Google / google/gemini-4-carbon

Gemini 4 Carbon - access through LLMTR

The 1M token context window is shared between input and output, and Google caps a single response at 65K tokens when streamed and 32K without streaming. Built for multi-step coding and debugging, architectural changes across large codebases, open-ended science problems, and analysis of documents that mix images, video, audio, or PDF. Thinking is tunable across low, medium, and high, with medium as the default, and thinking tokens bill as output. It has the same unit prices as Gemini 4 Argon but spends more thinking tokens on the same task and answers more slowly; for interactive coding and fast loops, Argon is the better fit. Introductory pricing applies through 30 November 2026; standard rates take over on 1 December 2026.

Technical specifications

Canonical IDgoogle/gemini-4-carbon
ProviderGoogle
Context window1,048,576 tokens
OperationsCHAT_COMPLETIONS, RESPONSES
Modalitiesimage, video, audio, text

Pricing

An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.

OperationMetricUnitPrice
CHAT_COMPLETIONSINPUT_TEXTPER_1M_TOKENS$2.00
CHAT_COMPLETIONSCACHE_READPER_1M_TOKENS$0.100000
CHAT_COMPLETIONSOUTPUT_TEXTPER_1M_TOKENS$10.00
CHAT_COMPLETIONSINPUT_AUDIOPER_1M_TOKENS$2.00
CHAT_COMPLETIONSINPUT_IMAGEPER_1M_TOKENS$2.00
CHAT_COMPLETIONSINPUT_VIDEOPER_1M_TOKENS$2.00
CHAT_COMPLETIONSTOOL_CALLPER_1K_CALLS$14.00
RESPONSESINPUT_TEXTPER_1M_TOKENS$2.00

Example usage

With existing OpenAI SDK flows, change only the base URL and model identifier.

curl https://llmtr.com/v1/chat/completions   -H "Authorization: Bearer llmtr-your_key"   -H "Content-Type: application/json"   -d '{"model":"google/gemini-4-carbon","messages":[{"role":"user","content":"Hello"}]}'

Related models