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

Real-world Visual Understanding on RealWorldQA

81.4Accuracy

InternVL2.5-8B-GenRecal

Updated 23d ago

Evaluation Results

Method	Links
InternVL2.5-8B-GenRecal 2025.06		81.4
Qwen2.5-VL-72B + CARES 2025.10		79
InternVL2.5-8B-GenRecal 2025.06		78.8
InternVL2.5-78B 2025.06		78.7
Qwen2-VL-72B 2025.06		77.8
Qwen2.5-VL-72B 2025.10		77
SkillGraph 2026.04		76.4
SkillGraph 2026.04		76.1
InternVL2.5-8B-GenRecal 2025.06		75.5
GPT-4o (0513) 2025.06		75.4
SkillGraph 2026.04		75.4
Random 2026.04		75.3
Complete 2026.04		75.2
SkillGraph 2026.04		74.9
SkillGraph 2026.04		74.7
SkillGraph 2026.04		74.4
SkillGraph 2026.04		74.1
Layered 2026.04		73.9
Vanilla MoE 2026.04		73.72
Molmo-72B 2025.06		73.7
SkillGraph 2026.04		73.7
MACS (Ours) 2026.04		73.58
Linear 2026.04		73.4
SkillGraph 2026.04		73.4
SkillGraph 2026.04		73.4
Centralized 2026.04		73.2
MACS (w/o Expanded) 2026.04		73.15
Random 2026.04		73.1
Complete 2026.04		72.9
InternVL2.5-8B-GenRecal 2025.06		72.5
InternVL2-76B 2025.06		72.2
SkillGraph 2026.04		72.2
Layered 2026.04		72.2
LLaVA-OneVision-72B 2025.06		71.9
SkillGraph 2026.04		71.9
Linear 2026.04		71.6
Centralized 2026.04		71.5
DirectAnswer 2026.04		71.4
Qwen3-VL-Seg 2026.05		71.2
SkillGraph 2026.04		71.1
SkillGraph 2026.04		70.9
Qwen3-VL 2026.05		70.9
DirectAnswer 2026.04		70.7
SkillGraph 2026.04		70.6
SkillGraph 2026.04		70.5
Random 2026.04		70.4
SkillGraph 2026.04		70.3
Complete 2026.04		70.3
SkillGraph 2026.04		70.1
NVLM-72B 2025.06		69.9
SkillGraph 2026.04		69.8
Layered 2026.04		69.7
SkillGraph 2026.04		69.6
CAI-MoE (Expanded) 2026.04		69.46
Random 2026.04		69.4
Linear 2026.04		69.3
Complete 2026.04		69.2
Centralized 2026.04		69.2
DirectAnswer 2026.04		68.6
Layered 2026.04		68.5
Linear 2026.04		68.1
Vanilla MoE 2026.04		68.07
InternVL3-8B 2025.10		68
InternVL3-8B + CARES 2025.10		68
InternVL3-8B + CARES-AR 2025.10		68
Qwen3-VL 2026.05		68
MACS (Ours) 2026.04		67.96
Centralized 2026.04		67.9
CAI-MoE (Token Drop) 2026.04		67.89
MACS (w/o Expanded) 2026.04		67.63
Gemini-1.5-Pro 2025.06		67.5
DirectAnswer 2026.04		67.2
InternVL-3.5 2026.05		66.3
InternVL2-8B 2025.09		65.5
PTP 2025.09		65.3
CAI-MoE (Expanded) 2026.04		65.18
Vanilla MoE 2026.04		64.87
Top-k Routing 2025.11		64.77
MACS (Ours) 2026.04		64.69
MACS (w/o Expanded) 2026.04		64.12
CAI-MoE (Token Drop) 2026.04		63.42
MoDES 2025.11		63.1
DiEP 2025.11		62.56
Gemma3-27B 2025.12		62.5
MoDES 2025.11		62.48
MC-MoE 2025.11		62.13
MoDES 2025.11		61.73
NAEE 2025.11		61.47
Gemma3-4B + AuditDM 2025.12		61.4
Top-k Routing 2025.11		61.34
CAI-MoE (Expanded) 2026.04		61.05
GPT-4o 2025.10		61
GPT-4o + CARES 2025.10		61
MC-MoE 2025.11		60.28
Claude-3.5-Sonnet 2025.06		60.1
NAEE 2025.11		59.47
DiEP 2025.11		58.74
CAI-MoE (Token Drop) 2026.04		58.34
Gemma3-12B 2025.12		58.3
PTP 2025.09		57.5

Showing 100 of 110 rows