docker-compose file error

[202252197]

ERROR An error occurred: 1 error(s) decoding:

      * 'services[wren-ai-service].ports[0]' expected a map, got 'string

[cyyeh]

@202252197 hi, have u tried launching Wren AI using release artifact?

Please check the official installation method here: https://docs.getwren.ai/oss/installation#using-wren-ai-launcher

[Ahmed-ao]

I used the link provided above to install wren ai and got the same bug! it occurs when you try to use your own local model instead of default gpt api.

[joanteixi]

I guess that you didn't create the .env file with all the parameters. Use this file as example: https://github.com/Canner/WrenAI/blob/main/docker/.env.example

[Ahmed-ao]

I used your .env file and it worked. However the is a problem with yaml config file; it doesn't connect to local model. I'm trying to use Deepseek:14b locally with Ollama. and I followed the instruction provided here https://docs.getwren.ai/oss/installation/custom_llm

Yaml Config

`type: llm
provider: litellm_llm
timeout: 600
models:

• model: openai/deepseek-r1:14b
  api_base: http://docker.host.internal:11434/v1
  api_key_name: LLM_OLLAMA_API_KEY
  kwargs:
  temperature: 0.8
  n: 1

  for better consistency of llm response

  seed: 0
  max_tokens: 4096
  response_format:
  type: text
• model: openai/deepseek-r1:14b
  api_base: http://docker.host.internal:11434/v1
  api_key_name: LLM_OLLAMA_API_KEY
  kwargs:
  temperature: 0.8
  n: 1

  for better consistency of llm response

  seed: 0
  max_tokens: 4096
  response_format:
  type: text

---

type: embedder
provider: litellm_embedder
models:

• model: openai/nomic-embed-text
  dimension: 768
  url: http://host.docker.internal:11434/v1
  timeout: 120

---

type: engine
provider: wren_ui
endpoint: http://localhost:3000

---

type: engine
provider: wren_ibis
endpoint: http://localhost:8000
source: bigquery
manifest: '' # base64 encoded string of the MDL
connection_info: '' # base64 encoded string of the connection info

---

type: engine
provider: wren_engine
endpoint: http://localhost:8080
manifest: ''

---

type: document_store
provider: qdrant
location: http://qdrant:6333
embedding_model_dim: 768
timeout: 120
recreate_index: true

---

type: pipeline
pipes:

• name: deepseek_pipline
  llm: litellm_llm.openai/deepseek-r1:14b
  embedder: litellm_embedder.openai/nomic-embed-text

  other pipeline configurations

• name: db_schema_indexing
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: historical_question_indexing
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: table_description_indexing
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: db_schema_retrieval
  llm: litellm_llm.openai/deepseek-r1:14b
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: historical_question_retrieval
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: sql_generation
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: sql_correction
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: followup_sql_generation
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: sql_summary
  llm: litellm_llm.openai/deepseek-r1:14b
• name: sql_answer
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: sql_breakdown
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: sql_expansion
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: sql_explanation
  llm: litellm_llm.openai/deepseek-r1:14b
• name: sql_regeneration
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: semantics_description
  llm: litellm_llm.openai/deepseek-r1:14b
• name: relationship_recommendation
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: question_recommendation
  llm: litellm_llm.openai/deepseek-r1:14b
• name: question_recommendation_db_schema_retrieval
  llm: litellm_llm.openai/deepseek-r1:14b
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: question_recommendation_sql_generation
  llm: litellm_llm.openai/deepseek-r1:14b
  engine: wren_ui
• name: chart_generation
  llm: litellm_llm.openai/deepseek-r1:14b
• name: chart_adjustment
  llm: litellm_llm.openai/deepseek-r1:14b
• name: intent_classification
  llm: litellm_llm.openai/deepseek-r1:14b
  embedder: litellm_embedder.openai/nomic-embed-text
  document_store: qdrant
• name: data_assistance
  llm: litellm_llm.openai/deepseek-r1:14b
• name: sql_pairs_indexing
  document_store: qdrant
  embedder: litellm_embedder.openai/nomic-embed-text
• name: sql_pairs_deletion
  document_store: qdrant
  embedder: litellm_embedder.openai/nomic-embed-text
• name: sql_pairs_retrieval
  document_store: qdrant
  embedder: litellm_embedder.openai/nomic-embed-text
  llm: litellm_llm.openai/deepseek-r1:14b
• name: preprocess_sql_data
  llm: litellm_llm.openai/deepseek-r1:14b
• name: sql_executor
  engine: wren_ui
• name: sql_question_generation
  llm: litellm_llm.openai/deepseek-r1:14b
• name: sql_generation_reasoning
  llm: litellm_llm.openai/deepseek-r1:14b

---

settings:
host: 127.0.0.1
port: 5556
column_indexing_batch_size: 50
table_retrieval_size: 10
table_column_retrieval_size: 100
query_cache_maxsize: 1000
allow_using_db_schemas_without_pruning: false
query_cache_ttl: 3600
langfuse_host: https://cloud.langfuse.com
langfuse_enable: true
logging_level: INFO
development: false
`

[paopa]

Hi @Ahmed-ao, here is my config.yaml for Ollama model and embedder. I think you can base it on it and modify it to your version.

models:
- api_base: http://host.docker.internal:11434/
  kwargs:
    n: 1
    temperature: 0
  model: ollama/phi4
provider: litellm_llm
timeout: 120
type: llm
---
models:
- api_base: http://host.docker.internal:11434/
  model: ollama/nomic-embed-text
  timeout: 120
provider: litellm_embedder
type: embedder