MiniCPM-V-2.6-8B: benchmark results
Provider: OpenBMB.
Aggregate Rating 1544 ± 46, rank #1776 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| VNU-Bench - Multi-source Summary | 50.76 | Accuracy (%; question type T10, 4-option multiple choice) | 50 |
| VNU-Bench - Event Comparison | 46.51 | Accuracy (%; question type T2, 4-option multiple choice) | 41.7 |
| VNU-Bench - Temporal Development | 46.2 | Accuracy (%; question type T8, 4-option multiple choice) | 41.7 |
| PAGE Bench - Final Result Score | 8.12 | Final Result Score (0-100; 0.3 point and command match + 0.3 | 37.5 |
| VNU-Bench - Cross-source Evidence Integration | 48.54 | Accuracy (%; question type T6, 4-option multiple choice) | 33.3 |
| PAGE Bench | 6.48 | Overall Score (0-100; mean of the middle-process and final-r | 31.2 |
| VNU-Bench - Multimodal Presentation Comparison | 52.46 | Accuracy (%; question type T5, 4-option multiple choice) | 25 |
| PAGE Bench - Middle Process Score | 4.84 | Middle Process Score (0-100; 0.6 task success + 0.2 step suc | 18.8 |
| VNU-Bench | 46.39 | Accuracy (%; overall over the ten question types, 4-option m | 16.7 |
| VNU-Bench - Narrative Reconstruction | 43.39 | Accuracy (%; question type T9, 4-option multiple choice) | 16.7 |
| VNU-Bench - Visual Comparison | 48.93 | Accuracy (%; question type T3, 4-option multiple choice) | 16.7 |
| EmoBench-M | 46.5 | Average Score | 12 |
Interactive version: theaggregate.ai/model?slug=minicpm-v-2-6-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.