List your company

Multimodal Research on In-house 1

52.5AccuracyGemini-3-Flash
7.7819.393142.61Apr 6, 2026

Evaluation results

MethodLinks
Gemini-3-FlashAgent Type=Direct Answer2026.0452.5
GPT-5.4Agent Type=Direct Answer2026.0445.1
MIAAgent Type=Memory-based Search Agent2026.0431.8
Gemini-2.5-ProAgent Type=Direct Answer2026.0430.8
Unsupervised MIAAgent Type=Memory-based Search Agent2026.0429.8
Qwen2.5-VL-32B+ReACTAgent Type=Search Agent2026.0428.8
GPT-4oAgent Type=Direct Answer2026.0425.6
MementoAgent Type=Memory-based Search Agent2026.0422.7
ExpeLAgent Type=Memory-based Search Agent2026.0419.7
Qwen2.5-VL-32BAgent Type=Direct Answer2026.0418.6
ReasoningBankAgent Type=Memory-based Search Agent2026.0418.6
No MemoryAgent Type=Memory-based Search Agent2026.0415.9
MMSearch-R1Agent Type=Search Agent2026.0413.6
RAGAgent Type=Memory-based Search Agent2026.0412.5
Mem0Agent Type=Memory-based Search Agent2026.0412.5
A-MemAgent Type=Memory-based Search Agent2026.0412.5
Qwen2.5-VL-7BAgent Type=Direct Answer2026.049.5
Qwen2.5-VL-7B+ReACTAgent Type=Search Agent2026.049.5