Our new X account is live! Follow @wizwand_team for updates

Multimodal Understanding on MME-RealWorld Lite

67.3Overall Score

InternVL3-38B

Updated 3d ago

Evaluation Results

Method	Links
InternVL3-38B 2026.02		67.3	-	-
InternVL3-78B 2026.02		65.4	-	-
TikArt 2026.02		56.97	60.48	53.47
GPT-5 2026.02		56.22	56.94	55.47
Monet-7B 2026.02		55.5	58.34	51.07
SkiLa 2025.12		54.1	56.6	50.2
TikArt (w/o Observation) 2026.02		53.67	58.43	48.91
Gemini-2.5-pro 2026.02		53.48	52.1	50.73
TikArt (w/o Zoom Action) 2026.02		52.56	57.91	47.2
Gemini-2.5-flash 2026.02		52.07	49.7	47.32
GPT-4o 2025.12		52	54.4	48.3
Qwen3-VL-Instruct 2026.02		51.82	56.3	60.78
Deepeyes 2026.02		50.79	55.3	46.29
TikArt (w/o Segment Action) 2026.02		50.69	53.51	47.87
SkiLa-V 2025.12		50.2	51.7	47.7
LVR 7B* 2025.12		49.1	52.7	43.5
TikArt (w/o GRPO) 2026.02		48.6	50.65	46.55
LLaVA-OneVision 7B 2025.12		48.5	52.8	42.7
Vision-R1 7B* 2025.12		47.9	49.8	44.9
Direct SFT 2025.12		47.5	49.6	44.3
Qwen2.5-VL-32B 2026.02		45.96	-	-
GPT-4o 2026.02		45.73	49.81	53.89
TPRU-7B 2026.02		45.34	-	-
TPRU-32B 2026.02		44.92	-	-
Qwen2.5-VL-7B 2026.02		44.55	-	-
Qwen2.5-VL 7B 2025.12		44.3	48.6	37.7
GPT-5-nano 2026.02		42.82	46.11	49.4
Qwen2.5-VL-3B 2026.02		41.94	-	-
Qwen3-VL-Instruct 2026.02		41.64	49.02	34.27
Qwen3-VL-Thinking 2026.02		39.43	44.05	34.8
TPRU-3B 2026.02		39.19	-	-
Qwen3-VL-Instruct 2026.02		39.07	45.07	51.07
LLaVA-OneVision-1.5 2026.02		37.61	38.69	36.53
GPT-4o-mini 2025.12		37.4	38.8	35.2