Syntra AI

All AI features route through OpenRouter, supporting 200+ models from OpenAI, Anthropic, Google, Meta, Mistral, and more.

Setup

1. •Set OPENROUTER_API_KEY in environment
2. •ai_create_config — define a model configuration
3. •Use the config ID for inference calls

Create a config

{
  "input_modality": "text",
  "output_modality": "text",
  "provider": "openrouter",
  "model_id": "anthropic/claude-sonnet-4-5-20250929",
  "system_prompt": "You are a helpful assistant."
}

AI model configs

• •ai_list_configs — list all configs
• •ai_create_config — create a new config (model, provider, modality, system prompt)
• •ai_get_config / ai_update_config / ai_delete_config — CRUD operations

Inference

Chat completion

{
  "config_id": "uuid-of-text-config",
  "messages": [
    { "role": "system", "content": "You are a coding assistant." },
    { "role": "user", "content": "Explain closures in JavaScript." }
  ],
  "temperature": 0.7,
  "max_tokens": 1000
}

Image generation

{
  "config_id": "uuid-of-image-config",
  "prompt": "A futuristic city at sunset, cyberpunk style",
  "n": 1,
  "size": "1024x1024"
}

Embeddings

{
  "config_id": "uuid-of-embedding-config",
  "input": "The quick brown fox jumps over the lazy dog"
}

Or batch: "input": ["text one", "text two", "text three"]

Vector search (RAG)

Syntra includes built-in vector storage powered by pgvector.

Workflow

1. •ai_create_collection — create a vector collection
2. •ai_upsert_documents — add documents (embeddings auto-generated)
3. •ai_vector_search — semantic search by text query

Collection management

• •ai_list_collections / ai_get_collection / ai_delete_collection
• •Each collection has: name, embedding_model, dimensions, distance_metric

Distance metrics

Document operations

• •ai_upsert_documents — insert or update documents with auto-embedding
• •ai_delete_documents — remove by IDs
• •ai_vector_search — search by text query with optional metadata filter
• •ai_vector_search_by_vector — search by raw embedding vector

Reference docs

• •For the complete RAG workflow: see vector-workflow.md