The Mac mini docker deployment project wrenai-wren-ai-service-1 reported an error: AttributeError: 'NoneType' object has no attribute 'get_generator'

[G337-BeiJing-Ratel]

Hi team,
Now i start wren and having the error below on "wren-ai-service-1" pod

2025-06-24 20:42:46 I0624 12:42:46.623 8 wren-ai-service:190] Using OpenAI API-compatible LLM: Qwen3-14B
2025-06-24 20:42:46 I0624 12:42:46.623 8 wren-ai-service:191] Using OpenAI API-compatible LLM model kwargs: {'n': 1, 'temperature': 0.6, 'top_p': 0.95, 'top_k': 20, 'response_format': {'type': 'text'}}
2025-06-24 20:42:48 ERROR: Traceback (most recent call last):
2025-06-24 20:42:48 File "/app/.venv/lib/python3.12/site-packages/starlette/routing.py", line 692, in lifespan
2025-06-24 20:42:48 async with self.lifespan_context(app) as maybe_state:
2025-06-24 20:42:48 File "/usr/local/lib/python3.12/contextlib.py", line 204, in aenter
2025-06-24 20:42:48 return await anext(self.gen)
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 File "/app/.venv/lib/python3.12/site-packages/fastapi/routing.py", line 133, in merged_lifespan
2025-06-24 20:42:48 async with original_context(app) as maybe_original_state:
2025-06-24 20:42:48 File "/usr/local/lib/python3.12/contextlib.py", line 204, in aenter
2025-06-24 20:42:48 return await anext(self.gen)
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 File "/app/.venv/lib/python3.12/site-packages/fastapi/routing.py", line 133, in merged_lifespan
2025-06-24 20:42:48 async with original_context(app) as maybe_original_state:
2025-06-24 20:42:48 File "/usr/local/lib/python3.12/contextlib.py", line 204, in aenter
2025-06-24 20:42:48 return await anext(self.gen)
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 File "/src/main.py", line 32, in lifespan
2025-06-24 20:42:48 app.state.service_container = create_service_container(pipe_components, settings)
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 File "/src/globals.py", line 53, in create_service_container
2025-06-24 20:42:48 "semantics_description": generation.SemanticsDescription(
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 File "/src/pipelines/generation/semantics_description.py", line 217, in init
2025-06-24 20:42:48 "generator": llm_provider.get_generator(
2025-06-24 20:42:48 ^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-06-24 20:42:48 AttributeError: 'NoneType' object has no attribute 'get_generator'
2025-06-24 20:42:48
2025-06-24 20:42:48 ERROR: Application startup failed. Exiting.
2025-06-24 20:54:41 INFO: Started server process [9]
2025-06-24 20:54:41 INFO: Waiting for application startup.

.env file:

`COMPOSE_PROJECT_NAME=wrenai
PLATFORM=linux/amd64

PROJECT_DIR=/Users/wangpeng/ai/wrenai/WrenAI

WREN_ENGINE_PORT=8080
WREN_ENGINE_SQL_PORT=7432
WREN_AI_SERVICE_PORT=5555
WREN_UI_PORT=3000
IBIS_SERVER_PORT=8000
WREN_UI_ENDPOINT=http://wren-ui:${WREN_UI_PORT}

QDRANT_HOST=qdrant
SHOULD_FORCE_DEPLOY=1

OPENAI_API_KEY=

LLM_PROVIDER=openai_llm
LLM_OPENAI_API_KEY=1234ddfg
LLM_OPENAI_API_BASE=http://xx.xx.xx.xx:9000/v1
GENERATION_MODEL=Qwen3-14B

EMBEDDER_PROVIDER=openai_like_embedder
EMBEDDING_MODEL=BAAI/bge-m3
EMBEDDING_MODEL_DIMENSION=1024
EMBEDDER_OPENAI_API_KEY=sk-xxxxxxxxxxxxxxx
EMBEDDER_OPENAI_API_BASE=https://api.siliconflow.cn/v1/embeddings

WREN_PRODUCT_VERSION=0.24.0
WREN_ENGINE_VERSION=0.16.4
WREN_AI_SERVICE_VERSION=0.24.0
IBIS_SERVER_VERSION=0.16.4
WREN_UI_VERSION=0.29.2
WREN_BOOTSTRAP_VERSION=0.1.5

USER_UUID=

POSTHOG_API_KEY=phc_nhF32aj4xHXOZb0oqr2cn4Oy9uiWzz6CCP4KZmRq9aE
POSTHOG_HOST=https://app.posthog.com
TELEMETRY_ENABLED=true

GENERATION_MODEL=gpt-4o-mini
LANGFUSE_SECRET_KEY=
LANGFUSE_PUBLIC_KEY=

HOST_PORT=3000
AI_SERVICE_FORWARD_PORT=5555

EXPERIMENTAL_ENGINE_RUST_VERSION=false
`

config.yaml file:
`type: llm
provider: openai_llm
models:

