SUTRA with FastAPI
This guide walks you through building a FastAPI application integrated with SUTRA models (V2 or R0) to create scalable, RESTful AI-powered endpoints. Leverage SUTRA’s multilingual and reasoning capabilities to build APIs for chat, reasoning, or custom AI workflows.
📦 Step 1: Install Dependencies
Install FastAPI, Uvicorn (ASGI server), and the OpenAI SDK:
# SUTRA models are OpenAI API compatible
!pip install -qU fastapi uvicorn openai🔐 Step 2: Authenticate with Your API Key
Obtain your SUTRA API key from https://developer.two.ai and store it securely (e.g., as an environment variable).
import os
SUTRA_API_KEY = os.getenv("SUTRA_API_KEY", "YOUR_SUTRA_API_KEY")⚙️ Step 3: Set Up FastAPI with SUTRA Client
Create a FastAPI application and initialize the SUTRA client using the OpenAI SDK.
from fastapi import FastAPI
from openai import OpenAI
from pydantic import BaseModel
app = FastAPI(title="SUTRA API with FastAPI")
# Initialize SUTRA client
client = OpenAI(
api_key=SUTRA_API_KEY,
base_url="https://api.two.ai/v2"
)
# Define request model
class ChatRequest(BaseModel):
prompt: str
model: str = "sutra-v2" # Default to sutra-v2, can use sutra-r0
max_tokens: int = 1024
temperature: float = 0.7💬 Step 4: Create a Basic Chat Endpoint
Add an endpoint to handle chat requests using sutra-v2 for multilingual responses.
@app.post("/chat")
async def chat(request: ChatRequest):
try:
response = client.chat.completions.create(
model=request.model,
messages=[
{"role": "system", "content": "You are a helpful AI that answers concisely in the requested language."},
{"role": "user", "content": request.prompt}
],
max_tokens=request.max_tokens,
temperature=request.temperature
)
return {"response": response.choices[0].message.content}
except Exception as e:
return {"error": str(e)}🧠 Step 5: Create a Reasoning Endpoint with SUTRA-R0
Add an endpoint for structured reasoning using sutra-r0, ideal for complex queries like legal or analytical tasks.
class ReasoningRequest(BaseModel):
prompt: str
max_tokens: int = 1024
temperature: float = 0.7
@app.post("/reason")
async def reason(request: ReasoningRequest):
try:
response = client.chat.completions.create(
model="sutra-r0",
messages=[
{"role": "system", "content": "You are a reasoning expert. Break down the query into logical steps before answering."},
{"role": "user", "content": request.prompt}
],
max_tokens=request.max_tokens,
temperature=request.temperature
)
return {"response": response.choices[0].message.content}
except Exception as e:
return {"error": str(e)}🚀 Step 6: Run the FastAPI Application
Run the application using Uvicorn:
uvicorn main:app --host 0.0.0.0 --port 8000Test the endpoints using curl or a tool like Postman:
- Chat Endpoint:
curl -X POST "http://localhost:8000/chat" -H "Content-Type: application/json" -d '{"prompt": "Explain AI benefits in Spanish", "model": "sutra-v2"}'- Reasoning Endpoint:
curl -X POST "http://localhost:8000/reason" -H "Content-Type: application/json" -d '{"prompt": "If a contract states \"Party A is liable unless Party B notifies within 7 days,\" who is liable if notice is given on day 8?"}'🛠 Troubleshooting
- Invalid API Key: Ensure
SUTRA_API_KEYis set correctly in your environment or code. Avoid hardcoding in production. - Model Not Found: Verify you’re using
sutra-v2orsutra-r0. SUTRA-V1 was deprecated on March 22, 2025. - Server Errors: Check that Uvicorn is running and the port (e.g., 8000) is not blocked.
- Rate Limits: If you hit rate limits, reduce request frequency or contact TWO AI at https://www.two.ai/support.
📎 Resources
Use FastAPI + SUTRA to build scalable, RESTful APIs with multilingual chat and advanced reasoning capabilities for production-ready AI applications.