Deep Search on XBench DeepSearch (Accuracy)
73AccuracyWeb2BigTable
Evaluation Results
| Method | Links | |
|---|---|---|
| Web2BigTableSystem Category=Web2BigTable (Ours), Backbone=GPT-5 mini, Tool=Gemini 3 Flash2026.04 | 73 | |
| Minimax-M2System Category=Foundation Models with Tools2026.04 | 72 | |
| MiroFlowSystem Category=Deep Research System, Backbone=GPT-52026.04 | 72 | |
| DeepSeek-V3.2System Category=Foundation Models with Tools2026.04 | 71 | |
| GLM-4.5System Category=Foundation Models with Tools2026.04 | 70 | |
| Web2BigTableSystem Category=Web2BigTable (Ours), Backbone=GPT-5 mini, Tool=GLM-4.5 (OpenRouter)2026.04 | 70 | |
| Kimi-ResearcherSystem Category=Deep Research System2026.04 | 69 | |
| Web2BigTableSystem Category=Web2BigTable (Ours), Backbone=GPT-5 mini, Tool=DeepSeek-V3.2 (OpenRouter)2026.04 | 69 | |
| Web2BigTableSystem Category=Web2BigTable (Ours), Backbone=GPT-5 mini, Tool=Minimax-M2 (OpenRouter)2026.04 | 67 | |
| Claude-4.5-SonnetSystem Category=Foundation Models with Tools2026.04 | 66 | |
| DeepSeek-V3.2System Category=Foundation Models with Tools, Interface=OpenRouter2026.04 | 65 | |
| Minimax-M2System Category=Foundation Models with Tools, Interface=OpenRouter2026.04 | 64 | |
| GLM-4.5System Category=Foundation Models with Tools, Interface=OpenRouter2026.04 | 64 | |
| DeepMiner-32B-RLSystem Category=Open-Source Agentic Model2026.04 | 62 | |
| Gemini-2.5-ProSystem Category=Foundation Models with Tools2026.04 | 56 | |
| WebShaper-32BSystem Category=Open-Source Agentic Model2026.04 | 54.6 | |
| OAgentsSystem Category=Deep Research System, Backbone=Claude-3-72026.04 | 54.5 | |
| WebSailor-32BSystem Category=Open-Source Agentic Model2026.04 | 53.3 | |
| AFM-32B-RLSystem Category=Open-Source Agentic Model2026.04 | 52 | |
| Gemini DeepResearchSystem Category=Deep Research System2026.04 | 50 | |
| WebDancer-QwQSystem Category=Open-Source Agentic Model2026.04 | 40 | |
| GPT-5 miniSystem Category=Foundation Models with Tools2026.04 | 35 | |
| Gemini 3 FlashSystem Category=Foundation Models with Tools2026.04 | 28 | |
| OpenAI DeepResearchSystem Category=Deep Research System2026.04 | 26.6 |