Skip to content

Benchmark

SWE-fficiency

Coding benchmarks compare models on programming tasks. Scores are percentages. Higher is better.

Models measured
1
Scores recorded
2
Sources
2
IQ weight
No weight published

Score distribution

34.8%
Mean
34.8%
Median
0%
Std dev
34.8%
10th percentile
34.8%
90th percentile
34.8%
Min

Leaderboard

#ModelScore
1MiniMax M3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim34.8%
2MiniMax M3Claude Code · 1C2G · 2h· Reported by MiniMax Research· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.34.8%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Data by LLM Stats