List your company

General Deep Research Tool Use on GAIA (success rate)

80.3Success RateGemini-3-Pro
20.08435.71751.3566.983Mar 10, 2026

Evaluation results

MethodLinks
Gemini-3-ProCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0380.3
GPT-OSS-120BCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0366
Claude-4-SonnetCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0363.7
DIVE-8B (RL)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0361.2
DeepSeek-V3.2-ExpCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0361
Gemini-2.5-ProCategory=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0360.2
Kimi-K2-0905Category=Frontier (≫8B), Temperature=1, Top-p=1, Number of runs=42026.0360
WebExplorer-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0350
DIVE-8B (SFT)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0349.3
EnvScaler-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0325.8
SWE-Dev-8BCategory=8B Baselines, Temperature=1, Top-p=1, Number of runs=42026.0323.2
Qwen3-8B (base)Category=Ours, Temperature=1, Top-p=1, Number of runs=42026.0322.4