Gemini 4 Carbon via the Tokenator API

What the model is for

Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex reasoning work. It takes text, image, audio, video and file input, supports long decode of up to 1M output tokens through continuation, and controls thinking depth with low, medium and high reasoning effort.

Specification

Full nameGemini 4 Carbon
API IDgemini-4-carbon
Model vendorGemini
TypeText model (chat / completions)
Context window1.05M tokens
Max output32.8K tokens
Image inputyes
Tokenator usage multiplier2×
Output token multiplier3×
BadgeNEW
Input modalitiestext, image, audio, video, file
Aliasescursor-gemini-4-carbon
Supported API formats/v1/chat/completions /v1/responses /v1/messages
Upstream providers1
Current statusavailable
Data updated2026-11-13

Benchmarks

Intelligence index60
Vals Finance Index (Vals AI)67%
Vals Index (Vals AI)70.5%
Vals Legal Index (Vals AI)75.9%
Vals RSI Index (Vals AI)27.7%
Terminal-Bench 4.0 (Vals AI)63.1%
Vibe Code Bench v1.1 (Vals AI)89.1%
EMB v1.1 (Vals AI)76.5%
Finance Agent v2.1 (Vals AI)67.2%
Tax Agent Bench (Vals AI)51.8%
HLAB-Vals-1.1 (Vals AI)17.5%
Legal Research Bench (Vals AI)55.8%

Artificial Analysis and Vals AI data loaded into the catalog for this model.

Upstream providers

ProviderMultiplierOutput token multiplier
AiPartners Ltd2×3×

A request goes to the first available provider by priority; if it fails, Tokenator switches to the next one.

Example request

curl
curl https://api.tokenator.top/v1/chat/completions \
  -H "Authorization: Bearer sk-your-tokenator-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-4-carbon",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Compatible tools

The model can be used in any tool that accepts a custom base URL: Claude Code, Codex CLI, OpenCode, Cursor, Cline, Kilo Code, Cherry Studio, DeepSeek Harness, Hermes, Pi, OpenClaw.

FAQ about Gemini 4 Carbon

What is the API ID of Gemini 4 Carbon?

gemini-4-carbon — put this into the model field of your request.

What context window does Gemini 4 Carbon have?

1.05M tokens; max output is 32.8K tokens.

How many tokens will a request cost?

(input × 2×) + (output × 3×). See the documentation for details.

Which tools support this model?

Any tool that accepts a custom base URL: Claude Code, Codex CLI, OpenCode, Cursor, Cline, Kilo Code. Setup guides are in the integrations section.