Gemini 4 is Google DeepMind's new frontier model series. Gemini 4 Argon is the first release: a reasoning model with a 1M-token unified context and output window for long-horizon coding, knowledge work, and cybersecurity.Muapi exposes Gemini 4 Argon through one simple endpoint with text and image input, an optional system prompt, token-based billing at Google's official rates, and the same asynchronous submit-and-poll contract as every other model.
Gemini 4 Argon is available now via API
Try Gemini 4 Argon in the playground, or call it from your own code with a Muapi API key. More Gemini 4 models will be added to this page as they launch.
Google DeepMind's frontier reasoning model with a 1M-token unified context and output window. Accepts text and an optional image plus a system prompt, billed per token.
Gemini 4 Argon is the first model in Google DeepMind's Gemini 4 series. It is a frontier reasoning model designed for long-horizon coding, knowledge work, and cybersecurity analysis, with a 1M-token unified context and output window that lets you pass in whole repositories, contracts, or research corpora and receive long, structured answers in one call.
Muapi wraps it as a single gemini-4-argon endpoint. Send a prompt with an optional image_url and system_prompt, then poll the shared result endpoint. For faster, cheaper everyday chat, see Gemini 3.8 Flash.
Work across very large inputs and generate long outputs without chunking your data.
Multi-step planning and analysis for hard coding, research, and decision-support tasks.
Include an image alongside your prompt to ask about screenshots, diagrams, or documents.
Steer persona, tone, and output format with a dedicated system_prompt field.
Pay $2.00 per million input tokens and $10.00 per million output tokens, with no subscription.
Plan, implement, and refactor across large codebases using the full 1M-token context.
Analyze long reports, contracts, and research papers and produce structured summaries.
Audit code and configuration for vulnerabilities and draft remediation steps.
Use Argon as the reasoning core of multi-step agents and automated workflows.
Ask questions about screenshots, charts, and scanned pages by attaching an image.
| Feature | Gemini 4 Argon | Gemini 3.8 Flash |
|---|---|---|
| Positioning | Frontier reasoning | Fast, low-cost chat |
| Context window | 1M-token unified context and output | Large context, optimized for latency |
| Input | Text, image, system prompt | Text, image, system prompt |
| Price (input / output per 1M tokens) | $2.00 / $10.00 | $2.50 / $12.50 |
| Best for | Long-horizon coding and deep analysis | Real-time assistants and high-volume tasks |
Submit a JSON request with your Muapi API key, save the request ID, and poll the shared result endpoint for the completed text.
curl -X POST https://api.muapi.ai/api/v1/gemini-4-argon \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"Review this design doc and list the three biggest risks with mitigations","system_prompt":"You are a staff engineer. Be concise and specific."}'
# Response: {"request_id":"REQUEST_ID"}Call gemini-4-argon with a prompt, and optionally a system_prompt and image_url.
curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \ -H "x-api-key: YOUR_API_KEY" # Poll until status is completed, then read the generated text from outputs.
Poll until the task is completed, then read the generated text from outputs.
Gemini 4 Argon is the first model in Google DeepMind's Gemini 4 series, a frontier reasoning model with a 1M-token unified context and output window for long-horizon coding, knowledge work, and cybersecurity.
Billing is per token at $2.00 per million input tokens and $10.00 per million output tokens, with a minimum charge of $0.0002 per call. There is no subscription.
Yes. Pass an image_url with your prompt to ask questions about a screenshot, diagram, or photographed document.
The endpoint is asynchronous. You receive a request_id immediately, then poll the predictions result endpoint or provide a webhook URL to be notified when the text is ready.
This page covers the Gemini 4 series. Argon is the first model available, and additional Gemini 4 models will be listed here as they launch.
Argon is the frontier reasoning tier built for deep, long-context work, while Flash is optimized for low-latency, high-volume chat. Both use the same prompt, image_url, and system_prompt fields.
Add frontier reasoning with a 1M-token context to your product with one API.