OpenVLM MMT-Bench - CIM: leaderboard
The "CIM" 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 | Grok 2 (1212) | 85 | #496 |
| 2 | Claude 3.5 Sonnet | 81.7 | #350 |
| 3 | GPT-4.1 Nano | 76.7 | #731 |
| 4 | Gemini 1.5 Pro | 76.7 | #445 |
| 5 | InternVL3-78B | 73.3 | #352 |
| 6 | Qwen 2 VL 72B | 73.3 | #460 |
| 7 | Claude 3 Haiku | 73.3 | #803 |
| 8 | Claude 3.7 Sonnet | 71.7 | #249 |
| 9 | Gemini 1.5 Flash | 71.7 | #618 |
| 10 | Aquila-VL-2B | 70 | #1238 |
| 11 | Pixtral-12B | 68.3 | #808 |
| 12 | InternVL3-38B | 66.7 | #405 |
| 13 | InternVL3-14B | 65 | #505 |
| 14 | GPT-4.1 (2025-04-14) | 61.7 | #250 |
| 15 | Gemma 3 4B | 55 | #1089 |
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-cim · How It Works · Data refreshed daily, snapshot 2026-10-20.