List your company

Multimodal Search on FVQA

84AccuracyGemini-3.1-Pro
34.80847.57960.3573.121May 8, 2026

Evaluation results

MethodLinks
Gemini-3.1-ProEvaluation Mode=Agentic Workflow2026.05841.3
HyperEyes-235BTraining Stage=RL, Evaluation Mode=Agentic Workflow2026.0581.41.5
HyperEyes-235BTraining Stage=SFT, Evaluation Mode=Agentic Workflow2026.0580.31.9
HyperEyes-30BTraining Stage=RL, Evaluation Mode=Agentic Workflow2026.0579.31.7
Kimi-K2.5Evaluation Mode=Agentic Workflow2026.0576.52.5
HyperEyes-30BTraining Stage=SFT, Evaluation Mode=Agentic Workflow2026.0576.12
Claude-Opus-4.6Evaluation Mode=Agentic Workflow2026.0574.51.3
VDREvaluation Mode=Multimodal Deep Search Agents2026.0574.212.7
Qwen3-VL-235BEvaluation Mode=Agentic Workflow2026.0570.21.7
WebWatcherEvaluation Mode=Multimodal Deep Search Agents2026.0564.34
Gemini-3.1-ProEvaluation Mode=Direct Answer2026.0562.7—
DeepEyes-V2Evaluation Mode=Multimodal Deep Search Agents2026.0560.62.8
Claude-Opus-4.6Evaluation Mode=Direct Answer2026.0560.1—
Kimi-K2.5Evaluation Mode=Direct Answer2026.0559.6—
MMSearch-R1Evaluation Mode=Multimodal Deep Search Agents2026.0558.41.3
Qwen3-VL-30BEvaluation Mode=Agentic Workflow2026.05582
Qwen3-VL-235BEvaluation Mode=Direct Answer2026.0544.2—
Qwen3-VL-30BEvaluation Mode=Direct Answer2026.0536.7—