Gemini code execution

by GoogleCode execution

Code execution is Google's Gemini tool for writing and running Python while it answers. Gemini runs the code, reads the result and, when it errors, can rewrite it up to five times, with libraries such as NumPy, pandas and Matplotlib already installed. Each run is capped at 30 seconds, and it combines with Google Search grounding. On Opper it works on the Gemini generateContent endpoint, passed through as Google defines it.

UseGemini code executionwithin

Add Gemini code execution on Opper

Set up with agents

Paste this into your coding agent (Claude Code, Cursor, Codex, and more) and it will set up and build with Opper for you.

Get your API key

Key info

Tool

codeExecution

Endpoint

generateContent

In your code

1curl "https://api.opper.ai/v3/compat/v1beta/models/gemini/gemini-4-argon:generateContent" \
2  -H "Authorization: Bearer $OPPER_API_KEY" \
3  -H "Content-Type: application/json" \
4  -d '{"contents": [{"role": "user", "parts": [{"text": "Sum the first 50 primes and show your working."}]}],
5       "tools": [{"codeExecution": {}}]}'

Use Gemini code execution with these models

Step-by-step setup for Gemini code execution on each of these models through Opper.

Gemini 4 Argon Gemini 3.8 Flash Gemini 3.7 Flash Gemini 3.6 Flash Gemini 3.5 Flash-Lite Gemini 3.5 Flash Gemini 3.1 Pro Preview Gemini 2.5 Pro See all 700+ models

More tools on Opper

OpenAI code interpreter

GPT models write and run Python in a sandbox, on the Responses endpoint.

Codecode_interpreter

Claude code execution

Claude runs code in Anthropic's sandbox, on Anthropic's own route.

Codecode_execution

Claude computer use

Claude drives a desktop through screenshots, clicks and typing in your agent loop.

Codecomputer_toolset