langchain - 💡(How to fix) Fix [Ollama][qwen] OllamaLLM.invoke() returns empty string with qwen3.5:9b model [4 comments, 3 participants]

langchain2026-04-22 04:12:36

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

langchain-ai/langchain#36936•Fetched 2026-04-23 07:23:13

View on GitHub

Comments

Participants

Timeline

Reactions

Author

Participants

Timeline (top)

labeled ×5commented ×3issue_type_added ×1mentioned ×1

Problem Description: When using OllamaLLM from langchain-ollama to call the qwen3.5:9b model, the invoke() method returns an empty string.

Expected Behavior: Should return a non-empty Chinese response.

Actual Behavior: Returns an empty string.

Additional Information:

ollama run qwen3.5:9b "你好" works normally from the command line
Direct requests call to the Ollama API works normally
Upgrading/downgrading langchain-ollama versions (0.1.3 → 0.2.0 → 1.1.0) has no effect
Setting num_predict=512, streaming=False, client_kwargs={"timeout": 120} has no effect
Using ChatOllama instead of OllamaLLM also has no effect

Error Message

Error Message and Stack Trace (if applicable)

响应: ''(Empty string, no error message)

Root Cause

Problem Description: When using OllamaLLM from langchain-ollama to call the qwen3.5:9b model, the invoke() method returns an empty string.

Expected Behavior: Should return a non-empty Chinese response.

Actual Behavior: Returns an empty string.

Additional Information:

ollama run qwen3.5:9b "你好" works normally from the command line
Direct requests call to the Ollama API works normally
Upgrading/downgrading langchain-ollama versions (0.1.3 → 0.2.0 → 1.1.0) has no effect
Setting num_predict=512, streaming=False, client_kwargs={"timeout": 120} has no effect
Using ChatOllama instead of OllamaLLM also has no effect

Fix Action

Fix / Workaround

This is a bug, not a usage question.
I added a clear and descriptive title that summarizes this issue.
I used the GitHub search to find a similar question and didn't find it.
I am sure that this is a bug in LangChain rather than my code.
The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
This is not related to the langchain-community package.
I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Code Example

from langchain_ollama import OllamaLLM

llm = OllamaLLM(model="qwen3.5:9b", temperature=0.3)
response = llm.invoke("你好，请用一句话回答")
print(f"响应: '{response}'")

# Expected output: non-empty string
# Actual output: Response: '' (empty string)


# Comparison: Directly calling Ollama API works normally 

import requests

response = requests.post(
"http://localhost:11434/api/generate",
json = {"model": "qwen3.5:9b", "prompt": "你好", "stream": False} timeout=120
)
print(response.json().get("response", ""))
# Normal response in Chinese

---

响应: ''(Empty string, no error message)

RAW_BUFFERClick to expand / collapse

Submission checklist

This is a bug, not a usage question.
I added a clear and descriptive title that summarizes this issue.
I used the GitHub search to find a similar question and didn't find it.
I am sure that this is a bug in LangChain rather than my code.
The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).
This is not related to the langchain-community package.
I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Package (Required)

Related Issues / PRs

No response

Reproduction Steps / Example Code (Python)

from langchain_ollama import OllamaLLM

llm = OllamaLLM(model="qwen3.5:9b", temperature=0.3)
response = llm.invoke("你好，请用一句话回答")
print(f"响应: '{response}'")

# Expected output: non-empty string
# Actual output: Response: '' (empty string)


# Comparison: Directly calling Ollama API works normally 

import requests

response = requests.post(
"http://localhost:11434/api/generate",
json = {"model": "qwen3.5:9b", "prompt": "你好", "stream": False} timeout=120
)
print(response.json().get("response", ""))
# Normal response in Chinese

Error Message and Stack Trace (if applicable)

响应: ''(Empty string, no error message)

Description

Problem Description: When using OllamaLLM from langchain-ollama to call the qwen3.5:9b model, the invoke() method returns an empty string.

Expected Behavior: Should return a non-empty Chinese response.

Actual Behavior: Returns an empty string.

Additional Information:

ollama run qwen3.5:9b "你好" works normally from the command line
Direct requests call to the Ollama API works normally
Upgrading/downgrading langchain-ollama versions (0.1.3 → 0.2.0 → 1.1.0) has no effect
Setting num_predict=512, streaming=False, client_kwargs={"timeout": 120} has no effect
Using ChatOllama instead of OllamaLLM also has no effect

System Info

System Information

OS: Windows OS Version: 10.0.26200 Python Version: 3.9.6 (tags/v3.9.6:db3ff76, Jun 28 2021, 15:26:21) [MSC v.1929 64 bit (AMD64)]

Package Information

langchain_core: 0.3.84 langchain: 0.3.28 langchain_community: 0.3.31 langsmith: 0.4.37 langchain_ollama: 0.3.10 langchain_text_splitters: 0.3.11

Other Dependencies

ollama: 0.21.0

extent analysis

TL;DR

The issue with OllamaLLM from langchain-ollama returning an empty string when calling the qwen3.5:9b model can be fixed by investigating the differences in how the invoke() method and the direct requests call to the Ollama API handle the request, particularly focusing on headers, parameters, and encoding.

Guidance

Compare Request Details: Compare the request headers, parameters, and body sent by OllamaLLM's invoke() method with those sent by the direct requests call to identify any discrepancies that could cause the empty response.
Check Encoding and Decoding: Verify that the encoding and decoding of the prompt and response are correctly handled in both scenarios, as character encoding issues could lead to empty or malformed responses.
Inspect invoke() Method: Look into the implementation of the invoke() method in OllamaLLM to ensure it correctly constructs and sends the request to the Ollama API, paying attention to any potential issues with the model specification, temperature, or other parameters.
Test with Different Models: Attempt to use OllamaLLM with different models to see if the issue is specific to the qwen3.5:9b model or if it's a more general problem with the invoke() method.

Example

No specific code example can be provided without modifying the existing OllamaLLM class or the invoke() method. However, adding logging or debugging statements to the invoke() method to inspect the request being sent and the response being received could help identify the issue.

Notes

The problem might be related to how the langchain-ollama package interacts with the Ollama API, possibly due to differences in request construction or parameter handling. The fact that direct requests calls and command-line usage of ollama work correctly suggests the issue is isolated to the OllamaLLM implementation.

Recommendation

Apply a workaround by using the direct requests call to the Ollama API until the issue with OllamaLLM is resolved, as this approach has been shown to work correctly in the provided example.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #ssr #latency issue #model loading #dependency error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

langchain - 💡(How to fix) Fix [Ollama][qwen] OllamaLLM.invoke() returns empty string with qwen3.5:9b model [4 comments, 3 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Error Message and Stack Trace (if applicable)

Root Cause

Fix Action

Fix / Workaround

Code Example

Submission checklist

Package (Required)

Related Issues / PRs

Reproduction Steps / Example Code (Python)

Error Message and Stack Trace (if applicable)

Description

System Info

System Information

Package Information

Other Dependencies

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

TRENDING

langchain - 💡(How to fix) Fix [Ollama][qwen] OllamaLLM.invoke() returns empty string with qwen3.5:9b model [4 comments, 3 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Error Message and Stack Trace (if applicable)

Root Cause

Fix Action

Fix / Workaround

Code Example

Submission checklist

Package (Required)

Related Issues / PRs

Reproduction Steps / Example Code (Python)

Error Message and Stack Trace (if applicable)

Description

System Info

System Information

Package Information

Other Dependencies

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

RELATED_DISCOVERY

TRENDING