Docs
Search
Ctrl K
Models
Chat
Compare
Providers
Apps
Rankings
Sign Up
Sign Up
FrontierSWE Benchmark Leaderboard | Phaseo
FrontierSWE
FrontierSWE
AI benchmark results and model performance
Summary
▼
Summary
▼
Type: percentage
Code
View benchmark source
Recorded Results
1
Average Score
81.20%
Scores Over Time
Individual benchmark scores plotted by date.
Models Using This Benchmark
Organisation
Model
Reported
Top Score
Info
Self Reported
Source
Moonshot
Kimi K3
16 Jul 2026
81.20%
KimiCode harness; dominance score recomputed from raw scores with the official evaluation script, current July 16, 2026.
Yes
Source
Score Range
81.20% - 81.20%
Leading Model (lowest score)
81.20% - Kimi K3
