mirror of
https://github.com/langgenius/dify-official-plugins.git
synced 2026-07-21 17:45:23 -04:00
When the Qwen3 model is configured to support stream function calling within Dify's vLLM plugin, the Agent reports an error: "Unterminated string starting at: line 1 column 20 (char 19)". #380
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 @vinland100 on GitHub (Jun 18, 2025).
Self Checks
Dify version
1.4.3
Cloud or Self Hosted
Self Hosted (Docker)
Steps to reproduce
using vllm deployment qwen3-32b:
docker run -dit --name vllm-qwen3-32b --runtime nvidia --gpus all -v "$HOME/.cache/modelscope:/root/.cache/modelscope" --env "VLLM_USE_MODELSCOPE=true" -p 8001:8000 --ipc=host vllm/vllm-openai:latest --model Qwen/Qwen3-32B --enable-auto-tool-choice --tool-call-parser hermes --enable-reasoning --tensor-parallel-size 2 --gpu-memory-utilization 0.6deployment Dify v1.4.3 using docker, set vllm plugin in Dify, set stream function calling as support.
then you see this error.
✔️ Expected Behavior
work normally
❌ Actual Behavior
when set to not support stream function calling, it works good, but it looks weird, the reasoning content not present first, must to click it to present:

@vinland100 commented on GitHub (Jun 18, 2025):
This error still exists when using the OpenAI-API-compatible plugin.
@dosubot[bot] commented on GitHub (Aug 19, 2025):
Hi, @vinland100. I'm Dosu, and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.
Issue Summary:
Next Steps:
Thank you for your understanding and contribution!