• api_base: http://2xx.xx9.xxx.xx:9000/v1
  model: Qwen3-14B
  alias: default
  timeout: 600
  kwargs:
  n: 1
  temperature: 0.6 # Recommended for thinking mode
  top_p: 0.95
  top_k: 20
  response_format:
  type: text

---

type: embedder
provider: openai_like_embedder
models:

• model: BAAI/bge-m3
  alias: default
  api_base: https://api.siliconflow.cn/v1/embeddings
  timeout: 120

---

type: engine
provider: wren_ui
endpoint: http://wren-ui:3000

---

type: engine
provider: wren_ibis
endpoint: http://ibis-server:8000

---

type: document_store
provider: qdrant
location: http://qdrant:6333
embedding_model_dim: 3072 # put your embedding model dimension here
timeout: 120
recreate_index: true

---

type: pipeline
pipes:

• name: db_schema_indexing
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: historical_question_indexing
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: table_description_indexing
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: db_schema_retrieval
  llm: openai_llm.Qwen3-14B
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: historical_question_retrieval
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: sql_generation
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
• name: sql_correction
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
  document_store: qdrant
• name: followup_sql_generation
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
• name: sql_answer
  llm: openai_llm.Qwen3-14B
• name: semantics_description
  llm: openai_llm.Qwen3-14B
• name: relationship_recommendation
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
• name: question_recommendation
  llm: openai_llm.Qwen3-14B
• name: question_recommendation_db_schema_retrieval
  llm: openai_llm.Qwen3-14B
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: question_recommendation_sql_generation
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
• name: chart_generation
  llm: openai_llm.Qwen3-14B
• name: chart_adjustment
  llm: openai_llm.Qwen3-14B
• name: intent_classification
  llm: openai_llm.Qwen3-14B
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: misleading_assistance
  llm: openai_llm.Qwen3-14B
• name: data_assistance
  llm: litellm_llm.qwen3-fast
• name: sql_pairs_indexing
  document_store: qdrant
  embedder: openai_like_embedder.BAAI/bge-m3
• name: sql_pairs_retrieval
  document_store: qdrant
  embedder: openai_like_embedder.BAAI/bge-m3
  llm: openai_llm.Qwen3-14B
• name: preprocess_sql_data
  llm: openai_llm.Qwen3-14B
• name: sql_executor
  engine: wren_ui
• name: user_guide_assistance
  llm: openai_llm.Qwen3-14B
• name: sql_question_generation
  llm: openai_llm.Qwen3-14B
• name: sql_generation_reasoning
  llm: litellm_llm.qwen3-thinking
• name: followup_sql_generation_reasoning
  llm: litellm_llm.qwen3-thinking
• name: sql_regeneration
  llm: openai_llm.Qwen3-14B
  engine: wren_ui
• name: instructions_indexing
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: instructions_retrieval
  embedder: openai_like_embedder.BAAI/bge-m3
  document_store: qdrant
• name: sql_functions_retrieval
  engine: wren_ibis
  document_store: qdrant
• name: project_meta_indexing
  document_store: qdrant
• name: sql_tables_extraction
  llm: openai_llm.Qwen3-14B

---

settings:
engine_timeout: 30
column_indexing_batch_size: 50
table_retrieval_size: 10
table_column_retrieval_size: 100
allow_intent_classification: true
allow_sql_generation_reasoning: true
allow_sql_functions_retrieval: true
enable_column_pruning: false
max_sql_correction_retries: 3
query_cache_maxsize: 1000
query_cache_ttl: 3600
langfuse_host: https://cloud.langfuse.com
langfuse_enable: true
logging_level: DEBUG
development: true
historical_question_retrieval_similarity_threshold: 0.9
sql_pairs_similarity_threshold: 0.7
sql_pairs_retrieval_max_size: 10
instructions_similarity_threshold: 0.7
instructions_top_k: 10
`

docker-compose.yaml :
`volumes:
data:

networks:
wren:
driver: bridge

services:
bootstrap:
image: ghcr.io/canner/wren-bootstrap:${WREN_BOOTSTRAP_VERSION}
restart: on-failure
platform: ${PLATFORM}
environment:
DATA_PATH: /app/data
volumes:
- ${PROJECT_DIR}/data:/app/data
command: /bin/sh /app/init.sh

wren-engine:
image: ghcr.io/canner/wren-engine:${WREN_ENGINE_VERSION}
restart: on-failure
platform: ${PLATFORM}
expose:
- ${WREN_ENGINE_PORT}
- ${WREN_ENGINE_SQL_PORT}
volumes:
- ${PROJECT_DIR}/data:/usr/src/app/etc
- ${PROJECT_DIR}/data:/app/data
networks:
- wren
depends_on:
- bootstrap

