code_interpreter

Python-focused sandbox for data analysis, visualization, and general code execution. Boots in ~60ms with a full scientific Python stack pre-installed.

Specs

Property	Value
Template ID	code_interpreter
Default Timeout	300s (5 min)
CPU	2 vCPU
Memory	2 GB
Disk	10 GB
Boot Time	~60ms
Internet	Enabled
Working Directory	/home/user

Python Environment

• Python 3.11 with pip
• Virtual environment pre-activated
• All libraries installed at sandbox build time (no install delay at runtime)

Pre-installed Libraries

Category	Libraries
Data	numpy, pandas, polars, scipy
Visualization	matplotlib, seaborn, plotly
ML	scikit-learn, xgboost
Web	requests, beautifulsoup4, httpx
Utilities	pillow, openpyxl, pyyaml, jinja2

TIP

You can install additional packages at runtime with pip install, but pre-installed libraries are available instantly with no startup cost.

Use Cases

• Data analysis and transformation (CSV, JSON, Excel)
• Chart and visualization generation
• Mathematical and statistical computations
• Machine learning model training and inference
• Web scraping and API data collection
• File format conversion

Create a Sandbox

POST/sandboxes

curl -X POST https://api.sandbase.ai/sandboxes \
  -H "Authorization: Bearer sk-sb-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "templateID": "code_interpreter",
    "timeout": 300
  }'

Response:

{
  "sandboxID": "sbx_01abc...",
  "templateID": "code_interpreter",
  "clientID": "SandBase",
  "status": "running",
  "startedAt": "2024-07-01T12:00:00Z",
  "endAt": "2024-07-01T12:05:00Z"
}

Run Code

Once the sandbox is running, execute Python code via the exec endpoint:

curl -X POST https://api.sandbase.ai/sandboxes/sbx_01abc.../processes \
  -H "Authorization: Bearer sk-sb-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "cmd": "python3",
    "args": ["-c", "import pandas as pd; print(pd.DataFrame({\"x\": [1,2,3]}).describe())"]
  }'

Full Example — Data Analysis Pipeline

import requests

BASE = "https://api.sandbase.ai"
HEADERS = {"Authorization": "Bearer sk-sb-YOUR_KEY"}

# 1. Create sandbox (~60ms)
sandbox = requests.post(f"{BASE}/sandboxes", headers=HEADERS, json={
    "templateID": "code_interpreter",
    "timeout": 300
}).json()

sandbox_id = sandbox["sandboxID"]

# 2. Run data analysis
code = """
import pandas as pd
import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

# Generate sample data
data = pd.DataFrame({
    'date': pd.date_range('2024-01-01', periods=90),
    'revenue': np.random.normal(1000, 200, 90).cumsum(),
    'users': np.random.poisson(50, 90)
})

# Analysis
print("=== Summary ===")
print(data.describe().to_string())
print(f"\\nTotal revenue: ${data['revenue'].iloc[-1]:,.0f}")
print(f"Avg daily users: {data['users'].mean():.1f}")

# Save chart
plt.figure(figsize=(10, 4))
plt.plot(data['date'], data['revenue'])
plt.title('Cumulative Revenue')
plt.savefig('/home/user/chart.png', dpi=100)
print("\\nChart saved to /home/user/chart.png")
"""

result = requests.post(
    f"{BASE}/sandboxes/{sandbox_id}/processes",
    headers=HEADERS,
    json={"cmd": "python3", "args": ["-c", code]}
).json()

print(result["stdout"])

# 3. Download the generated chart
chart = requests.get(
    f"{BASE}/sandboxes/{sandbox_id}/files?path=/home/user/chart.png",
    headers=HEADERS
)

# 4. Clean up
requests.delete(f"{BASE}/sandboxes/{sandbox_id}", headers=HEADERS)

Configuration Options

Option	Type	Default	Description
timeout	integer	300	Sandbox lifetime in seconds (max 3600)
metadata	object	{}	Custom key-value metadata
envVars	object	{}	Environment variables injected at boot

Custom timeout and metadata

{
  "templateID": "code_interpreter",
  "timeout": 900,
  "metadata": {
    "project": "quarterly-report",
    "user": "analyst-1"
  }
}

Environment variables

{
  "templateID": "code_interpreter",
  "timeout": 300,
  "envVars": {
    "DATABASE_URL": "postgresql://...",
    "API_TOKEN": "tok_..."
  }
}

Related

• Create Sandbox — Full create endpoint reference
• Execute Command — Run commands in a sandbox
• File Operations — Upload/download files
• Sandbox Lifecycle — States, timeouts, pause/resume
• All Templates — Overview of available templates