Welcome to the VisualOverload Leaderboard!

Below you will find the public leaderboard for the VisualOverload benchmark, which evaluates models on their ability to understand and reason about complex visual scenes. We seperate by models and 'special' inference techniques (e.g., special prompts, ICL, CoT etc.) to better understand the source of their performance.

The leaderboard ranks models based on their overall accuracy across a six tasks (activity recognition, attribute recognition, counting, OCR, reasoning, and global scene recognition). We provide an aggregate score (Total) as well as individual scores on three distinct splits per difficulty (Easy, Medium, Hard), and each task.


o3	No	74.0	69.8	36.7	61.0	75.1	94.7	99.9	80.2	19.8	69.5
o4 mini	No	70.0	76.5	38.3	62.7	67.8	93.7	100.0	80.6	17.2	69.1
Gemini 2.0 Flash	No	76.0	71.1	41.7	57.6	56.6	92.1	99.5	77.3	19.5	68.1
InternVL3 38B	No	76.7	78.5	35.4	45.8	69.8	92.2	99.7	81.8	7.2	67.6
InternVL3 78B	No	78.0	80.5	34.7	31.4	65.4	93.7	99.7	80.6	8.1	66.8
Horizon Alpha	No	57.3	74.5	35.6	48.3	63.9	93.2	99.7	76.8	10.7	65.7
Qwen2.5-VL 72B	No	67.3	74.5	35.1	72.9	53.2	90.5	99.8	76.2	13.0	65.7
InternVL3 8B	No	66.0	67.8	32.2	42.4	59.0	93.4	100.0	75.1	8.4	63.9
Qwen2.5-VL 32B	No	60.0	70.5	30.8	60.2	61.5	90.3	99.9	72.5	12.1	63.5
LLaVA-OneVision 72B	No	66.0	69.8	30.9	39.0	57.1	91.8	99.8	73.8	4.4	62.7
InternVL3 14B	No	66.7	69.1	30.6	41.5	57.1	91.1	99.8	73.5	5.1	62.5
Qwen2.5-VL 7B	No	63.3	69.1	34.9	55.9	49.8	85.3	99.8	71.2	9.3	61.5
LLaVA-OneVision 7B	No	60.7	57.7	28.4	29.7	54.1	88.2	99.6	68.0	4.7	58.3
DeepSeek VL2	No	65.3	63.8	25.9	46.6	58.5	81.8	99.9	66.7	4.0	57.7
Llama 4 Scout	No	58.7	65.8	31.1	37.3	62.0	78.8	99.4	64.0	14.0	57.5
InternVL3 2B	No	50.0	58.4	30.4	39.0	49.8	80.3	99.9	62.6	6.3	55.3
InternLM-XComposer2.5	No	48.0	51.7	22.7	35.6	45.9	87.3	99.5	61.8	8.8	54.3
Qwen2.5-VL 3B	No	60.7	61.7	25.9	49.2	43.9	77.5	98.9	61.9	4.9	54.1
InternLM-XComposer2-4KHD	No	50.7	53.7	25.4	31.4	42.4	83.6	99.2	61.1	6.7	53.4
Gemma 3 27B	No	51.3	46.3	18.1	39.8	50.7	86.3	99.6	57.7	8.8	53.2
VILA HD 1.5K	No	54.0	57.7	25.9	21.2	52.2	79.4	99.4	61.0	4.0	53.1
LFM2 VL 1.6B	No	49.3	55.7	25.2	28.0	44.4	79.5	99.3	60.1	5.1	51.9
InternVL3 1B	No	48.0	57.0	27.2	25.4	35.1	77.5	98.7	57.9	5.6	50.6
PaliGemma 2 10B	No	48.7	52.3	23.6	5.1	42.4	81.8	98.3	56.3	6.3	50.3
Gemma 3 12B	No	48.7	42.3	16.5	30.5	47.8	82.7	99.5	54.4	6.5	49.9
VILA HD 4K	No	54.0	48.3	22.5	11.0	49.3	74.5	99.0	55.8	4.0	48.5
LLaVA-NeXT 13B	No	44.0	43.6	17.0	6.8	41.5	75.8	99.0	46.9	3.7	45.1
Gemma 3n E4B	No	40.0	23.5	19.3	22.9	41.0	73.9	92.7	50.5	9.1	44.1
SmolVLM	No	42.7	41.6	17.2	28.0	32.2	67.3	96.3	47.5	4.2	42.0
LLaVA 1.5 13B	No	41.3	39.6	13.8	3.4	42.9	71.6	95.6	43.5	2.8	42.0
PaliGemma 2 28B	No	40.0	49.0	17.4	5.9	40.0	66.1	92.0	47.9	8.1	41.5
DeepSeek VL2 Tiny	No	54.7	47.7	22.5	35.6	37.1	54.2	88.8	45.8	4.4	41.2
LFM2 VL 450M	No	35.3	47.0	22.9	20.3	27.8	59.5	94.5	43.9	8.6	39.7
LLaVA-NeXT 7B	No	44.7	41.6	19.1	8.5	40.5	54.0	93.2	39.5	2.6	37.5
Gemma 3n E2B	No	32.0	26.2	15.0	19.5	35.6	53.2	83.2	39.9	8.6	33.9
LLaVA 1.5 7B	No	35.3	43.6	13.2	3.4	39.5	43.2	87.3	35.1	2.3	30.8
PaliGemma 2 3B	No	42.0	53.0	20.4	8.5	24.9	32.7	78.7	42.6	6.3	29.0