AI KNOWLEDGE DESK

Models · entities · concepts · comparisons · practical tools

GETLLMS.ORG

🚀Function Overview

Google announced gemini-4-carbon on October 28, 2026, and launched it to Developer API general availability as a paid model on November 10, 2026. It processes text, images, audio, video, and PDFs across a 1,048,576-token context window, outputting text up to 65,536 tokens synchronously or up to 1,048,576 tokens via optional long decode. Carbon supports low, medium, and high thinking modes with medium set as default. Developer API pricing is $2 per million input tokens and $10 per million output tokens through November 30, 2026, transitioning to $4 per million input tokens and $20 per million output tokens on December 1, 2026.

Key Features

  • Google’s most capable reasoning model for complex software engineering and long-horizon tasks.
  • Unified 1,048,576-token context window accepting text, image, audio, video, and PDF inputs
  • Carbon supports low, medium, and high thinking levels. Medium is the Developer API default.
  • In the Interactions API, one non-streaming hop on Carbon or Argon is capped at 32,768 tokens. Streaming hops and Neon or Hydrogen non-streaming hops can use up to 65,536. Long decode can continue within the shared 1,048,576-token context.

Use Cases

  • •Refactoring complex enterprise software systems and resolving cross-repository dependencies
  • •Work through multi-step debugging tasks and verify the fix with regression tests.
  • •Analyzing extensive code repositories alongside technical audiovisual specifications and system logs

⚙️Input Parameters

contents

array

Gemini generateContent contents with text or supported media parts

generationConfig

object

Documented output and thinking configuration; ordinary synchronous output is distinct from long decode

💡Usage Examples

Example 1

Input Parameters

{
  "contents": [
    {
      "role": "user",
      "parts": [
        {
          "text": "Review the supplied project notes. List confirmed findings, open questions, and the checks that would resolve them."
        }
      ]
    }
  ]
}

Output Results

Illustrative generateContent request. Inspect candidates and usageMetadata after running it with your own API access.