Multipl-E MBPP
Multilingual
text
+
+
+
+
About
MultiPL-E MBPP extends the Mostly Basic Python Problems (MBPP) benchmark to multiple programming languages through the MultiPL-E framework. This variant evaluates models' ability to solve basic programming problems across diverse languages, testing fundamental coding skills, algorithm implementation, and programming language syntax understanding.
+
+
+
+
Evaluation Stats
Total Models3
Organizations1
Verified Results0
Self-Reported3
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
3 models
Top Score
65.7%
Average Score
60.0%
High Performers (80%+)
0Top Organizations
#1Meta
3 models
60.0%
+
+
+
+
Leaderboard
3 models ranked by performance on Multipl-E MBPP
| License | Links | ||||
|---|---|---|---|---|---|
| Jul 23, 2024 | Llama 3.1 Community License | 65.7% | |||
| Jul 23, 2024 | Llama 3.1 Community License | 62.0% | |||
| Jul 23, 2024 | Llama 3.1 Community License | 52.4% |
