Google: Gemini 4 Carbon
Chat-50%google/gemini-4-carbonGemini 4 Carbon is Google’s reasoning model for complex coding, research and agent workflows. It supports text, image, audio, video and PDF input, tool calling and structured output. Input, thinking and output share a 1,048,576-token window. Foreground output is limited to 32,768 tokens without streaming or 65,536 with streaming. Thinking levels are low, medium (default) and high. Only implicit context caching is supported; extended output requires continuation support.
Context Window
1M
Max Output Tokens
66K
Released
2026-11-10
Capabilities
VisionFunction CallingReasoningPrompt CachingWeb SearchAudio InputVideo InputPDF Input
Available Providers
Gemini
Supported Protocols
openaigemini
Providers
Gemini
provider.type: "gemini"-50%
Input Tokens
$2/M
$4/M
Output Tokens
$10/M
$20/M
Cache Read
$0.1/M
$0.2/M
Audio Input
$2/M
$4/M
Cached Image
$0.1/M
$0.2/M
Cached Audio
$0.1/M
$0.2/M
Cached Video
$0.1/M
$0.2/M
Web Search
$0.014/R
Protocols
openai
/v1/chat/completionsgemini
Documentation
How to call Gemini 4 Carbon on OfoxAI: model IDs, request parameters and the errors you are most likely to hit.
- Gemini API guideModel IDs, choosing an endpoint, code examples and common errors.
- Thinking and reasoning settingsTurn thinking on or off, set the effort or token budget, and read the reasoning output.
- Web searchLet the model search the web and return sources with its answer.
- PDF, audio and structured outputSend PDF and audio as input and constrain the response to a schema.
Code Examples
from google import genaiclient = genai.Client(api_key="YOUR_OFOX_API_KEY",http_options={"api_version": "v1beta", "base_url": "https://api.ofox.io/gemini"},)response = client.models.generate_content(model="google/gemini-4-carbon",contents="Hello!",)print(response.text)
Related Models
Frequently Asked Questions
Google: Gemini 4 Carbon on Ofox.ai costs $4 per million input tokens and $20 per million output tokens. Pay-as-you-go, no monthly fees.
Google: Gemini 4 Carbon supports a context window of 1M tokens with max output of 66K tokens, allowing you to process large documents and maintain long conversations.
Simply set your base URL to https://api.ofox.io/v1 and use your Ofox API key. The API is OpenAI-compatible — just change the base URL and API key in your existing code.
Google: Gemini 4 Carbon supports the following capabilities: Vision, Function Calling, Reasoning, Prompt Caching, Web Search, Audio Input, Video Input, PDF Input. Access all features through the Ofox.ai unified API.