What you need to know about the Deep Research and Deep Research Max Update: - Can consults over 100...
TL;DR · AI 摘要
Google Deep Research 及 Max 版本支持多模态输入、百源检索与原生图表生成，Max 版单任务发起约 160 次搜索，性能在 DeepSearchQA 达 93.3%。
核心要点
- Deep Research Max 单任务可发起约 160 次搜索，显著提升信息覆盖广度。
- 支持 PDF、CSV、音视频等多模态输入，并能内联生成图表和信息图。
- 通过 MCP 协议可连接外部工具，扩展研究能力，适用于复杂分析场景。
- Can consults over 100 sources in one research task.
- Generates native charts and infographics inline.
- Accepts PDFs, CSVs, images, audio, and video inputs.
- Max version uses ~160 search queries per https://t.co/KuyspnpEJG" / X
Philipp Schmid on X: "What you need to know about the Deep Research and Deep Research Max Update: - Can consults over 100 sources in one research task. - Generates native charts and infographics inline. - Accepts PDFs, CSVs, images, audio, and video inputs. - Max version uses ~160 search queries per https://t.co/KuyspnpEJG" / X
Don’t miss what’s happening
Philipp Schmid 
What you need to know about the Deep Research and Deep Research Max Update: - Can consults over 100 sources in one research task. - Generates native charts and infographics inline. - Accepts PDFs, CSVs, images, audio, and video inputs. - Max version uses ~160 search queries per task. - 93.3% on DeepSearchQA (web research) - 85.9% on BrowseComp (finding hard facts) - Connect external tools via MCP (Model Context Protocol) Blog → https://blog.google/innovation-and-ai/models-and-research/gemini-models/next-generation-gemini-deep-research… Docs →http://ai.google.dev/gemini-api/doc s/deep-research… AI Studio → https://aistudio.google.com/prompts/new_ch at?model=deep-research-preview-04-2026…
·
1
16
10