Share your thoughts, 1 month free Claude Pro on usSee more

Video Question Answering on DoraVQA (test)

76.1Overall Accuracy

Gemini-3.0-Flash

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-3.0-Flash 2026.01		76.1	67.83	81.25	65.38	88.52
Gemini-3.0-Flash 2026.01		75.84	66.81	87.5	69.23	86.89
Qwen3-VL-8B + GRPO 2026.01		67.98	62.5	37.5	59.62	72.13
GPT-4V 2026.01		67.79	60.34	50	69.23	68.85
Gemini-2.5-Pro 2026.01		64.41	62.62	50	60.78	71.93
Qwen2-VL-7B + GRPO 2026.01		62.38	59.74	43.75	64.58	63.93
Qwen3-VL-8B 2026.01		58.08	52.16	43.75	41.18	65.57
InternVideo2.5-8B 2026.01		57.68	46.55	31.25	59.62	68.85
Qwen2-VL-7B 2026.01		56.74	53.02	43.75	51.92	60.66
LLaVA-Video-7B 2026.01		55.41	54.02	31.25	66	67.31
Qwen2-VL-2B + GRPO 2026.01		55.11	55.45	28.57	50	50.67
Qwen2-VL-2B 2026.01		41.36	38.77	37.5	38.46	45.9
Video-LLaVA-7B 2026.01		37.82	47.49	41.67	46.34	32.43