-batched-transactiondarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-compact-targeted-updateperf-table-hasvaluespatial-maintained-setsqlite-typed-writetransfer-lifecycle-cleanupfile-data-root-confinementnet-per-player-budgetnet-secure-entity-actionperf-angle-zerorealm-shared-authorityspawn-preventive-rate-limitents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnetworkvar-entity-stateprediction-one-shot-effecttimer-delayed-validity
12	OpenCode	mimo-v2.5-free	Medium	33.3%	85.7%	40.5%	6/5/7/2	28.9s
#12 mimo-v2.5-free Medium OpenCode Pass rate33.3% 6 pass · 18 scored Coverage85.7% 18 / 21 evidence-bearing Verified score40.5% 95% interval — · 1 run(s) pass@k33.3% 6/21 fixtures ever solved Outcome mix Pass6 Partial5 Wrong7 Invalid output2 Harness1 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 Attempts21 Scored18 Passed6 Fixtures solved6/21 Run facts Avg time 28.9s Repeat Single-shot Evidence share 18/21 (85.7%) Verified runs 1 95% interval — Raw label opencode/mimo-v2.5-free@medium Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 6/21 passed in linked run sqlite-typed-writeperf-string-table-concatprediction-one-shot-effectdarkrp-shop-npcnet-bounded-chunk-transfernet-compact-targeted-updatenet-per-player-budgetrealm-shared-authorityspatial-maintained-setsqlite-batched-transactionfile-data-root-confinementnet-bounded-decompressionnet-secure-entity-actionspawn-preventive-rate-limittransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclehudpaint-hot-pathnetworkvar-entity-stateperf-pairs-ipairs-fortimer-delayed-validity
13	OpenCode	mimo-v2.5-free	High	35.3%	73.9%	37.0%	6/5/6/6	23.0s
#13 mimo-v2.5-free High OpenCode Pass rate35.3% 6 pass · 17 scored Coverage73.9% 17 / 23 evidence-bearing Verified score37.0% 95% interval — · 1 run(s) pass@k35.3% 6/23 fixtures ever solved Outcome mix Pass6 Partial5 Wrong6 Invalid output6 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 Attempts23 Scored17 Passed6 Fixtures solved6/23 Run facts Avg time 23.0s Repeat Single-shot Evidence share 17/23 (73.9%) Verified runs 1 95% interval — Raw label opencode/mimo-v2.5-free@high Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 6/23 passed in linked run hook-object-lifecycleperf-string-table-concatperf-table-hasvaluespatial-maintained-setspawn-preventive-rate-limitsqlite-batched-transactiondarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-compact-targeted-updatenetworkvar-entity-staterealm-shared-authorityfile-data-root-confinementnet-per-player-budgetnet-secure-entity-actionsqlite-typed-writetransfer-lifecycle-cleanupents-iterator-readonlyhudpaint-hot-pathperf-seq-tablesperf-table-emptyprediction-one-shot-effecttimer-delayed-validity
14	OpenCode	nemotron-3-ultra-free	Medium	35.0%	80.0%	36.0%	7/4/9/5	15.4s
#14 nemotron-3-ultra-free Medium OpenCode Pass rate35.0% 7 pass · 20 scored Coverage80.0% 20 / 25 evidence-bearing Verified score36.0% 95% interval — · 1 run(s) pass@k35.0% 7/25 fixtures ever solved Outcome mix Pass7 Partial4 Wrong9 Invalid output5 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 Attempts25 Scored20 Passed7 Fixtures solved7/25 Run facts Avg time 15.4s Repeat Single-shot Evidence share 20/25 (80.0%) Verified runs 1 95% interval — Raw label opencode/nemotron-3-ultra-free@medium Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 7/25 passed in linked run hudpaint-hot-pathperf-seq-tablesperf-string-table-concatspatial-maintained-setspawn-preventive-rate-limitdarkrp-shop-npcnet-bounded-chunk-transfernet-bounded-decompressionnet-compact-targeted-updateperf-table-hasvalueprediction-one-shot-effectrealm-shared-authoritysqlite-batched-transactionsqlite-typed-writefile-data-root-confinementnet-per-player-budgetnet-secure-entity-actiontransfer-lifecycle-cleanupents-iterator-readonlyhook-object-lifecyclenetworkvar-entity-stateperf-for-vs-whileperf-steamid-cacheperf-table-randomtimer-delayed-validity
15	OpenCode	north-mini-code-free	High	41.2%	56.7%	26.7%	7/2/8/10	20.5s
#15 north-mini-code-free High OpenCode Pass rate41.2% 7 pass · 17 scored Coverage56.7% 17 / 30 evidence-bearing Verified score26.7% 95% interval — · 1 run(s) pass@k41.2% 7/30 fixtures ever solved Outcome mix Pass7 Partial2 Wrong8 Invalid output10 Harness3 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 Attempts30 Scored17 Passed7 Fixtures solved7/30 Run facts Avg time 20.5s Repeat Single-shot Evidence share 17/30 (56.7%) Verified runs 1 95% interval — Raw label opencode/north-mini-code-free@high Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 7/30 passed in linked run darkrp-shop-npcnet-bounded-chunk-transferspatial-maintained-setfile-data-root-confinementnet-compact-targeted-updateperf-pairs-ipairs-forperf-seq-tablesperf-string-table-concatperf-surface-textperf-table-hasvalueprediction-one-shot-effectspawn-preventive-rate-limittransfer-lifecycle-cleanupnet-bounded-decompressionnet-secure-entity-actionnetworkvar-entity-stateperf-angle-zerorealm-shared-authoritysqlite-batched-transactionsqlite-typed-writetimer-delayed-validityhudpaint-hot-pathnet-per-player-budgetents-iterator-readonlyhook-object-lifecycleperf-local-colorperf-localplayer-cacheperf-player-cache-hooksperf-ply-table-indexperf-surface-drawrect
16	OpenCode	north-mini-code-free	Low	37.5%	50.0%	20.3%	6/1/9/14	22.5s
#16 north-mini-code-free Low OpenCode Pass rate37.5% 6 pass · 16 scored Coverage50.0% 16 / 32 evidence-bearing Verified score20.3% 95% interval — · 1 run(s) pass@k37.5% 6/32 fixtures ever solved Outcome mix Pass6 Partial1 Wrong9 Invalid output14 Harness2 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 Attempts32 Scored16 Passed6 Fixtures solved6/32 Run facts Avg time 22.5s Repeat Single-shot Evidence share 16/32 (50.0%) Verified runs 1 95% interval — Raw label opencode/north-mini-code-free@low Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 6/32 passed in linked run net-bounded-chunk-transfersqlite-typed-writefile-data-root-confinementhook-object-lifecyclehudpaint-hot-pathnet-compact-targeted-updatenet-per-player-budgetperf-ents-iteratorperf-for-vs-whileperf-hudpaint-cacheperf-seq-tablesperf-string-table-concatperf-table-count-hashprediction-one-shot-effectspatial-maintained-setspawn-preventive-rate-limitdarkrp-shop-npcents-iterator-readonlynet-bounded-decompressionnet-secure-entity-actionnetworkvar-entity-stateplayer-iteratorrealm-shared-authoritysqlite-batched-transactiontransfer-lifecycle-cleanupperf-looking-atperf-msg-vs-printperf-pairs-ipairs-forperf-plain-lua-vs-fnperf-surface-drawrectperf-table-hasvaluetimer-delayed-validity
17	OpenCode	north-mini-code-free	Medium	12.5%	64.0%	12.0%	2/2/12/9	23.6s
#17 north-mini-code-free Medium OpenCode Pass rate12.5% 2 pass · 16 scored Coverage64.0% 16 / 25 evidence-bearing Verified score12.0% 95% interval — · 1 run(s) pass@k12.5% 2/25 fixtures ever solved Outcome mix Pass2 Partial2 Wrong12 Invalid output9 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 Attempts25 Scored16 Passed2 Fixtures solved2/25 Run facts Avg time 23.6s Repeat Single-shot Evidence share 16/25 (64.0%) Verified runs 1 95% interval — Raw label opencode/north-mini-code-free@medium Run id c56989a5-0aa1-4fb5-be76-721322569d56 Latest run fixtures 2/25 passed in linked run hook-object-lifecyclenet-bounded-decompressionnet-compact-targeted-updateperf-ents-iteratorperf-seq-tablesperf-string-table-concatperf-table-hasvalueprediction-one-shot-effectspatial-maintained-setdarkrp-shop-npcents-iterator-readonlyhudpaint-hot-pathnet-bounded-chunk-transfernet-secure-entity-actionnetworkvar-entity-stateperf-meta-vs-argrealm-shared-authorityspawn-preventive-rate-limitsqlite-batched-transactionsqlite-typed-writetransfer-lifecycle-cleanupfile-data-root-confinementnet-per-player-budgetperf-pairs-ipairs-fortimer-delayed-validity

