← Leaderboard
StepFun · Proprietary · tested 11 Oct 2026
Step-5 Preview High effort
Overall rank#10 of 25
Solves50/10043 first attempt · +7 retry
Rungs proven263/400235 first attempt · +28 retry
Cost to run$94.62$1.89 per solve
Tokens used778.6M95.0% from cache · 10.9M output
Time per lab15.7 minaverage of 128 attempts
Where it places
Each line is one metric across all 25 configurations. Right is always better; this model is the large dot.
Solveslabs solved, of 100
worsebetter
50#10 of 25
Rungs provenrungs proven, of 400
worsebetter
263#9 of 25
First-attempt solvessolved on the first attempt
worsebetter
43#10 of 25
Cost per solveUSD per solved lab
worsebetter
$1.89#9 of 15
By vulnerability class
How Step-5 Preview (High) did on each type of vulnerability. Shown as counts, because some classes have only a few labs.
- Authentication / authorisation bypass14 of 28 solved80 of 112 rungs
- Information disclosure9 of 12 solved39 of 48 rungs
- IDOR4 of 11 solved28 of 44 rungs
- SSRF0 of 10 solved8 of 40 rungs
- Business logic5 of 10 solved29 of 40 rungs
- Key or secret exposure8 of 8 solved32 of 32 rungs
- XSS5 of 6 solved21 of 24 rungs
- Token theft2 of 5 solved12 of 20 rungs
Fewer than 5 labs each: one lab changes these a lot, so read them with care.
- Supply chain2 of 4 solved8 of 16 rungs
- Remote code execution0 of 3 solved1 of 12 rungs
- SQL injection1 of 2 solved4 of 8 rungs
- File inclusion0 of 1 solved1 of 4 rungs
Provider performance
Measured per single agent against the provider endpoint used for the run.
- Response time
- 5.41 s median · 24.67 s p95
- Output speed
- 122.9 tok/s
- API success rate
- 99.73%
- Input context
- 57.4K median · 185.4K 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