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

Counting on CountBench

93.8Accuracy

SAM3

Updated 3d ago

Evaluation Results

Method	Links
SAM3 2026.01		93.8
Qwen2.5-VL 2025.02		93.6
Qwen2.5-VL-72B 2025.12		93.6
Qwen3-VL (Thinking) 2026.01		92.46
GLM-4.6V 2026.01		92.06
Seed-1.5-VL (Thinking) 2026.01		91.85
Molmo 2025.02		91.2
CodeDance-7B 2025.12		91.2
GLM-4.6V Flash 2026.01		90.22
Claude-4-Sonnet 2026.02		90.2
Claude-3.5 Sonnet 2025.02		89.7
Gemini2.5-Pro 2026.02		89.7
Qwen3-VL Thinking 2026.01		88.85
STEP3-VL-10B (PaCoRe) 2026.01		88.8
STEP3-VL-10B 2026.01		88.75
STEP3-VL-10B (SeRe) 2026.01		88.75
Youtu-VL 2026.01		88.6
GPT-4o 2025.02		87.9
GPT-4o 2025.12		87.9
Qwen2.5-VL-32B 2025.12		87.8
Gemini-2.5 (Pro) 2026.01		87.78
Qwen2.5-VL 2026.02		86.7
GPT-4o 2026.02		85.7
Gemini 1.5-Pro 2025.02		85.5
Visual Para-Thinker 2026.02		85.4
Thyme-VL-7B 2025.12		84.8
GPT5-mini 2026.02		84.4
Sequential 2026.02		83.7
MiMo-VL RL-2508 2026.01		83.6
Majority voting@4 2026.02		83.2
Qwen2.5-VL 2026.02		83.1
Llava-OneVision-7B 2025.12		82.3
InternVL 3.5 2026.01		82.18
Deepeyes-7B 2025.12		80.4
InternVL3-8B 2025.12		80.3
InternVL-3.5 2026.01		79.4
Qwen3-VL 2026.01		78.4
Qwen2.5-VL-7B 2025.12		76.5
MiniCPM-V 4.5 2026.01		76.4
Visual Para-Thinker 2026.02		72.6
InternVL2.5 2025.02		72.1
Sequential 2026.02		70.4
Majority voting@4 2026.02		69.8
Qwen2.5-VL 2026.02		69.2
InternVL3.5 8B 2026.01		68.6
InternVL2.5-8B 2025.12		55.9
MiniCPM-V 4.5 2026.01		50.1
InternVL3.5 8B 2026.01		43.9
Singh et al. 2026.01		34.2
Zhang et al. 2026.01		30.7
CLIP-ViT-B/32 2026.01		29.2
Random 2026.01		10