Gemini 3.8 Flash (medium)

gemini-3-8-flash-medium · 28 samples · intelligence / raw

Scoring

Choose the quality benchmark and final score formula for this model.

Cost adjustment

Tune formulas that account for Cost per Task.

Score over time

56.056.356.656.957.209-03 00:0009-03 11:3009-03 23:0009-04 10:3009-04 22:00 Run #2012 · Score · X: 2026-09-03 00:00:25 · Y: 56.6Run #2013 · Score · X: 2026-09-03 01:00:25 · Y: 56.6Run #2014 · Score · X: 2026-09-03 02:00:26 · Y: 56.6Run #2015 · Score · X: 2026-09-03 03:00:24 · Y: 56.6Run #2016 · Score · X: 2026-09-03 04:00:24 · Y: 56.6Run #2017 · Score · X: 2026-09-04 00:00:27 · Y: 56.6Run #2018 · Score · X: 2026-09-04 01:00:14 · Y: 56.6Run #2019 · Score · X: 2026-09-04 02:00:13 · Y: 56.6Run #2020 · Score · X: 2026-09-04 03:00:14 · Y: 56.6Run #2021 · Score · X: 2026-09-04 04:00:13 · Y: 56.6Run #2022 · Score · X: 2026-09-04 05:00:13 · Y: 56.6Run #2023 · Score · X: 2026-09-04 06:00:13 · Y: 56.6Run #2024 · Score · X: 2026-09-04 07:00:13 · Y: 56.6Run #2025 · Score · X: 2026-09-04 08:00:15 · Y: 56.6Run #2026 · Score · X: 2026-09-04 09:00:14 · Y: 56.6Run #2027 · Score · X: 2026-09-04 10:00:14 · Y: 56.6Run #2028 · Score · X: 2026-09-04 11:00:13 · Y: 56.6Run #2029 · Score · X: 2026-09-04 12:00:15 · Y: 56.6Run #2030 · Score · X: 2026-09-04 13:00:14 · Y: 56.6Run #2031 · Score · X: 2026-09-04 14:00:15 · Y: 56.6Run #2032 · Score · X: 2026-09-04 15:00:14 · Y: 56.6Run #2033 · Score · X: 2026-09-04 16:00:13 · Y: 56.6Run #2034 · Score · X: 2026-09-04 17:00:14 · Y: 56.6Run #2035 · Score · X: 2026-09-04 18:00:13 · Y: 56.6Run #2036 · Score · X: 2026-09-04 19:00:14 · Y: 56.6Run #2037 · Score · X: 2026-09-04 20:00:13 · Y: 56.6Run #2038 · Score · X: 2026-09-04 21:00:14 · Y: 56.6Run #2039 · Score · X: 2026-09-04 22:00:13 · Y: 56.6

Cost per task over time

$0.400$0.405$0.410$0.415$0.42009-03 00:0009-03 11:3009-03 23:0009-04 10:3009-04 22:00 Run #2012 · Cost per task · X: 2026-09-03 00:00:25 · Y: $0.410Run #2013 · Cost per task · X: 2026-09-03 01:00:25 · Y: $0.410Run #2014 · Cost per task · X: 2026-09-03 02:00:26 · Y: $0.410Run #2015 · Cost per task · X: 2026-09-03 03:00:24 · Y: $0.410Run #2016 · Cost per task · X: 2026-09-03 04:00:24 · Y: $0.410Run #2017 · Cost per task · X: 2026-09-04 00:00:27 · Y: $0.410Run #2018 · Cost per task · X: 2026-09-04 01:00:14 · Y: $0.410Run #2019 · Cost per task · X: 2026-09-04 02:00:13 · Y: $0.410Run #2020 · Cost per task · X: 2026-09-04 03:00:14 · Y: $0.410Run #2021 · Cost per task · X: 2026-09-04 04:00:13 · Y: $0.410Run #2022 · Cost per task · X: 2026-09-04 05:00:13 · Y: $0.410Run #2023 · Cost per task · X: 2026-09-04 06:00:13 · Y: $0.410Run #2024 · Cost per task · X: 2026-09-04 07:00:13 · Y: $0.410Run #2025 · Cost per task · X: 2026-09-04 08:00:15 · Y: $0.410Run #2026 · Cost per task · X: 2026-09-04 09:00:14 · Y: $0.410Run #2027 · Cost per task · X: 2026-09-04 10:00:14 · Y: $0.410Run #2028 · Cost per task · X: 2026-09-04 11:00:13 · Y: $0.410Run #2029 · Cost per task · X: 2026-09-04 12:00:15 · Y: $0.410Run #2030 · Cost per task · X: 2026-09-04 13:00:14 · Y: $0.410Run #2031 · Cost per task · X: 2026-09-04 14:00:15 · Y: $0.410Run #2032 · Cost per task · X: 2026-09-04 15:00:14 · Y: $0.410Run #2033 · Cost per task · X: 2026-09-04 16:00:13 · Y: $0.410Run #2034 · Cost per task · X: 2026-09-04 17:00:14 · Y: $0.410Run #2035 · Cost per task · X: 2026-09-04 18:00:13 · Y: $0.410Run #2036 · Cost per task · X: 2026-09-04 19:00:14 · Y: $0.410Run #2037 · Cost per task · X: 2026-09-04 20:00:13 · Y: $0.410Run #2038 · Cost per task · X: 2026-09-04 21:00:14 · Y: $0.410Run #2039 · Cost per task · X: 2026-09-04 22:00:13 · Y: $0.410

