Gemini 3 Pro Preview

Chat

Built by Google · United States · deepmind.google

Google chat model.

How to use

model: "google/gemini-3-pro-preview"

Cap quality here and save

Hard questions go to this model, easy ones drop to something cheaper. Set a savings target (0–0.8) and the router tunes itself until realized savings converge on it.

{
  "model": "celon/auto",
  "celon": {
    "anchor": "google/gemini-3-pro-preview",
    "savings_target": 0.4
  }
}

Naming a model directly gets you that model — savings controls don't apply.

Context

1.0M

Max output

—

Input $/1M

$2.00

Output $/1M

$12.00

Features

Image inputTool callingJSON

Serving providers

Google

Providers

Price, latency, and uptime per provider serving this model. Celon routes each request to the fastest and cheapest of them. Click a row for regions and data policies.

Loading provider metrics…

Performance

Benchmark quality scores, and measured speed per provider.

Quality

Epoch Capabilities IndexComposite over 50+ benchmarks

153.2

GPQA DiamondPhD-level science questions

92.6%

AIMECompetition mathematics

91.4%

FrontierMathResearch-level mathematics

37.6%

SWE-bench VerifiedReal GitHub issue resolution

72.9%

Terminal-BenchAgentic terminal tasks

69.4%

ARC-AGI-2Abstract reasoning

31.1%

SimpleQA VerifiedFactual accuracy

72.9%

HLEHumanity's Last Exam

37.5%

At each model's best reasoning effort. Benchmarks it wasn't run on are omitted.

Speed

Loading provider metrics…