Image Captioning on ICBench short captions (test)
80.13FluencyChatGPT-5
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ChatGPT-5Evaluator=Human2026.04 | 80.13 | 88.55 | 70.78 | 1 | — | — | — | — | — | — | — | |
| Gemini-3.0-FlashEvaluator=Human2026.04 | 74.36 | 84.57 | 77.97 | 2 | — | — | — | — | — | — | — | |
| InternVL3.5Evaluator=Human2026.04 | 70.67 | 81.81 | 68.52 | 3 | — | — | — | — | — | — | — | |
| DeepSeekVL2Evaluator=Human2026.04 | 67.69 | 79.52 | 66.44 | 5 | — | — | — | — | — | — | — | |
| Qwen3-VLEvaluator=Human2026.04 | 65.04 | 77.42 | 73.6 | 4 | — | — | — | — | — | — | — | |
| ChatGPT-5Evaluator=ITIScore2026.04 | 63.37 | 62.32 | 59.84 | 1 | — | — | — | — | — | — | — | |
| MiniCPM-V2.6Evaluator=Human2026.04 | 62.42 | 71.13 | 64.46 | 6 | — | — | — | — | — | — | — | |
| Gemini-3.0-FlashEvaluator=ITIScore2026.04 | 60.94 | 60.53 | 58.72 | 2 | — | — | — | — | — | — | — | |
| InternVL3.5Evaluator=ITIScore2026.04 | 60.04 | 58.99 | 56.71 | 3 | — | — | — | — | — | — | — | |
| Ovis2.5Evaluator=Human2026.04 | 59.84 | 73.34 | 62.42 | 7 | — | — | — | — | — | — | — | |
| Qwen3-VLEvaluator=ITIScore2026.04 | 59.17 | 56.9 | 57.46 | 4 | — | — | — | — | — | — | — | |
| DeepSeekVL2Evaluator=ITIScore2026.04 | 58.25 | 58.7 | 55.49 | 5 | — | — | — | — | — | — | — | |
| MiniCPM-V2.6Evaluator=ITIScore2026.04 | 57.92 | 54.2 | 52.79 | 6 | — | — | — | — | — | — | — | |
| LLama3.2-VEvaluator=Human2026.04 | 56.91 | 75.37 | 60.07 | 8 | — | — | — | — | — | — | — | |
| Ovis2.5Evaluator=ITIScore2026.04 | 56.22 | 55.81 | 53.83 | 7 | — | — | — | — | — | — | — | |
| LLama3.2-VEvaluator=ITIScore2026.04 | 55.33 | 56.89 | 54.98 | 8 | — | — | — | — | — | — | — | |
| mPLUG-Owl3Evaluator=ITIScore2026.04 | 54.09 | 53.48 | 51.92 | 9 | — | — | — | — | — | — | — | |
| mPLUG-Owl3Evaluator=Human2026.04 | 53.46 | 68.33 | 56.91 | 9 | — | — | — | — | — | — | — | |
| LLaVA-1.5Evaluator=ITIScore2026.04 | 51.72 | 51.36 | 49.18 | 10 | — | — | — | — | — | — | — | |
| LLaVA-1.5Evaluator=Human2026.04 | 46.94 | 61.71 | 46.36 | 10 | — | — | — | — | — | — | — | |
| ITIScoreEvaluation Protocol=Spearman Rank Correlation (SRCC) with Human2026.04 | 0.983 | 1 | 0.933 | 1 | — | — | — | — | — | — | — | |
| GPT-5Evaluation Protocol=Spearman Rank Correlation (SRCC) with Human2026.04 | 0.917 | 0.9 | 0.917 | — | — | — | — | — | — | — | — | |
| FLEUREvaluation Protocol=Spearman Rank Correlation (SRCC) with Human2026.04 | 0.883 | 0.683 | 0.5 | — | — | — | — | — | — | — | — | |
| ChatGPT-5Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 52.23 | 51.37 | 46.8 | 46.13 | 42.59 | 42.36 | |
| CLIP-SMethod Category=Reference-free2026.04 | — | — | — | — | — | 35.42 | 35.31 | 19.81 | 18.73 | 7.91 | 6.87 | |
| DeepSeekVL2 (small)Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 39.74 | 39.44 | 32.83 | 32.7 | 33.34 | 33.35 | |
| FLEURMethod Category=Reference-free2026.04 | — | — | — | — | — | 36.18 | 35.77 | 22.69 | 21.56 | 6.13 | 5.03 | |
| Gemini-3.0-FlashMethod Category=Advanced MLLM2026.04 | — | — | — | — | — | 49.07 | 48.48 | 48.98 | 48.28 | 37.86 | 37.62 | |
| InternVL3Model Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 42.58 | 42.08 | 29.07 | 29.16 | 26.22 | 26.29 | |
| InternVL3.5Model Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 47.57 | 46.83 | 44.38 | 43.83 | 37.96 | 37.82 | |
| ITIScoreBase Model=DeepSeekVL2 (small), Method Category=Instruction-tuned MLLM2026.04 | — | — | — | — | — | 62.4 | 62.4 | 61.24 | 61.25 | 57.79 | 57.79 | |
| ITIScoreBase Model=Qwen3-VL (8B), Method Category=Instruction-tuned MLLM2026.04 | — | — | — | — | — | 64.04 | 64.04 | 59.19 | 59.2 | 60.54 | 60.54 | |
| ITIScoreBase Model=InternVL3.5 (8B), Method Category=Instruction-tuned MLLM2026.04 | — | — | — | — | — | 64.48 | 64.49 | 63.21 | 63.21 | 62.1 | 62.1 | |
| LLama3.2-VModel Parameters=11B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 40.66 | 40.27 | 39.47 | 39.24 | 36.06 | 36.01 | |
| LLaVA-1.5Model Parameters=7B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 25.99 | 26.04 | 21.06 | 21.45 | 21.21 | 21.55 | |
| MiniCPM-V2.6Model Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 28.53 | 27.08 | 24.42 | 23.14 | 24.66 | 23.46 | |
| mPLUG-Owl3Model Parameters=7B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 30.47 | 30.3 | 33.92 | 33.72 | 26.68 | 26.72 | |
| Ovis2.5Model Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 25.59 | 24.43 | 18.69 | 18.06 | 17.81 | 17.26 | |
| PAC-SMethod Category=Reference-free2026.04 | — | — | — | — | — | 40.47 | 39.32 | 21.42 | 20.82 | 10.23 | 6.68 | |
| Qwen2.5-VLModel Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 44.9 | 44.48 | 33.91 | 33.79 | 29.19 | 29.16 | |
| Qwen3-VLModel Parameters=8B, Method Category=Advanced MLLM2026.04 | — | — | — | — | — | 48.43 | 47.71 | 41.25 | 40.88 | 37.96 | 37.79 | |
| UMICMethod Category=Reference-free2026.04 | — | — | — | — | — | 32.17 | 31.94 | 18.43 | 17.21 | 6.88 | 5.67 |