Chat Directly with Your Documents and Files

Learn how to attach PDFs, spreadsheets, images, and Word docs to a conversation and get grounded answers, summaries, and analysis in WorkLLM.

Document Chat lets you upload files directly into a WorkLLM conversation and ask questions about them. Instead of copying and pasting content into a chat window, you attach the file and the AI reads it in full — answering questions, generating summaries, extracting data, and surfacing insights grounded in the actual document. You can attach multiple files to a single conversation and ask questions that draw across all of them.

Supported file types

WorkLLM supports a wide range of document and media formats:

Category	Supported File Types	Coming Soon	Coming Later
Documents & Office Files	.ppt,.pptx,.eml,.pdf,.doc,.docx,.xls,.xlsx,.csv,.txt,.rtf,.md	.epub,.mobi, .msg	.odt,.ods,.odp,.one,.enex,.key
Spreadsheets, Data, Analytics	.xml,.log,.jsonl/.ndjson,.xls,.xlsx,.csv,.json,.yaml,.yml,.sql		.parquet,.xlsm,.xlsb
Presentations & Design Assets	.ppt,.pptx	.svg	.odp,,.ai,.psd,.fig,.drawio,.dio,.vsdx, .key
Images	.png,.jpg,.jpeg,.webp	.gif,.tiff,.bmp,.heic,.heif
Audio & Video		Video Links (Youtube - 5 hours max video),.mp4,.mov,.webm,.avi,.mkv,.mp3,.ogg,.flac,.aac	Video LInks (Vimeo)
Code & Developer Files	.sh,.bash,.xml,.java,.c,.cpp,.cs,.php,.rb,.go,.rs,.ipynb,.py,.js,.ts,.tsx,.jsx,.html,.css,.json,.yaml,.yml,.sql		.R,.swift,.kt,.kts,.scala,.pl,.erl,.ex,.exs,.ini,.cfg,.env
Archives & Bundles		.zip, .tar,.tar.gz,.tgz,.7z,.rar
Knowledge Base & Web	.html,.md,.txt		SharePoint integration, Git repo integration

Image understanding (PNG, JPG, etc.) requires a multimodal model such as GPT-4o or Gemini 1.5 Pro. WorkLLM automatically selects a compatible model when you attach an image, or you can select one manually from the model picker.

Attaching files to a conversation

You can attach files using drag-and-drop or the attachment button.

Drag and drop

Open a chat and drag one or more files from your computer directly onto the chat window. A drop zone appears when you hover over it. Release to upload. Files are processed immediately and are ready to query once the upload completes.

Attachment button

1. Click the attachment icon

   Click the paperclip icon to the left of the message input field.

2. Select your files

   A file picker opens. Select one or more files from your computer. Hold Shift or Cmd/Ctrl to select multiple files at once.

3. Confirm and upload

   Click Open. Your files upload and appear as thumbnails above the message input. Once processing completes, you can start asking questions.

Asking questions about your documents

After files are attached, type your question in the input field and send it. The AI generates a response grounded in the content of your uploaded files.

Here are common ways to interact with attached documents:

Summarization

Ask for a summary of the document or a specific section:

• “Summarize this report in three bullet points.”
• “What are the key takeaways from section 4?”
• “Give me an executive summary of this proposal.”

Data extraction

Pull specific information from spreadsheets or structured documents:

• “What was the total revenue in Q3 across all regions?”
• “List all action items assigned to the engineering team.”
• “Extract all dates and deadlines mentioned in this contract.”

Analysis and comparison

Ask the AI to interpret, evaluate, or compare content:

• “What are the risks identified in this document?”
• “Compare the assumptions in document 1 and document 2.”
• “Does this contract include a liability cap clause? If so, what is it?”

Question and answer

Ask specific questions the way you would ask a colleague who had read the file:

• “What is the renewal date for this agreement?”
• “Who is listed as the primary contact?”
• “Does this policy cover remote employees?”

Using multiple files in one conversation

You can attach multiple files (up to 10) to a single prompt and ask questions that reference all of them. This is useful for comparing documents, consolidating information from multiple sources, or cross-referencing related materials.

To add more files after the conversation has started, click the paperclip icon again and select additional files. The AI treats all attached files as part of the same context.

When working with multiple files, name them clearly before uploading. WorkLLM displays file names in the thread so you can reference them directly in your prompts — for example, “According to the Q3 report, what was…”

File size and attachment limits

Limit	Value
Maximum file size per file	25 MB
Maximum files per prompt	10 files
Maximum total upload size per conversation	125 MB

Very large files (close to the 25 MB limit) may take longer to process before they are ready to query. If the AI seems unaware of content from a large file, wait for the processing indicator to clear and try again.

Can I share a document conversation with my team?

Yes. Document Chat conversations follow the same sharing rules as standard chats. Use the Share option to share with teammates. Attached files remain available to all thread participants.

Are uploaded files stored permanently?

Files are retained for the duration of the conversation and subject to your organization’s data retention settings. Contact your Admin to review your organization’s retention policy.

Can the AI edit or modify my uploaded files?

No. The AI reads and analyzes your files but does not modify them. Original files are never altered.

What if my document is password-protected?

Password-protected files cannot be processed. Remove the password protection before uploading, or use a version of the file without password restrictions.