google/gemini-4-carbon

Model release OpenRouter 模型 Source language: Chinese

OpenRouter 于 2026-11-11 收录该模型(收录时间不等于模型正式发布日期);上下文:1,048,576 tokens;输入价:$2.0000/百万 tokens;输出价:$10.0000/百万 tokens;定位(来源描述):Gemini 4 Carbon is a frontier Gemini 4 model from Google for coding, agentic and complex…

Full text

Google: Gemini 4 Carbon

google/gemini-4-carbon

Compare

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.

Modalities

In / out price

50% off

$2 / $10per 1M

Context

1.0M

Released

Nov 10, 2026

Compare

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput

50% off

$4.00$2.00$20.00$10.00$0.20$0.10$0.20$0.107.84s30 tok/s

99.24%

50% off

$4.00$2.00$20.00$10.00$0.20$0.10$0.20$0.106.66s45 tok/s

97.65%

Not used in Standard routing:Why these endpoints are not used

Flex50% off

$2.00$1.00$10.00$5.00$0.10$0.05$0.10$0.055.99s44 tok/s

98.78%

50% off

$4.40$2.20$22.00$11.00$0.22$0.11$0.22$0.116.10s31 tok/s

97.50%

50% off

$4.40$2.20$22.00$11.00$0.22$0.11$0.22$0.114.15s43 tok/s

100.00%

Priority50% off

$7.20$3.60$36.00$18.00$0.36$0.18$0.36$0.1816.85s26 tok/s

100.00%

Priority50% off

$7.20$3.60$36.00$18.00$0.36$0.18$0.36$0.187.40s33 tok/s

100.00%

Provider50% offInput / Output /M

Input $2.00 / Output $10.0050% off

30 tok/s99.24% uptime

Input $2.00 / Output $10.0050% off

45 tok/s97.65% uptime

Not used in Standard routing:Why these endpoints are not used

Input $1.00 / Output $5.0050% off

Flex

44 tok/s98.78% uptime

Input $2.20 / Output $11.0050% off

31 tok/s97.50% uptime

Input $2.20 / Output $11.0050% off

43 tok/s100.00% uptime

Input $3.60 / Output $18.0050% off

Priority

26 tok/s100.00% uptime

Input $3.60 / Output $18.0050% off

Priority

33 tok/s100.00% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.6445/M tokens

Output

$10.36/M tokens

Gemini 4 Carbon — Price history

Price history

No effective price data for this selection

$0.640$10.2172.5%84.5%

$0.660$14.6983.0%6.5%

Not used in Standard routing:Why these endpoints are not used

Flex

$0.548$5.0247.8%5.1%

$0.607$11.0376.4%0.1%

Priority

$0.640$18.7286.8%0.7%

Priority

$0.889$34.8691.4%3.2%

$1.047$11.4659.2%0.1%

ProviderInput / Output /M

Token share

Effective in $0.640 / Effective out $10.21

Token share

84.5%

Effective in $0.660 / Effective out $14.69

Token share

6.5%

Not used in Standard routing:Why these endpoints are not used

Effective in $0.548 / Effective out $5.02

FlexToken share

5.1%

Effective in $0.607 / Effective out $11.03

Token share

0.1%

Effective in $0.640 / Effective out $18.72

PriorityToken share

0.7%

Effective in $0.889 / Effective out $34.86

PriorityToken share

3.2%

Effective in $1.047 / Effective out $11.46

Token share

0.1%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

45tok/s

P50, best across providers

Latency

4.15s

P50, best provider

Reasoning effortLatency / throughput

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)The model was reachable. Request routed to a provider.

100.00%

Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.

98.92%

Availability over the last 3 days

Last 72 hours

Availability 98.92%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

98.34%

Without Routing

96.24%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for Google: Gemini 4 Carbon (Artificial Analysis, Design Arena and Vals AI)

SourceBenchmarkScore

Artificial AnalysisGemini 4 Carbon (High) Intelligence Index60.0

Artificial AnalysisGemini 4 Carbon (High) AA-LCR79.3%

Artificial AnalysisGemini 4 Carbon (High) GDPval-AA61.6%

Artificial AnalysisGemini 4 Carbon (High) Terminal-Bench 4.065.2%

Artificial AnalysisGemini 4 Carbon (High) AA-Omniscience Accuracy58.3%

Artificial AnalysisGemini 4 Carbon (High) AA-Omniscience Non-Hallucination Rate88.6%

Design ArenaGemini 4 Carbon Agents Arena Mobile Apps Elo1288

Design ArenaGemini 4 Carbon Agents Arena Pdf Elo1340

Design ArenaGemini 4 Carbon Agents Arena Spreadsheet Elo1224

Design ArenaGemini 4 Carbon Models Arena 3D Elo1427

Design ArenaGemini 4 Carbon Models Arena Code Categories Elo1348

Design ArenaGemini 4 Carbon Models Arena Game Development Elo1333

Design ArenaGemini 4 Carbon Models Arena UI Component Elo1347

Design ArenaGemini 4 Carbon Models Arena Website Elo1319

Vals AIGemini 4 Carbon EMB v1.176.5%

Vals AIGemini 4 Carbon Finance Agent v2.167.2%

Vals AIGemini 4 Carbon Vals Finance Index67.0%

Vals AIGemini 4 Carbon HLAB-Vals-1.117.5%

Vals AIGemini 4 Carbon Vals Legal Index75.9%

Vals AIGemini 4 Carbon Legal Research Bench55.8%

Vals AIGemini 4 Carbon Vals RSI Index27.7%

Vals AIGemini 4 Carbon Tax Agent Bench51.8%

Vals AIGemini 4 Carbon Terminal-Bench 4.063.1%

Vals AIGemini 4 Carbon Vals Index70.5%

Vals AIGemini 4 Carbon Vibe Code Bench 1-100 v1.149.7%

Vals AIGemini 4 Carbon Vibe Code Bench v1.189.1%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

1.

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

28.7Btokens

2.

pi

There are many coding agents, but this one is yours.

23.4Btokens

3.

omp

new

20Btokens

4.

Codex

A coding agent that helps you build and ship with AI

16.7Btokens

5.

Cursor

new

9.26Btokens

Rest of OpenRouter

100B

Top 5 Apps

14.1B

Activity

Token volume and request traffic to this model over time.

API

Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.

Google: Gemini Pro versions

Available through EU in-region routing

Google: Gemini 4 Carbon is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).

Providers in EuropeGoogle VertexPrices in Europe

Google Vertex: $2.20/M tokens input, $11/M tokens output

Base URLhttps://eu.openrouter.ai/api/v1PlanRequires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing

Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs

Frequently asked questions

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.

Gemini 4 Carbon costs $2.00/M input tokens and $10.00/M output tokens, with separate rates for Cache Read at $0.10/M tokens, Cache Write at $0.1875/M tokens, Image Input at $2.00/M tokens, Input Audio at $2.00/M tokens, Input Audio Cache at $0.10/M tokens, and Web Search at $14.00/1K calls.

Gemini 4 Carbon has a 1,048,576 token context window. It supports up to 1,048,576 completion tokens.

Yes. Gemini 4 Carbon accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

Gemini 4 Carbon accepts audio, files such as PDFs, images, text, and video as input and returns text.

Gemini 4 Carbon is served by 2 providers on OpenRouter: Google AI Studio and Google Vertex. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

Gemini 4 Carbon was released on November 10, 2026.

Sources

← Back to the daily brief