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

gmod.darkrp-shop-npc.v1
gmod.ents-iterator-readonly.v1
gmod.ents-iterator.v1
gmod.file-data-root-confinement.v1
gmod.hook-add.v1
gmod.hook-object-lifecycle.v1
gmod.hudpaint-hot-path.v1
gmod.isvalid.v1
gmod.net-bounded-chunk-transfer.v1
gmod.net-bounded-decompression.v1
gmod.net-compact-targeted-update.v1
gmod.net-per-player-budget.v1
gmod.net-receivers.v1
gmod.net-secure-entity-action.v1
gmod.networkvar-entity-state.v1
gmod.perf.angle-zero.v1
gmod.perf.config-var.v1
gmod.perf.darkrpvar.v1
gmod.perf.disttosqr.v1
gmod.perf.ents-iterator.v1
gmod.perf.find-ents-near.v1
gmod.perf.find-players-near.v1
gmod.perf.for-vs-while.v1
gmod.perf.hook-once.v1
gmod.perf.hudpaint-cache.v1
gmod.perf.local-color.v1
gmod.perf.local-vs-global.v1
gmod.perf.localplayer-cache.v1
gmod.perf.looking-at.v1
gmod.perf.meta-vs-arg.v1
gmod.perf.msg-vs-print.v1
gmod.perf.pairs-ipairs-for.v1
gmod.perf.plain-lua-vs-fn.v1
gmod.perf.player-cache-hooks.v1
gmod.perf.ply-table-index.v1
gmod.perf.seq-tables.v1
gmod.perf.setdrawcolor-split.v1
gmod.perf.sqlite-vs-mysql.v1
gmod.perf.steamid-cache.v1
gmod.perf.string-table-concat.v1
gmod.perf.surface-drawrect.v1
gmod.perf.surface-text.v1
gmod.perf.table-count-hash.v1
gmod.perf.table-empty.v1
gmod.perf.table-hasvalue.v1
gmod.perf.table-literal.v1
gmod.perf.table-random.v1
gmod.perf.x-times-x.v1
gmod.player-iterator.v1
gmod.prediction-one-shot-effect.v1
gmod.realm-shared-authority.v1
gmod.spatial-maintained-set.v1
gmod.spawn-preventive-rate-limit.v1
gmod.sqlite-batched-transaction.v1
gmod.sqlite-typed-write.v1
gmod.timer-delayed-validity.v1
gmod.transfer-lifecycle-cleanup.v1

Downloads

Results

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Outcome mix

Attempt funnel

Run facts

Latest run fixtures

Fixtures