Gemini 4 Carbon is a large language model made by Google, added to the catalog on 10 Nov 2026. It has a context window of 1,048,576 tokens and costs $2.00 per 1M input tokens and $10.00 per 1M output tokens. It accepts text, images, files, audio and video and produces text. Typical uses: coding tasks, complex reasoning and analyzing supplied media and files.
Price checked · source: OpenRouter
What Gemini 4 Carbon is
Google Gemini 4 Carbon is a Gemini 4 model from Google described for coding, agentic work and complex reasoning. It accepts text, images, audio, video and files, and produces text. This combination allows prompts to include written material alongside supported media and documents, with text returned as the model’s output. The catalog lists reasoning, tool use, structured outputs and response-format support. These capabilities provide options for reasoning tasks, using tools and shaping responses into a requested format. The listed information does not identify particular tools or describe specific workflows, so those details depend on the surrounding application.
Gemini 4 Carbon has a context length of 1,048,576 tokens and a maximum output of 943,718 tokens. The context figure describes the amount of material it can take into account in a request or exchange, while the output limit indicates the listed upper bound for text generation. This capacity suits work that brings together substantial context, such as coding prompts that include extensive source material, questions about provided files, or reasoning tasks involving long inputs. Its acceptance of image, audio and video also suits tasks based on those supplied media, although the listed output is text rather than generated media. Structured-output and response-format support can be relevant when results need a defined text format. The model is described for agentic work, but the catalog does not specify the actions or tools involved.
Key features
- Accepts text, image, audio, video and file input
- Supports a context length of 1,048,576 tokens
- Produces text, with a maximum output of 943,718 tokens
- Supports tools, reasoning, structured outputs and response formats
Typical uses
- Coding tasks
- Complex reasoning
- Analyzing supplied media and files
- Generating structured text responses
Gemini 4 Carbon pricing
US dollars, list price through OpenRouter.
| Item | Standard | Batch |
|---|---|---|
| Input, per 1M tokens | $2.00 | $1.00 |
| Output, per 1M tokens | $10.00 | $5.00 |
| Cached input read, per 1M tokens | $0.10 | $0.05 |
| Cache write, per 1M tokens | $0.188 | $0.188 |
| Image input, per image | $0.000002 | $0.000001 |
| Web search, per search | $0.014 | $0.014 |
| Variant | Model id |
|---|---|
| Batch | google/gemini-4-carbon:batch |
A variant is the same model under another id, with prices or limits of its own. Its prices are the columns of the list above.
No price change since it was added on 10 Nov 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0040 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.058 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.080 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.315 |
Gemini 4 Carbon specifications
| Model id | google/gemini-4-carbon |
|---|---|
| Version | gemini-4-carbon-20261110 |
| Context window | 1,048,576 tokensAbout 786,432 English words |
| Max output | 943,718 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 10 Nov 2026 |
| Input types | audio, file, image, text, video |
|---|---|
| Output types | text |
| Tool calling | Yes |
| Structured outputs | Yes |
| Reasoning effort control | Yes |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Gemini 4 Carbon yet.
Compare Gemini 4 Carbon
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.