98.5
The large value shows the leading mean for this attempt. Rankings are calculated across repeated attempts rather than from this result alone.
- Attempt
- flash-2004
- Model
- google/gemini-3-flash-preview
- Questions asked
- 101
- Seed
- 2004
- Fingerprint
- 721f8778a8c63ffc497ba983
- When
- 2026-09-06T20:58:18.087748+00:00
- Duration
- 18.3 min
- Cost
- 3.60 USD billed
- Tokens
- 268.4k prompt · 1.75M completion · 2.02M total
moonshotai/kimi-k2
mean on this sampled test
· 0.171 USD billed
· 14.7k / 70.9k / 85.5k tok
· 10.2 s / question
91.6
qwen/qwen3.6-flash
mean on this sampled test
· 0.226 USD billed
· 15.5k / 198.1k / 213.6k tok
· 10.4 s / question
76.3
qwen/qwen3.7-flash
mean on this sampled test
· 0.028 USD billed
· 15.5k / 210.5k / 225.9k tok
· 19.7 s / question
75.6
qwen/qwen3.8-flash
mean on this sampled test
· 0.033 USD billed
· 18.9k / 64.1k / 83.1k tok
· 7.6 s / question
86.3
moonshotai/kimi-k2.5
mean on this sampled test
· 0.585 USD billed
· 14.6k / 247.5k / 262.1k tok
· 22.7 s / question
90.1
moonshotai/kimi-k2.6
mean on this sampled test
· 0.470 USD billed
· 14.6k / 142.1k / 156.7k tok
· 13.2 s / question
82.4
google/gemini-2.5-flash
mean on this sampled test
· 0.133 USD billed
· 13.8k / 51.4k / 65.2k tok
· 3.5 s / question
90.8
google/gemini-3.5-flash
mean on this sampled test
· 0.862 USD billed
· 13.8k / 93.4k / 107.3k tok
· 3.9 s / question
90.1
google/gemini-3.6-flash
mean on this sampled test
· 0.360 USD billed
· 13.8k / 93.3k / 107.1k tok
· 4.3 s / question
87.8
google/gemini-3.7-flash
mean on this sampled test
· 0.178 USD billed
· 13.8k / 44.6k / 58.4k tok
· 3.1 s / question
93.9
google/gemini-3.8-flash
mean on this sampled test
· 0.220 USD billed
· 13.8k / 56.0k / 69.8k tok
· 2.6 s / question
92.4
qwen/qwen3.5-flash-02-23
mean on this sampled test
· 0.062 USD billed
· 15.5k / 235.7k / 251.1k tok
· 12.4 s / question
84.7
deepseek/deepseek-v4-flash
mean on this sampled test
· 0.011 USD billed
· 14.7k / 36.2k / 50.8k tok
· 5.3 s / question
92.4
google/gemini-2.5-flash-lite
mean on this sampled test
· 0.023 USD billed
· 13.8k / 54.6k / 68.5k tok
· 1.7 s / question
87.0
google/gemini-3.1-flash-lite
mean on this sampled test
· 0.044 USD billed
· 13.8k / 27.1k / 40.9k tok
· 1.6 s / question
94.7
google/gemini-3.5-flash-lite
mean on this sampled test
· 0.080 USD billed
· 13.8k / 30.2k / 44.0k tok
· 1.4 s / question
91.6
google/gemini-3-flash-preview
mean on this sampled test
· 0.092 USD billed
· 13.8k / 28.4k / 42.2k tok
· 2.7 s / question
98.5
deepseek/deepseek-v4-flash-0731
mean on this sampled test
· 0.018 USD billed
· 20.2k / 70.0k / 90.1k tok
· 3.7 s / question
88.5