default

afi / afi · cli / proxy / Default profile / OpenAI · gpt 5.6 sol high / afi 0.30.0

P50 across 3 scored trials

3 recorded; malformed and unscored trials remain inspectable but do not participate in aggregate metrics

Recall55%

Model-judged precisionN/A

Total bill$0.2368

Duration2m 51s

Per-metric distributions

Exact Type-7 percentiles over complete scored samples; partial scored samples remain unavailable; non-scored trials are disclosed below

Recall 3/3 scored trials measured

P50 55% P95 59.5%

Model-judged precision 0/3 scored trials measured

N/A no qualified judge decided these findings, so only locality was measured

N/A · 0 measured

Total bill 3/3 scored trials measured

P50 $0.2368 P95 $0.3209

Duration 3/3 scored trials measured

P50 2m 51s P95 4m 35s

Metric	Availability	Measured	Min	P50	P95	Max
Findings `finding_count`	Measured	3/3	`11`	`11`	`11.9`	`12`
Defects found `found`	Measured	3/3	`11`	`11`	`11.9`	`12`
Defects missed `missed`	Measured	3/3	`8`	`9`	`9`	`9`
Unkeyed findings `unkeyed`	Measured	3/3	`0`	`0`	`0`	`0`
Intended findings `intended`	Measured	3/3	`0`	`0`	`0`	`0`
Recall `recall`	Measured	3/3	`55%`	`55%`	`59.5%`	`60%`
Model-judged precision `precision`	N/A no qualified judge decided these findings, so only locality was measured	0/3	`-`	`-`	`-`	`-`
F1 `f1`	N/A no qualified judge decided these findings, so only locality was measured	0/3	`-`	`-`	`-`	`-`
Tier 1 recall `tier_1`	Measured	3/3	`100%`	`100%`	`100%`	`100%`
Tier 2 recall `tier_2`	Measured	3/3	`42.9%`	`42.9%`	`55.7%`	`57.1%`
Tier 3 recall `tier_3`	Measured	3/3	`50%`	`50%`	`50%`	`50%`
Tier 4 recall `tier_4`	Measured	3/3	`50%`	`50%`	`50%`	`50%`
Security recall `category_security`	Measured	3/3	`50%`	`50%`	`50%`	`50%`
Defect recall `category_defect`	Measured	3/3	`61.5%`	`69.2%`	`76.2%`	`76.9%`
Maintainability recall `category_maintainability`	Measured	3/3	`0%`	`0%`	`0%`	`0%`
Performance recall `category_performance`	Measured	3/3	`33.3%`	`33.3%`	`63.3%`	`66.7%`
Median anchor distance `anchor_median`	Measured	3/3	`0`	`0`	`0`	`0`
Worst anchor distance `anchor_max`	Measured	3/3	`0`	`0`	`2.6999999999999997`	`3`
Refusals a judge overturned `anchor_missed`	Not recorded no judging pass has looked beyond the scored slack	0/3	`-`	`-`	`-`	`-`
Review bill `review_bill`	Measured	3/3	`$0.1998`	`$0.2368`	`$0.3209`	`$0.3303`
Judge bill `judge_bill`	N/A no qualified judge decided these findings, so only locality was measured	0/3	`-`	`-`	`-`	`-`
Total bill `total_bill`	Measured	3/3	`$0.1998`	`$0.2368`	`$0.3209`	`$0.3303`
Tokens `tokens`	Measured	3/3	`1436`	`1574`	`1689.2`	`1702`
Duration `seconds`	Measured	3/3	`2m 5s`	`2m 51s`	`4m 35s`	`4m 47s`
Carried findings `carried`	Not recorded the reviewer reported no reuse figure	0/3	`-`	`-`	`-`	`-`

Compatibility identity

Any change to these inputs creates another group

Reviewer: afi
Reviewer tool: afi · cli
Subject: proxy
Reviewed SHA: 9b51f95ef609a219e211e37b082cd2e6913190e0
Key fingerprint: bd331c0b144e
Settings fingerprint: 89cd13e3b4e0
Configuration ID: config-323a87a9eb99b1e3
Build: afi 0.30.0 / b0f313c0b59a66ecc7612396dc8db0ea5da13a7a
Harness: bench 1 / 348a7e367b8c52f1ce72828ed09c15b0e9fac38b dirty
Adapter: afi 1 / sha256:a1a935298956020ee6d767a756847abb4c00694c88c2eb80d454886ebc4acf8c
Build ID: build-13a6e5be87bc4a53
Cohort ID: cohort-fdab2adfb8a4bd39
Comparison ID: comparison-004ca4a8ec71f5cd

Trials

Select View evidence to inspect one run; the static table is paginated at 100 trials

								Details
afi afi · cli	default Default profile · OpenAI · gpt 5.6 sol high	proxy	31 Aug 2026, 15:54	55%	N/A	11	scored	View evidence
afi afi · cli	default Default profile · OpenAI · gpt 5.6 sol high	proxy	31 Aug 2026, 15:48	55%	N/A	11	scored	View evidence
afi afi · cli	default Default profile · OpenAI · gpt 5.6 sol high	proxy	31 Aug 2026, 15:46	60%	N/A	12	scored	View evidence

benchee benchee-dashboard-1