364a34d fix: correct unpacking order of distances, documents, and metadatas in hybrid search query
📊 Changes
1 file changed (+1 additions, -1 deletions)
View changed files
📝backend/open_webui/retrieval/utils.py (+1 -1)
📄 Description
Changelog Entry
Description
Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
Fixed
Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
Additional Information
Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
The zip order is:
[(distances, metadatas, documents)], but the unpack order is
[distances, documents, metadatas] — these are mismatched.
After unpacking, what's actually in each variable is:
[distances] ← distances ✅
[documents]← metadatas ❌
[metadatas]← documents ❌
ifk<k_reranker:sorted_items=sorted(# Ordering here is different to how it is unpacked in the lines below, leading to mismatched values in the variables.zip(distances,metadatas,documents),key=lambdax:x[0],reverse=True)sorted_items=sorted_items[:k]ifsorted_items:distances,documents,metadatas=map(list,zip(*sorted_items))else:distances,documents,metadatas=[],[],[]
When the results are then processed in merge_and_sort_query_results, the isinstance(document, str) check always fails, leading to 0 results in the combined list.
defmerge_and_sort_query_results(query_results:list[dict],k:int)->dict:......fordistance,document,metadatainzip(distances,documents,metadatas):ifisinstance(document,str):# <--- the metadata value is accidentally getting unpacked into this variable.doc_hash=hashlib.sha256(document.encode()).hexdigest()# Compute a hash for uniquenessifdoc_hashnotincombined.keys():combined[doc_hash]=(distance,document,metadata)continue# if doc is new, no further comparison is needed# if doc is alredy in, but new distance is better, updateifdistance>combined[doc_hash][0]:combined[doc_hash]=(distance,document,metadata)
Screenshots or Videos
Before change:
The value in 'document' is a dict with all the document metadata values:
After change:
The value in 'document' is a string with the document chunk content:
Contributor License Agreement
By submitting this pull request, I confirm that I have read and fully agree to the Contributor License Agreement (CLA), and I am providing my contributions under its terms.
Note
Deleting the CLA section will lead to immediate closure of your PR and it will not be merged in.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.
## 📋 Pull Request Information
**Original PR:** https://github.com/open-webui/open-webui/pull/21562
**Author:** [@VasilyLebedev123](https://github.com/VasilyLebedev123)
**Created:** 2/18/2026
**Status:** ✅ Merged
**Merged:** 2/19/2026
**Merged by:** [@tjbck](https://github.com/tjbck)
**Base:** `dev` ← **Head:** `fix/hybrid-search-query-results-unpacking-order`
---
### 📝 Commits (1)
- [`364a34d`](https://github.com/open-webui/open-webui/commit/364a34d066265fac9010fdc18f5d281c20e9f040) fix: correct unpacking order of distances, documents, and metadatas in hybrid search query
### 📊 Changes
**1 file changed** (+1 additions, -1 deletions)
<details>
<summary>View changed files</summary>
📝 `backend/open_webui/retrieval/utils.py` (+1 -1)
</details>
### 📄 Description
<!--
⚠️ CRITICAL CHECKS FOR CONTRIBUTORS (READ, DON'T DELETE) ⚠️
1. Target the `dev` branch. PRs targeting `main` will be automatically closed.
2. Do NOT delete the CLA section at the bottom. It is required for the bot to accept your PR.
-->
# Changelog Entry
### Description
- Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
### Fixed
- Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
---
### Additional Information
Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
The zip order is:
[(distances, metadatas, documents)], but the unpack order is
[distances, documents, metadatas] — these are mismatched.
After unpacking, what's actually in each variable is:
[distances] ← distances ✅
[documents]← metadatas ❌
[metadatas]← documents ❌
```py
if k < k_reranker:
sorted_items = sorted(
# Ordering here is different to how it is unpacked in the lines below, leading to mismatched values in the variables.
zip(distances, metadatas, documents ), key=lambda x: x[0], reverse=True
)
sorted_items = sorted_items[:k]
if sorted_items:
distances, documents, metadatas = map(list, zip(*sorted_items))
else:
distances, documents, metadatas = [], [], []
```
When the results are then processed in merge_and_sort_query_results, the isinstance(document, str) check always fails, leading to 0 results in the combined list.
```py
def merge_and_sort_query_results(query_results: list[dict], k: int) -> dict:
......
for distance, document, metadata in zip(distances, documents, metadatas):
if isinstance(document, str): # <--- the metadata value is accidentally getting unpacked into this variable.
doc_hash = hashlib.sha256(
document.encode()
).hexdigest() # Compute a hash for uniqueness
if doc_hash not in combined.keys():
combined[doc_hash] = (distance, document, metadata)
continue # if doc is new, no further comparison is needed
# if doc is alredy in, but new distance is better, update
if distance > combined[doc_hash][0]:
combined[doc_hash] = (distance, document, metadata)
```
### Screenshots or Videos
Before change:
The value in 'document' is a dict with all the document metadata values:
<img width="780" height="445" alt="image" src="https://github.com/user-attachments/assets/a55a9ba9-a32e-4612-9180-e89905b41358" />
After change:
The value in 'document' is a string with the document chunk content:
<img width="842" height="392" alt="image" src="https://github.com/user-attachments/assets/b86e66b8-762e-4857-be27-4fc16023ffc4" />
### Contributor License Agreement
<!--
🚨 DO NOT DELETE THE TEXT BELOW 🚨
Keep the "Contributor License Agreement" confirmation text intact.
Deleting it will trigger the CLA-Bot to INVALIDATE your PR.
-->
By submitting this pull request, I confirm that I have read and fully agree to the [Contributor License Agreement (CLA)](https://github.com/open-webui/open-webui/blob/main/CONTRIBUTOR_LICENSE_AGREEMENT), and I am providing my contributions under its terms.
> [!NOTE]
> Deleting the CLA section will lead to immediate closure of your PR and it will not be merged in.
---
<sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
📋 Pull Request Information
Original PR: https://github.com/open-webui/open-webui/pull/21562
Author: @VasilyLebedev123
Created: 2/18/2026
Status: ✅ Merged
Merged: 2/19/2026
Merged by: @tjbck
Base:
dev← Head:fix/hybrid-search-query-results-unpacking-order📝 Commits (1)
364a34dfix: correct unpacking order of distances, documents, and metadatas in hybrid search query📊 Changes
1 file changed (+1 additions, -1 deletions)
View changed files
📝
backend/open_webui/retrieval/utils.py(+1 -1)📄 Description
Changelog Entry
Description
Fixed
Additional Information
Fix: Hybrid search returning 0 results due to a mismatch in query result unpacking order.
The zip order is:
[(distances, metadatas, documents)], but the unpack order is
[distances, documents, metadatas] — these are mismatched.
After unpacking, what's actually in each variable is:
[distances] ← distances ✅
[documents]← metadatas ❌
[metadatas]← documents ❌
When the results are then processed in merge_and_sort_query_results, the isinstance(document, str) check always fails, leading to 0 results in the combined list.
Screenshots or Videos
Before change:
The value in 'document' is a dict with all the document metadata values:
After change:
The value in 'document' is a string with the document chunk content:
Contributor License Agreement
By submitting this pull request, I confirm that I have read and fully agree to the Contributor License Agreement (CLA), and I am providing my contributions under its terms.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.