UncoverAI
AI Models

LLM · Google

Gemini 4 Carbongemini-4-carbon

In Signals: Nov 2026

Maker's stated useGoogle's most capable reasoning model, for complex engineering and scienceChecked 13 Nov 2026 Source

Google's model page lists gemini-4-carbon with a unified context window of 1,048,576 tokens and a synchronous max output of 65,536, or up to 1,048,576 with long decode. Inputs are text, image, video, audio, and PDF, and output is text. The pricing page's paid-tier standard rate is $2 input and $10 output per 1M tokens through 30 November 2026, then $4 and $20 from 1 December 2026. The stored figures are the introductory rates. The model is not available on the free tier. Paid tier lists Used to improve our products as No. The Gemini API changelog dated 10 November 2026 says gemini-4-carbon is generally available (https://ai.google.dev/gemini-api/docs/changelog). Gemini Enterprise Agent Platform lists the same model id, gemini-4-carbon.

Caching and batch

  • Live API No
  • Cache read $0.10
  • Cache write Not supported (implicit caching only)
  • Cache TTL is Not stated by vendor
  • Minimum cacheable tokens 4,096
  • Batch discount is not stated as a percent
  • Batch input / output $1.00 / $5.00

Paid tier. Standard input is $2 and output is $10 through 30 Nov 2026, then $4 and $20 from 1 Dec 2026. Context caching is $0.10 through 30 Nov 2026 and $0.20 from 1 Dec 2026. Implicit caching only. Explicit cache writes are not supported. Batch input is $1 and batch output is $5 through 30 Nov 2026, then $2 and $10 from 1 Dec 2026. Batch context caching is $0.05 through 30 Nov 2026 and $0.10 from 1 Dec 2026.

Checked 10 Nov 2026 Source Minimum length

Checked 10 Nov 2026https://ai.google.dev/gemini-api/docs/models/gemini-4-carbon