How to Intermediate · 3 min read

How to use Gemini 1.5 Pro 1M context window

Q: How to use Gemini 1.5 Pro 1M context window

Use the gemini-1.5-pro model with the OpenAI SDK v1 by specifying it in the model parameter and passing your input messages. The 1 million token context window allows you to send very large inputs or long conversations in a single request.

Quick answer

Use the gemini-1.5-pro model with the OpenAI SDK v1 by specifying it in the model parameter and passing your input messages. The 1 million token context window allows you to send very large inputs or long conversations in a single request.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the official OpenAI Python SDK and set your API key as an environment variable.

bash

pip install openai>=1.0

Step by step

Use the gemini-1.5-pro model with the OpenAI SDK v1 to send a chat completion request. The model supports a 1 million token context window, so you can pass very large inputs in the messages array.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Example large context input (truncated for demo)
large_context = """\
This is a very long conversation or document content repeated many times to simulate a large context window.
""" * 10000  # Repeat to increase tokens

messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": large_context}
]

response = client.chat.completions.create(
    model="gemini-1.5-pro",
    messages=messages
)

print(response.choices[0].message.content)

output

The assistant's response based on the large input context.

Common variations

You can use streaming to handle large outputs efficiently or switch to smaller context models like gemini-1.5-flash for shorter inputs. Async usage is also supported with the OpenAI Python SDK.

python

import asyncio
import os
from openai import OpenAI

async def main():
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

    messages = [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello with async!"}
    ]

    response = await client.chat.completions.acreate(
        model="gemini-1.5-pro",
        messages=messages
    )

    print(response.choices[0].message.content)

asyncio.run(main())

output

Hello! How can I assist you today?

Troubleshooting

If you receive a context length error, ensure your input tokens do not exceed 1 million tokens.
For performance issues, consider chunking your input or using streaming responses.
Verify your API key is set correctly in os.environ["OPENAI_API_KEY"].

✅

Key Takeaways

Use gemini-1.5-pro with OpenAI SDK v1 for 1 million token context support.
Pass large inputs in the messages array to leverage the extended context window.
Async and streaming calls improve handling of large inputs and outputs.
Always set your API key securely via environment variables.
Monitor token usage to avoid exceeding the 1M token limit.

Verified 2026-04 · gemini-1.5-pro, gemini-1.5-flash

Verify ↗