← Leaderboard
Xiaomi · Open weights · tested 8 Nov 2026
MiMo V2.6 Pro High effort
Overall rank#6 of 10
Solves8/1004 first attempt · +4 retry
Rungs proven112/40079 first attempt · +33 retry
Cost to run$41.76$5.22 per solve
Tokens used740.4M89.8% from cache · 14.7M output
Time per lab24.4 minaverage of 194 attempts
Where it places
Each line is one metric across all 10 ranked configurations. Right is always better; this model is the large dot.
Solveslabs solved, of 100
worsebetter
8#6 of 10
Rungs provenrungs proven, of 400
worsebetter
112#6 of 10
First-attempt solvessolved on the first attempt
worsebetter
4#6 of 10
Cost per solveUSD per solved lab
worsebetter
$5.22#4 of 9
By vulnerability class
How MiMo V2.6 Pro (High) did on each type of vulnerability. Shown as counts: each class has 10 labs, so one lab is 10% of a class.
- Access control3 of 10 solved24 of 40 rungs
- Authentication & sessions1 of 10 solved17 of 40 rungs
- Injection: queries & data0 of 10 solved6 of 40 rungs
- Code execution1 of 10 solved11 of 40 rungs
- Server-side access1 of 10 solved14 of 40 rungs
- Browser-side1 of 10 solved8 of 40 rungs
- Infrastructure & edge1 of 10 solved10 of 40 rungs
- Business logic & races0 of 10 solved6 of 40 rungs
- Secrets, crypto & supply chain0 of 10 solved11 of 40 rungs
- AI agents0 of 10 solved5 of 40 rungs
Provider performance
Measured per single agent against the provider endpoint used for the run.
- Response time
- 14.21 s median · 38.94 s p95
- Output speed
- 59.3 tok/s
- API success rate
- 98.90%
- Input context
- 45.0K median · 111.1K p95
About this result
Every model runs in the same neutral harness, so results compare the models themselves.
- Harness
- BB Arena neutral harness
- Safety refusals
- None