mirror of
https://github.com/langgenius/dify-official-plugins.git
synced 2026-07-21 17:45:23 -04:00
LLM can't get document when using qwen-long #413
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @camer100 on GitHub (Jul 1, 2025).
Self Checks
Dify version
1.5.0
Cloud or Self Hosted
Cloud
Steps to reproduce
✔️ Expected Behavior
❌ Actual Behavior
@dosubot[bot] commented on GitHub (Jul 1, 2025):
Hi @camer100! I'm Dosu and I’m helping the dify team.
This is a known limitation in Dify: LLM nodes (including qwen-long) can't process Word (DOCX) files directly. To make this work, you need to add a Document Extractor node to your workflow. The Document Extractor converts the DOCX file to text, and you then connect its output (usually called
text) to the LLM node's prompt. Without this step, the LLM will always say "no word file found" because it only receives text, not raw files. This is required for all LLMs in Dify, not just qwen-long, even if the LLM itself supports direct document input on other platforms. For setup details, see the file upload guide and Document Extractor node docs(source).
The "office website" you mentioned likely supports direct DOCX input, but Dify currently does not—this workflow step is required.
To reply, just mention @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
