Gemini 4 Carbon is Google's frontier multimodal model in the Gemini 4 family, designed for coding, agentic, and complex reasoning work. It accepts text, image, audio, video, and file input and returns text. It is available through Google's API and via OpenRouter. The price listed here is the OpenRouter listed price, checked Nov 18, 2026: $2.00 per million input tokens and $10.00 per million output tokens. No cached input rate is available from OpenRouter for this model.
Gemini 4 Carbon supports a 1,048,576-token context window and a maximum output of 943,718 tokens. It supports vision input, audio input, video input, function calling, and streaming — making it one of the most broadly multimodal models in the Gemini 4 family.
Within the Gemini 4 family, Carbon is the full-featured multimodal frontier variant. Gemini 4 Hydrogen ($1.50/$7.50) is the text-focused efficient variant. Gemini 4 Argon is the reasoning-focused variant. Carbon's multimodal input support makes it the natural choice for workflows involving images, audio, or video alongside text.
Data source: OpenRouter listing, checked Nov 18, 2026. Data confidence: Partial — benchmark scores are not available for this model at this time. The OpenRouter listed price may differ from Google's direct API pricing; check ai.google.dev for the current official rate.
Gemini 4 Carbon is suited to multimodal agentic workflows, complex coding tasks, and applications that require processing diverse input types — documents, images, audio, and video — within a single model call. Its 1M+ context window supports long multi-modal conversations and large document analysis.
Input price vs 143 frontier models tracked. Avg: $5.58/1M.
| Model ID | gemini-4-carbon |
| Provider | |
| Model Family | Gemini 4 |
| Tier | Frontier |
| Context Window | 1,048,576 tokens (1.0M) |
| Max Output Tokens | 943,718 tokens |
| Input Price | $2 / 1M tokens |
| Output Price | $10 / 1M tokens |
| Cached Input Price | Not available |
| Throughput | — |
| Latency (P50) | — |
| Intelligence Score | — |
| LiveBench Score | — |
| Cost per Task (LiveBench) | — |
| Data Confidence | Partial |
| HLE Score | — |
| GPQA Diamond | — |
| SWE-bench Score | — |
| OSWorld Score | — |
| BrowseComp Score | — |
| Terminal-Bench Score | — |
| AutoBench Score | — |
| Parameters | — |
| Vision Support | Yes |
| Function Calling | Yes |
| Streaming | Yes |
| Reasoning Model | No |
| Open Weights | No |
| Data Source | Static (verified) |
| Last Updated | 2026-11-18 |
Self-hosting gemini-4-carbon? These GPUs are well-matched for this model's compute and memory requirements.