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

Video Question Answering on Video-MME v1.0 (test)

65.3Accuracy (Long)

GPT-4o

Updated 12d ago

Evaluation Results

Method	Links
GPT-4o 2026.05		65.3	-	70.3	71.9
LongVU 2026.05		60.6	-	59.5	-
MUSEG-7B 2025.05		53.6	71.9	63.7	63.1
Qwen2.5-VL-7B-Instruct 2025.05		52.6	71.8	62.7	62.4
LLaVA-OV-7B + MuKV 2026.05		52.1	-	57.9	61.2
Qwen3-VL-4B + MuKV 2026.05		51	-	61.8	63.6
LLaVA-OV-7B + StreamMem 2026.05		50.1	-	56.6	59.4
Qwen2.5-VL-3B + MuKV 2026.05		50	-	61	60.8
Qwen3-VL-2B + MuKV 2026.05		49.8	-	60.8	62.6
Dispider 2026.05		49.7	-	53.7	57.2
Qwen3-VL-4B 2026.05		49.5	-	60.2	62.5
Qwen2.5-VL-3B + StreamMem 2026.05		49.1	-	60.1	59.5
LLaVA-OV-7B + LiveVLM 2026.05		48.8	-	56.4	57.3
LLaVA-OV-0.5B + MuKV 2026.05		48.5	-	41.4	49.1
Qwen3-VL-2B 2026.05		48.5	-	59	61.5
LLaVA-OV-0.5B + ReKV 2026.05		47.9	-	40.1	48.2
Qwen2.5-VL-3B 2026.05		47.2	-	58	60.3
LongVA 2026.05		46.2	-	50.4	52.6
LLaVA-OV-7B 2026.05		46.2	-	54.7	56.9
LLaVA-OV-0.5B 2026.05		46.2	-	39.7	45.4
MovieChat+ 2026.05		33.4	-	-	38.2
Qwen2.5-VL-3B + InfiniPot-V 2026.05		-	-	-	59.3