← Leaderboard
Xiaomi · Open weights · tested 8 Oct 2026
MiMo V2.6 Flash High effort
Overall rank#6 of 11
Solves49/10042 first attempt · +7 retry
Rungs proven260/400229 first attempt · +31 retry
Cost to run$3.43$0.07 per solve
Tokens used181.9M91.4% from cache · 5.6M output
Time per lab20.4 minaverage of 155 attempts
Where it places
Each line is one metric across all 11 configurations. Right is always better; this model is the large dot.
Solveslabs solved, of 100
worsebetter
49#6 of 11
Rungs provenrungs proven, of 400
worsebetter
260#5 of 11
First-attempt solvessolved on the first attempt
worsebetter
42#6 of 11
Cost per solveUSD per solved lab
worsebetter
$0.07#1 of 7
By vulnerability class
How MiMo V2.6 Flash (High) did on each type of vulnerability. Shown as counts, because some classes have only a few labs.
- Authentication / authorisation bypass16 of 28 solved83 of 112 rungs
- Information disclosure8 of 12 solved35 of 48 rungs
- IDOR2 of 11 solved24 of 44 rungs
- SSRF0 of 10 solved9 of 40 rungs
- Business logic6 of 10 solved31 of 40 rungs
- Key or secret exposure8 of 8 solved32 of 32 rungs
- XSS4 of 6 solved17 of 24 rungs
- Token theft3 of 5 solved17 of 20 rungs
Fewer than 5 labs each: one lab changes these a lot, so read them with care.
- Supply chain1 of 4 solved6 of 16 rungs
- Remote code execution0 of 3 solved1 of 12 rungs
- SQL injection1 of 2 solved5 of 8 rungs
- File inclusion0 of 1 solved0 of 4 rungs
Provider performance
Measured per single agent against the provider endpoint used for the run.
- Response time
- 25.48 s median · 67.79 s p95
- Output speed
- 31.2 tok/s
- API success rate
- 94.52%
- Input context
- 22.7K median · 79.2K 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