SUTRA with PydanticAI

SUTRA by TWO Platforms

SUTRA is a family of large multi-lingual language models (LMLMs) pioneered by Two Platforms. SUTRA’s dual-transformer approach extends the power of both MoE and Dense AI language model architectures, delivering cost-efficient multilingual capabilities for over 50+ languages. It powers scalable AI applications for conversation, search, and advanced reasoning, ensuring high-performance across diverse languages, domains and applications.

PydanticAI

PydanticAI is a Python agent framework designed to make it less painful to build production grade applications with Generative AI.

Get Your API Keys

Before you begin, make sure you have:

A SUTRA API key (Get yours at TWO AI's SUTRA API page)
Basic familiarity with Python and Jupyter notebooks

This notebook is designed to run in Google Colab, so no local Python installation is required.

SUTRA using PydanticAI

Install Requirements

# Install required packages
!pip install "pydantic-ai-slim[openai]" --quiet

Setup API Keys 🔑

import os
from google.colab import userdata

# Set the API key from Colab secrets
os.environ["SUTRA_API_KEY"] = userdata.get("SUTRA_API_KEY")

Initialize Sutra Model via PydanticAI

import os
import nest_asyncio
import asyncio
from pydantic_ai import Agent
from pydantic_ai.models.openai import OpenAIModel
from pydantic_ai.providers.openai import OpenAIProvider

# Required for running async in Colab
nest_asyncio.apply()

# ⚙️ Model Configuration
sutra_provider = OpenAIProvider(
    base_url="https://api.two.ai/v2",
    api_key=os.environ["SUTRA_API_KEY"]
)

sutra_v2 = OpenAIModel("sutra-v2", provider=sutra_provider)
sutra_r0 = OpenAIModel("sutra-r0", provider=sutra_provider)

v2_agent = Agent(sutra_v2)
r0_agent = Agent(sutra_r0)

Multilingual Content Generation

async def run_content_generation():
    print("🌐 Multilingual Content Generation\n")
    examples = {
        "Hindi": "Write a short story about a robot in Hindi",
        "Tamil": "Write a motivational speech for students in Tamil",
        "Japanese": "Write a haiku about spring in Japanese",
        "Arabic": "Write a children's story in Arabic",
        "French": "Write a paragraph about climate change in French"
    }
    for lang, prompt in examples.items():
        result = await v2_agent.run(prompt)
        print(f"\n[{lang}]\n{result.output}\n")

await run_content_generation()

Multilingual Translation

async def run_translation():
    print("🌐 Multilingual Translation\n")
    phrases = [
        "Knowledge is power",
        "The world is beautiful",
        "Unity in diversity"
    ]
    target_languages = ["Telugu", "Spanish", "Russian", "Chinese", "Swahili"]

    for phrase, lang in zip(phrases, target_languages):
        prompt = f"Translate this to {lang}: '{phrase}'"
        result = await v2_agent.run(prompt)
        print(f"\nTo {lang}:\n{result.output}")

await run_translation()

Reasoning Capabilities

async def run_reasoning():
    print("🧠 Logical and Mathematical Reasoning\n")

    # Logical reasoning in Greek
    logic_prompt = """
    Premise 1: All birds can fly.
    Premise 2: Penguins are birds.
    Conclusion: Penguins can fly.
    Is this argument valid? Explain in Greek.
    """
    result = await r0_agent.run(logic_prompt)
    print(f"\n[Logical Reasoning in Greek]\n{result.output}")

    # Math reasoning in German
    math_prompt = "Solve step by step and explain in German: If 3x + 6 = 21, what is x?"
    result = await r0_agent.run(math_prompt)
    print(f"\n[Math Reasoning in German]\n{result.output}")

await run_reasoning()

Code Generation in Multilingual Explanation

async def run_code_gen():
    print("💻 Code Generation with Explanation in Polish\n")
    prompt = "Write a Python function to check for prime number and explain in Polish"
    result = await v2_agent.run(prompt)
    print(result.output)

await run_code_gen()

SUTRA with PydanticAI

On this page