List your company

True-False Question Answering on Image Implication True-False 1.0 (test)

74AccuracyMetaphorStar-32B
-2.9617.023756.98Feb 11, 2026

Evaluation results

MethodLinks
MetaphorStar-32BParameters=32B2026.0274
MetaphorStar-7BParameters=7B2026.0270
Doubao-1.5-thinking-vision-pro2026.0262
MetaphorStar-3BParameters=3B2026.0262
Gemini-3.0-pro2026.0258
QwenVL-2.5-32BParameters=32B2026.0256
Gemini-2.5-flash2026.0256
Gemini-2.5-flash-thinking2026.0254
Claude-4.0-Sonnet2026.0252
QwenVL-2.5-72BParameters=72B2026.0250
GPT-4o2026.0250
GLM-4V-plus2026.0242
o4-mini2026.0242
GPT-4.12026.0240
GLM-4.1V-8BParameters=8B2026.0238
Claude-3.5-Sonnet2026.0238
GPT-4o-mini2026.0236
InternVL3-78BParameters=78B2026.0236
Grok-32026.0236
Grok-3-reasoning2026.0236
QwenVL-2.5-7BParameters=7B2026.0228
QVQ-72BParameters=72B2026.0228
QwenVL-2.5-3BParameters=3B2026.0220
DeepSeek-VL22026.0220
LLaVA-1.5-7BParameters=7B2026.020