Files
MiniCPM5-1B-Reasoning-Agent…/README.md

18 lines
422 B
Markdown
Raw Normal View History

---
library_name: transformers
tags: []
---
# 📊 Benchmark Results
| Benchmark | Metric | Score |
|---|---|---|
| ARC-Challenge | acc_norm | 39.25% |
| GSM8K | exact_match | 37.45% |
| MMLU | acc | 53.75% |
*Evaluated using `lm-evaluation-harness`.*
Note: v2 of this model coming soon with stronger reasoning and agentic capabilities. Will fix the "identity confusion" Users may experience with this current model.