Batch Operations

Batch Operations

ZeroDB batch endpoints let you execute multiple operations in a single request — reducing round-trips and improving throughput for bulk ingestion, migration, and agent pipelines.

Batch Query

Execute up to 50 table queries in one call:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/batch/query \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "operations": [
      {
        "table": "agents",
        "filters": {"status": "active"},
        "limit": 10
      },
      {
        "table": "sessions",
        "filters": {"user_id": "usr-123"},
        "limit": 5
      }
    ],
    "parallel": true,
    "stop_on_error": false
  }'

Batch Memory Upsert

Insert or update up to 500 memory records:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/memory/batch/upsert \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "records": [
      {
        "content": "User prefers concise responses",
        "memory_type": "semantic",
        "importance": 0.8,
        "tags": ["preference", "communication"]
      },
      {
        "content": "Project uses FastAPI + SQLAlchemy",
        "memory_type": "semantic",
        "importance": 0.9,
        "tags": ["tech-stack"]
      }
    ]
  }'

Batch Memory Delete

Delete up to 500 memory records by ID:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/memory/batch/delete \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "ids": [
      "mem-abc123",
      "mem-def456",
      "mem-ghi789"
    ]
  }'

Batch Memory Update

Partial-update up to 500 memory records:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/memory/batch/update \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "updates": [
      {
        "id": "mem-abc123",
        "importance": 0.95,
        "tags": ["preference", "high-priority"]
      }
    ]
  }'

Batch Vector Upsert

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/vectors/upsert-batch \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "vectors": [
      {
        "id": "vec-001",
        "vector_embedding": [0.021, -0.043, 0.118],
        "namespace": "documents",
        "metadata": {"source": "upload", "doc_id": "doc-42"}
      }
    ]
  }'

Batch Event Publish

Publish up to 100 events in one request:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/events/batch \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "events": [
      {
        "topic": "agent.task.completed",
        "event_payload": {"task_id": "task-001", "duration_ms": 1200}
      },
      {
        "topic": "agent.task.completed",
        "event_payload": {"task_id": "task-002", "duration_ms": 850}
      }
    ]
  }'

Batch RLHF Interactions

Log up to 1,000 RLHF interactions at once:

curl -X POST https://api.ainative.studio/api/v1/projects/{project_id}/database/rlhf/interactions/batch \
  -H "x-api-key: $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "interactions": [
      {
        "prompt": "Explain async/await",
        "response": "Async/await is a syntax for non-blocking I/O...",
        "metadata": {"model": "qwen-coder-32b"}
      },
      {
        "prompt": "What is a decorator in Python?",
        "response": "A decorator wraps a function to extend its behavior...",
        "metadata": {"model": "qwen-coder-32b"}
      }
    ]
  }'

Python SDK

import requests

API_KEY = "your-api-key"
PROJECT_ID = "your-project-id"
BASE = f"https://api.ainative.studio/api/v1/projects/{PROJECT_ID}/database"
headers = {"x-api-key": API_KEY, "Content-Type": "application/json"}

## Batch upsert memory
r = requests.post(f"{BASE}/memory/batch/upsert", headers=headers, json={
    "records": [
        {"content": f"Fact #{i}", "memory_type": "semantic", "importance": 0.7}
        for i in range(100)
    ]
})
print(f"Upserted: {r.json()['inserted']} records")

## Batch publish events
r = requests.post(f"{BASE}/events/batch", headers=headers, json={
    "events": [
        {"topic": "user.action", "event_payload": {"action": "click", "item": f"item-{i}"}}
        for i in range(50)
    ]
})
print(f"Published: {r.json()['published']} events")

Limits

Operation	Max per batch
Batch query	50 operations
Memory upsert	500 records
Memory delete	500 IDs
Memory update	500 records
Vector upsert	No fixed limit (per vector store capacity)
Event publish	100 events
RLHF interactions	1,000 interactions

Endpoints

Method	Path	Description
POST	/database/batch/query	Execute batch table queries (parallel optional)
POST	/database/memory/batch/upsert	Batch insert/update memory records
POST	/database/memory/batch/delete	Batch delete memory by ID list
POST	/database/memory/batch/update	Batch partial-update memory records
POST	/database/vectors/upsert-batch	Batch upsert vectors
POST	/database/events/batch	Batch publish events
POST	/database/rlhf/interactions/batch	Batch log RLHF interactions