Time per task over time

1.70m1.71m1.72m1.74m1.75m09-03 00:0009-03 11:3009-03 23:0009-04 10:3009-04 22:00 Run #2012 · Time per task · X: 2026-09-03 00:00:25 · Y: 1.74mRun #2013 · Time per task · X: 2026-09-03 01:00:25 · Y: 1.74mRun #2014 · Time per task · X: 2026-09-03 02:00:26 · Y: 1.74mRun #2015 · Time per task · X: 2026-09-03 03:00:24 · Y: 1.74mRun #2016 · Time per task · X: 2026-09-03 04:00:24 · Y: 1.74mRun #2017 · Time per task · X: 2026-09-04 00:00:27 · Y: 1.74mRun #2018 · Time per task · X: 2026-09-04 01:00:14 · Y: 1.74mRun #2019 · Time per task · X: 2026-09-04 02:00:13 · Y: 1.74mRun #2020 · Time per task · X: 2026-09-04 03:00:14 · Y: 1.74mRun #2021 · Time per task · X: 2026-09-04 04:00:13 · Y: 1.74mRun #2022 · Time per task · X: 2026-09-04 05:00:13 · Y: 1.74mRun #2023 · Time per task · X: 2026-09-04 06:00:13 · Y: 1.74mRun #2024 · Time per task · X: 2026-09-04 07:00:13 · Y: 1.74mRun #2025 · Time per task · X: 2026-09-04 08:00:15 · Y: 1.74mRun #2026 · Time per task · X: 2026-09-04 09:00:14 · Y: 1.74mRun #2027 · Time per task · X: 2026-09-04 10:00:14 · Y: 1.74mRun #2028 · Time per task · X: 2026-09-04 11:00:13 · Y: 1.74mRun #2029 · Time per task · X: 2026-09-04 12:00:15 · Y: 1.74mRun #2030 · Time per task · X: 2026-09-04 13:00:14 · Y: 1.74mRun #2031 · Time per task · X: 2026-09-04 14:00:15 · Y: 1.74mRun #2032 · Time per task · X: 2026-09-04 15:00:14 · Y: 1.74mRun #2033 · Time per task · X: 2026-09-04 16:00:13 · Y: 1.74mRun #2034 · Time per task · X: 2026-09-04 17:00:14 · Y: 1.74mRun #2035 · Time per task · X: 2026-09-04 18:00:13 · Y: 1.74mRun #2036 · Time per task · X: 2026-09-04 19:00:14 · Y: 1.71mRun #2037 · Time per task · X: 2026-09-04 20:00:13 · Y: 1.71mRun #2038 · Time per task · X: 2026-09-04 21:00:14 · Y: 1.71mRun #2039 · Time per task · X: 2026-09-04 22:00:13 · Y: 1.71m
Run?Fetched?Score?Quality?Cost/task?Time/task?Intel?Agent?MMMU%?
#2039 2026-09-04 22:00:13 56.6 56.6 $0.410 1.71m 56.6 - 84.2
#2038 2026-09-04 21:00:14 56.6 56.6 $0.410 1.71m 56.6 - 84.2
#2037 2026-09-04 20:00:13 56.6 56.6 $0.410 1.71m 56.6 - 84.2
#2036 2026-09-04 19:00:14 56.6 56.6 $0.410 1.71m 56.6 - 84.2
#2035 2026-09-04 18:00:13 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2034 2026-09-04 17:00:14 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2033 2026-09-04 16:00:13 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2032 2026-09-04 15:00:14 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2031 2026-09-04 14:00:15 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2030 2026-09-04 13:00:14 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2029 2026-09-04 12:00:15 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2028 2026-09-04 11:00:13 56.6 56.6 $0.410 1.74m 56.6 - 84.2
#2027 2026-09-04 10:00:14 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2026 2026-09-04 09:00:14 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2025 2026-09-04 08:00:15 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2024 2026-09-04 07:00:13 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2023 2026-09-04 06:00:13 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2022 2026-09-04 05:00:13 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2021 2026-09-04 04:00:13 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2020 2026-09-04 03:00:14 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2019 2026-09-04 02:00:13 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2018 2026-09-04 01:00:14 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2017 2026-09-04 00:00:27 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2016 2026-09-03 04:00:24 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2015 2026-09-03 03:00:24 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2014 2026-09-03 02:00:26 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2013 2026-09-03 01:00:25 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2
#2012 2026-09-03 00:00:25 56.6 56.6 $0.410 1.74m 56.6 48.5 84.2