Google: Gemini 3.1 Pro Preview

Reasoning-first Gemini preview for agentic coding and complex problem solving

gemini-3.1-pro-preview

AvailableGoogleChatCached input 90% less

Input / Output-50%: $2.00 / $12.00$1.00 / $6.00
Context: 1M
Max output: 64K
Modalities: VisionChat
Capabilities: Tool usePrompt CacheReasoning

About Gemini 3.1 Pro Preview

Gemini 3.1 Pro is Google's reasoning-first Pro model in the Gemini 3.1 line, available as a preview. Google describes advanced intelligence, complex problem-solving, and strong agentic capabilities. It reads text and images, thinks before answering, calls tools, returns schema-bound JSON, and supports caching. It is the pick when a Flash model's answers on a hard task are not good enough.

Where it works well

Built-in thinking suits multi-step problems, debugging, and planning where a wrong first answer is costly.
Agentic coding is a stated focus, with tool calling for edit-and-test loops.
Images and text can be combined in one prompt for diagram, chart, and UI analysis.
Schema-bound JSON output and caching support production pipelines.

When to choose another model

It is a preview, so Google may change behavior or retire it in favor of a stable Pro release.
For simple, high-volume calls the Flash and Flash-Lite models answer faster and are the better fit.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Text to textGemini API

POST/v1beta/models/gemini-3.1-pro-preview:generateContent

API format:

curl "https://api.tokenlab.sh/v1beta/models/gemini-3.1-pro-preview:generateContent" \
  -H "Content-Type: application/json" \
  -H "x-goog-api-key: sk-xxx" \
  -d '{
    "contents": [
      {"role": "user", "parts": [{"text": "Hello!"}]}
    ]
  }'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Input tokens <= 200K · Input token range · <=200k

per 1M tokens

Official price: Input $2.00 / Output $12.00 / Cache read $0.20
Verified price: Input $1.00 / Output $6.00 / Cache read $0.10
Discount: -50%

Input tokens > 200K · Input token range · >200k

per 1M tokens

Official price: Input $4.00 / Output $18.00 / Cache read $0.40
Verified price: Input $2.00 / Output $9.00 / Cache read $0.20
Discount: -50%

Prompt cache pricing

Cache read

Official price: $0.20
Verified price: $0.10
Discount: -50%

Tool fees

Charged per call, only when the model uses the tool.

Web search

Official price: $0.014/search
Verified price: $0.007/search
Discount: -50%

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Input tokens <= 200K · Input token range · <=200k	Input $2.00 / Output $12.00 / Cache read $0.20	Input $1.00 / Output $6.00 / Cache read $0.10	-50%
Input tokens > 200K · Input token range · >200k	Input $4.00 / Output $18.00 / Cache read $0.40	Input $2.00 / Output $9.00 / Cache read $0.20	-50%
Prompt cache pricing
Cache read	$0.20	$0.10	-50%
Tool feesCharged per call, only when the model uses the tool.
Web search	$0.014/search	$0.007/search	-50%

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 13, 2026, 6:21 AM UTC

30-day success rate: 99.6%
7-day median latency: 4.8sn=310
7-day P95 latency: 31.2sn=310
30-day requests: 100+
Last active: 1 hour ago

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Gemini 3.1 Pro Preview in Console with a prompt ready to edit or send.

Help me try gemini-3.1-pro-preview with a short message at /v1beta/models/gemini-3.1-pro-preview:generateContent. Show the reply, latency, and cost.

Use cases

Best for

Reasoning
Vision

Hard debugging
Give it a failing build log and the relevant files, and ask for the root cause and a minimal fix.
Research synthesis
Combine several long documents into a reasoned comparison with open questions listed.
Agent planning
Use it as the planner that splits a task and hands steps to cheaper models.
Visual analysis
Read a dashboard screenshot or architecture diagram and explain what it implies.

Prompt examples

This stack trace appears only under load. The two modules involved are attached; find the likeliest race condition and propose a fix.

Compare these three vendor proposals on risk, cost structure, and lock-in, and recommend one with reasons.

Look at this architecture diagram and list the single points of failure.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Gemini 3.1 Pro good at?

Complex reasoning, agentic coding, and analysis across text and images. Google describes it as the model for advanced problem-solving. Use thinking-heavy settings for hard tasks and cheaper models for routine ones.

Is Gemini 3.1 Pro stable?

No. Google lists it as a preview, which means the behavior may change and the model may be replaced. Keep an evaluation set so you can detect shifts when that happens.

Does it call tools and return JSON?

Yes. Function calling and schema-constrained output are both supported, which is what makes it usable as the planner or reviewer in an automated agent pipeline.

When should I pick 3.8 Flash instead?

When latency matters more than the last bit of reasoning depth. 3.8 Flash is a Flash-tier model that Google also aims at long-running software tasks; compare both on your hardest real prompts.

How much does Gemini 3.1 Pro Preview cost?

On TokenLab, Gemini 3.1 Pro Preview costs Input $1.00 / Output $6.00 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

What are the context window and output limit of Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview accepts up to 1,048,576 tokens of context and returns up to 65,536 tokens in one response.

Which endpoint should Gemini 3.1 Pro Preview use?

Use https://api.tokenlab.sh/v1beta/models/gemini-3.1-pro-preview:generateContent for Gemini 3.1 Pro Preview. The request example below shows the matching code shape.

Which operations does Gemini 3.1 Pro Preview support?

Gemini 3.1 Pro Preview supports Text to text. Select an operation above to see its endpoint and request example.