List your company
ResearchFormerly WizWand

Visual Question Answering on CktBench

83.058AccuracyGPT 5.1
9.7192828.7591447.79966.83886Jun 5, 2026
Updated 4mo ago

Evaluation results

MethodLinks
GPT 5.1Reasoning Protocol=SPARC2026.0683.058
SPARC2026.0683.058
CLAUDE S4Reasoning Protocol=SPARC2026.0681.991
GPT 4OReasoning Protocol=SPARC2026.0680.829
QWEN3 I.Reasoning Protocol=SPARC2026.0680.581
GLM 4.5VReasoning Protocol=SPARC2026.0672.116
MATH.2026.0654.378
MATHSENSEI2026.0654.378
CLAUDE S4Reasoning Protocol=COT2026.0651.659
GPT 5.1Reasoning Protocol=COT2026.0646.221
GPT 5.1Reasoning Protocol=CODE2026.0645.5
GPT 4OReasoning Protocol=COT2026.0643.151
QWEN3 I.Reasoning Protocol=COT2026.0641.667
NOTEMR2026.0638.116
NOTEMR2026.0638.116
GIFOMR2026.0636.633
GIFOMR2026.0636.633
GLM 4.5VReasoning Protocol=COT2026.0631.662
CLAUDE S4Reasoning Protocol=CODE2026.0631.5
QWEN3 I.Reasoning Protocol=CODE2026.0627.012
REFLEC.2026.0622.255
REFLECTIVA2026.0622.255
GPT 4OReasoning Protocol=CODE2026.0614.03
GLM 4.5VReasoning Protocol=CODE2026.0612.54