ibis-server:
image: ghcr.io/canner/wren-engine-ibis:${IBIS_SERVER_VERSION}
restart: on-failure
platform: ${PLATFORM}
expose:
- ${IBIS_SERVER_PORT}
environment:
WREN_ENGINE_ENDPOINT: http://wren-engine:${WREN_ENGINE_PORT}
networks:
- wren

wren-ai-service:
image: ghcr.io/canner/wren-ai-service:${WREN_AI_SERVICE_VERSION}
restart: on-failure
platform: ${PLATFORM}
expose:
- ${WREN_AI_SERVICE_PORT}
ports:
- ${AI_SERVICE_FORWARD_PORT}:${WREN_AI_SERVICE_PORT}
environment:
PYTHONUNBUFFERED: 1
CONFIG_PATH: /app/data/config.yaml
env_file:
- ${PROJECT_DIR}/docker/.env
volumes:
- ${PROJECT_DIR}/docker/config.yaml:/app/data/config.yaml
- ${PROJECT_DIR}/wren-ai-service/src:/src
networks:
- wren
depends_on:
- qdrant

qdrant:
image: qdrant/qdrant:v1.11.0
restart: on-failure
expose:
- 6333
- 6334
volumes:
- ${PROJECT_DIR}/data:/qdrant/storage
networks:
- wren

wren-ui:
image: ghcr.io/canner/wren-ui:${WREN_UI_VERSION}
restart: on-failure
platform: ${PLATFORM}
environment:
DB_TYPE: sqlite
SQLITE_FILE: /app/data/db.sqlite3
WREN_ENGINE_ENDPOINT: http://wren-engine:${WREN_ENGINE_PORT}
WREN_AI_ENDPOINT: http://wren-ai-service:${WREN_AI_SERVICE_PORT}
IBIS_SERVER_ENDPOINT: http://ibis-server:${IBIS_SERVER_PORT}

  GENERATION_MODEL: ${GENERATION_MODEL}

  WREN_ENGINE_PORT: ${WREN_ENGINE_PORT}
  WREN_AI_SERVICE_VERSION: ${WREN_AI_SERVICE_VERSION}
  WREN_UI_VERSION: ${WREN_UI_VERSION}
  WREN_ENGINE_VERSION: ${WREN_ENGINE_VERSION}
  USER_UUID: ${USER_UUID}
  POSTHOG_API_KEY: ${POSTHOG_API_KEY}
  POSTHOG_HOST: ${POSTHOG_HOST}
  TELEMETRY_ENABLED: ${TELEMETRY_ENABLED}
  
  NEXT_PUBLIC_USER_UUID: ${USER_UUID}
  NEXT_PUBLIC_POSTHOG_API_KEY: ${POSTHOG_API_KEY}
  NEXT_PUBLIC_POSTHOG_HOST: ${POSTHOG_HOST}
  NEXT_PUBLIC_TELEMETRY_ENABLED: ${TELEMETRY_ENABLED}
  EXPERIMENTAL_ENGINE_RUST_VERSION: ${EXPERIMENTAL_ENGINE_RUST_VERSION}
  
  WREN_PRODUCT_VERSION: ${WREN_PRODUCT_VERSION}
ports:
  - ${HOST_PORT}:3000
volumes:
  - ${PROJECT_DIR}/data:/app/data
networks:
  - wren
depends_on:
  - wren-ai-service
  - wren-engine:

`
The version I'm using is 0.24，

Could you please take a look for me? thank you

[cyyeh]

@G337-BeiJing-Ratel please upload your config.yaml

[G337-BeiJing-Ratel]

config.txt

The yaml file cannot be uploaded. So I changed the file suffix. The content inside is the same

[cyyeh]

@G337-BeiJing-Ratel you should put provider prefix before the model name, please check config.yaml examples for reference: https://github.com/Canner/WrenAI/tree/main/wren-ai-service/docs/config_examples

Also, since there is alias defined in your config.yaml, you could simply declare litellm_llm.default and litellm_embedder.default in pipeline definitions

[G337-BeiJing-Ratel]

hi，My qwen3 model was deployed on my local graphics card machine using vllm, and the embedder model uses SiliconFlow. Can I also refer to the configuration of config.qwen3.yaml。

[G337-BeiJing-Ratel]

My current approach is to customize llm and embedder. In - service/WrenAI/wren - ai/SRC/will/LLM openai have been added to the directory. The py, and in - service/SRC/WrenAI/wren - ai/will/embedder openai_like.py has been newly added under this directory. I don't know if my approach is correct.Below is my custom llm file. Please check it. thank you

openai.py.txt

openai_like.py.txt

[haomiemie]

请问解决了吗？能否参考一下您解决后的配置文件