OpenVLM MMT-Bench - Color Contrast: leaderboard
The "Color Contrast" subtask of OpenVLM MMT-Bench. OpenCompass OpenVLM evaluation of MMT-Bench: multimodal tasks covering visual recognition, localization, OCR, counting, relation reasoning, visual math, and hallucination detection.
Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 207 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.7 Sonnet | 70 | #249 |
| 2 | InternVL3-8B | 60 | #620 |
| 3 | Pixtral-12B | 60 | #808 |
| 4 | Claude 3.5 Sonnet | 60 | #350 |
| 5 | Gemma 3 12B | 60 | #671 |
| 6 | Gemma 3 4B | 60 | #1089 |
| 7 | GPT-4.1 Nano | 55 | #731 |
| 8 | MiniCPM-V-2.6 | 55 | #836 |
| 9 | Grok 2 (1212) | 55 | #496 |
| 10 | InternVL3-38B | 50 | #405 |
| 11 | InternVL3-78B | 50 | #352 |
| 12 | InternVL2-8B | 50 | #817 |
| 13 | Gemma 3 27B | 50 | #607 |
| 14 | Aquila-VL-2B | 50 | #1238 |
| 15 | Claude 3 Haiku | 50 | #803 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=openvlm-mmt-bench-color-contrast · How It Works · Data refreshed daily, snapshot 2026-10-20.