dify - 💡(How to fix) Fix [Cloud / Free tier] Knowledge Retrieval returns empty result for every query, across multiple embedding providers, file formats, and chunking configs

Self Checks

• I have read the Contributing Guide and Language Policy.
• This is only for bug report, if you would like to ask a question, please head to Discussions.
• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report, otherwise it will be closed.
• 【中文用户 & Non English User】请使用英语提交，否则会被关闭 ：）
• Please do not modify this template :) and fill in all the required fields.

Dify version

Version1.14.1

Cloud or Self Hosted

Cloud

Steps to reproduce

1. On Dify Cloud free tier, create a new Knowledge Base.
2. Configure the KB:

• Index Method: High Quality
• Embedding Model: any (tested with both Google gemini-embedding-001 and OpenAI text-embedding-3-small)
• Retrieval Setting: Hybrid Search, Weighted Score, Semantic 0.7 / Keyword 0.3, Top K 3 or 4, Score Threshold OFF

1. Upload a document (any file type tested — PDFs and plain .txt both reproduce).
2. Wait for the document to be processed. Status changes to Available. Chunk preview shows reasonable text content.
3. Open Retrieval Testing within the KB UI.
4. Enter any query that should hit content in the document (e.g. for a Critical Illness definitions PDF, try critical illness).
5. Click Test.

✔️ Expected Behavior

Retrieval Testing returns at least one relevant chunk from the uploaded document, with a relevance score. The Knowledge Retrieval node in a chatflow returns a non-empty result array.

❌ Actual Behavior

Retrieval Testing returns no results. No chunks displayed, no error message. Knowledge Retrieval node in a chatflow returns:

{ "result": [] }

with PROCESS DATA showing all-zero usage:

{ "usage": { "prompt_tokens": 0, "prompt_unit_price": "0.0", "prompt_price_unit": "0.0", "prompt_price": "0.0", "completion_tokens": 0, "completion_unit_price": "0.0", "completion_price_unit": "0.0", "completion_price": "0.0", "total_tokens": 0, "total_price": "0.0", "currency": "USD", "latency": 0, "time_to_first_token": null, "time_to_generate": null } }

No error indicators anywhere — document status is Available, KB settings page shows no errors, embedding model is recognized.

KB was deleted and recreated from scratch between provider switches to rule out stale state from prior embedding model.