Emotion prediction on FI-2 ID VSA, VECBench
88.6AccuracyGPT-5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-5Model Category=Proprietary MLLM2025.12 | 88.6 | 88.4 | |
| EmoCaliberModel Category=Emotion-Oriented MLLM2025.12 | 88.1 | 88.16 | |
| Qwen3-VLModel Category=Open-Source MLLM2025.12 | 87.1 | 87.04 | |
| InternVL3.5Model Category=Open-Source MLLM2025.12 | 85.6 | 85.3 | |
| EmoVITModel Category=Emotion-Oriented MLLM2025.12 | 85.5 | 85.6 | |
| Qwen2.5-VLModel Category=Open-Source MLLM2025.12 | 84.9 | 84.82 | |
| Emotion-LLaMAModel Category=Emotion-Oriented MLLM2025.12 | 56.4 | 63.15 |