Cinematographic Video Captioning on CineCap Bench 1.0 (test) benchmark leaderboard · SOTA2 Research

Cinematographic Video Captioning on CineCap Bench 1.0 (test)

54.38Camera Movement CmpCineCap

Evaluation Results

Method	Links
CineCap2026.06		54.38	60.5	76.51	77.44	79.68	82.91	90.06	88.35	66.17	66.09	72.51	69.54	72.38	74.8	73.57
Gemini-2.5-Pro2026.06		40.23	43.09	51.07	47.92	48.11	47.37	80.92	78.61	40.24	43.55	48.9	45.44	52.03	52.53	52.28
Qwen3-VL-8B2026.06		34.48	31.66	45.49	38.52	36.76	33.4	61.12	51.14	33.31	36.73	44.19	41.21	44.09	38.6	41.16
Gemini-3.1-Pro2026.06		29.22	40.5	54.34	60.52	43.01	43.59	79	80.6	19.98	22.76	54.78	58.01	46.31	60.48	52.45
Qwen3-VL-30B2026.06		29.22	30.31	54.26	49.47	36.12	36.39	44.67	43.53	30.19	33.16	34.52	34.32	39.33	43.42	41.27
Qwen2.5-VL-72B2026.06		28.42	32.23	53.19	52.83	31.92	31.11	65.65	60.58	35.71	37.93	28.7	30.18	41.02	43.39	42.17
InternVL-3-8B2026.06		18.8	25.21	42.7	40.32	33.66	34.89	27.17	26.41	27.85	31.54	29.27	31.57	30.69	36.81	33.47
LLaVa-OV-7B2026.06		8.94	13.51	28.21	25.98	19.5	20.04	11.66	10.86	16.14	17.98	19.79	21.1	19.3	31.4	23.91
LLaVa-NV-7B2026.06		8.11	12.89	56.68	34.91	23.23	23.71	30.37	25.44	24.31	29.91	20.23	22.02	28.05	30.14	29.06
Tarsier-7B2026.06		8.06	13.11	37.57	35.51	27.38	28.28	15.81	14.92	17.83	21.47	22.95	25.23	22.64	31.05	26.19