CongRong-v2.0: benchmark results
Provider: iFlytek. Released 2025-05-30. Access: API.
Aggregate Rating 1701 ± 29, rank #342 of 10916 rated models, from 148 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MM-Vet - Math | 99.6 | Score | 100 |
| OpenVLM MMBench V1.1 CN - Overall | 88.3 | Accuracy (%) | 100 |
| OpenVLM MMMU - Electronics | 93.3 | Accuracy (%) | 100 |
| OpenVLM AI2D - Rock Strata | 95.1 | Accuracy (%) | 99.8 |
| OpenVLM MMMU - Basic Medical Science | 86.7 | Accuracy (%) | 99.8 |
| OpenVLM MMMU - Computer Science | 86.7 | Accuracy (%) | 99.8 |
| MMBench V1.1 | 88.1 | Accuracy (%) | 99.6 |
| OpenVLM Leaderboard | 80.7 | Avg Score | 99.6 |
| OpenVLM MMBench V1.1 CN - Physical Relation | 77.3 | Accuracy (%) | 99.6 |
| OpenVLM MMBench V1.1 EN - AR | 92.7 | Accuracy (%) | 99.6 |
| OpenVLM MMBench V1.1 EN - Physical Relation | 77.3 | Accuracy (%) | 99.6 |
| OpenVLM MMStar - Logical Reasoning | 83.6 | Accuracy (%) | 99.6 |
Interactive version: theaggregate.ai/model?slug=congrong-v2-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.