> ## Content Index
> Fetch the complete content index at: https://www.testingcatalog.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# xAI to add file attachment support in DeepSearch mode for Grok 3 on X
- URL: https://www.testingcatalog.com/xai-to-add-file-attachment-support-in-deepsearch-mode-for-grok-3-on-x/
- Published: 2025-02-23T21:19:28.000Z
- Updated: 2025-02-23T21:46:19.000Z
- Author: Alexey Shabanov
- Tags: Grok News, Latest AI News

### Key Points

- xAI is preparing to enable file attachments in DeepSearch mode for Grok 3 on X.
- This will allow users to provide additional context or data for Grok 3's web research.
- DeepSearch mode lets Grok 3 browse the web and synthesize information.
- This feature (file attachments) is not publicly available yet.

### What’s Happening

xAI, founded by Elon Musk, has announced plans to enhance its AI model [Grok](https://www.testingcatalog.com/tag/grok/) by adding file attachment capabilities to its DeepSearch mode on the X platform. This means users will soon be able to upload files like images and PDFs when using DeepSearch mode, which currently allows Grok 3 to search the web and provide detailed answers based on real-time information.

![Grok](https://storage.ghost.io/c/2a/1b/2a1b1782-8506-4d7d-bf53-ad3fb52e2a0f/content/images/2025/02/Screenshot-2025-02-23-at-15.09.39-1.png)

### Why It Matters

This update could make [Grok 3](https://www.testingcatalog.com/grok-3-is-now-the-most-powerful-ai-model-and-you-can-try-it-for-free/) more versatile, especially for tasks like scientific research or complex problem-solving, by letting it consider both web data and user-provided files. It’s a step toward making AI tools more integrated and context-aware.

### Additional Detail: Grok 3 Already Supports Attachments in "Think" Mode

Interestingly, Grok 3 already allows file attachments for analysis in its "Think", but this new feature will specifically integrate them into DeepSearch, potentially enhancing how it uses this data in web-based research.