Gemini Omni coming soonLearn more
EvoLink.AIEvoLink.AI

Gemini 3.1 Pro Preview API

EvoLink's Gemini 3.1 Pro Preview route targets top-tier coding, agentic reasoning, and deep analysis workflows — with 1M context, thinking mode, and a dedicated CustomTools variant for tool-heavy workloads.
Model Type:
✓Gemini 3.1 Pro
Price: 

$1.865 - 3.729(~ 126.8 - 253.6 credits) per 1M input tokens; $11.182 - 16.774(~ 760.4 - 1140.6 credits) per 1M output tokens

$0.187 - 0.374(~ 12.7 - 25.4 credits) per 1M cache read tokens

Google Search grounding charged separately per query.

Highest stability with guaranteed 99.9% uptime. Recommended for production environments.

Use the same API endpoint for all versions. Only the model parameter differs.

Gemini 3.1 Pro — Flagship Gemini Reasoning API

Gemini 3.1 Pro is EvoLink's flagship Gemini route for coding and reasoning tasks. Access it at $2 input / $12 output per 1M tokens with OpenAI SDK compatibility, thinking mode, and full tool use support.

example 1

What Makes Gemini 3.1 Pro Google's Top Model

Frontier Coding & Reasoning

Gemini 3.1 Pro is positioned for high-quality coding tasks. It performs competitively on public benchmarks such as SWE-bench and HumanEval alongside Claude Opus 4.6 and GPT 5.4.

Try Coding Tasks
example 2

Agentic Workflows & CustomTools

Built for multi-step agent loops with thinking mode, function calling, structured outputs, and a dedicated CustomTools variant (gemini-3.1-pro-customtools) for tool-heavy pipelines.

Build Agents
example 3

1M Context + Production Controls

Process entire codebases, long documents, and multi-turn conversations in a single request. Use caching, Batch API, and Google Search grounding for production-scale deployments.

Check Limits
example 4

Why Choose Gemini 3.1 Pro on EvoLink

Gemini 3.1 Pro offers strong reasoning at a lower price point than comparable models — and EvoLink adds OpenAI SDK compatibility, usage tracking, and unified billing.

Best Price-to-Performance Ratio

At $2/$12 per 1M tokens, Gemini 3.1 Pro is priced well below Claude Opus 4.6 ($15/$75). Check our benchmark comparisons for detailed performance data on specific tasks.

Drop-In OpenAI Replacement

Use the same /v1/chat/completions format and OpenAI SDK. Switch from GPT models by changing one line — the model string and API base URL.

Full Usage Transparency

Every response includes prompt_tokens, completion_tokens, reasoning_tokens, and cache usage. Track cost per request and reasoning depth at the token level.

How to Call Gemini 3.1 Pro Preview

Use the OpenAI SDK format and the gemini-3.1-pro-preview model string.

1

Step 1 - Set the Model

Use model: "gemini-3.1-pro-preview" in the request body.

2

Step 2 - Send Messages

Provide a messages array with role/content pairs (minimum length 1).

3

Step 3 - Inspect Output + Usage

Read choices[0].message.content and track usage.prompt_tokens, completion_tokens, and reasoning_tokens.

Open Developer Playground

Gemini 3.1 Pro Preview Specs

Technical details for EvoLink's flagship Gemini reasoning route

Model

Model ID

gemini-3.1-pro-preview (general) or gemini-3.1-pro-customtools (tool-heavy workloads).

Pricing

Input Pricing

$2.00 per 1M tokens (standard), $4.00 per 1M for prompts over 200K tokens.

Pricing

Output Pricing

$12.00 per 1M tokens (standard), $18.00 per 1M for high-volume prompts over 200K tokens.

Limits

Context Window

1,000,000 input tokens (1M context). Max output: 65,536 tokens.

Multimodal

Supported Inputs

Text, code, images, video, audio, PDF. Output is text only. Google Search grounding + URL context available.

Lifecycle

Upgrade from Earlier Gemini 3 Routes

Use this route when you want EvoLink's newer Gemini Pro option for coding, instruction following, and agentic tool use.

Gemini 3.1 Pro API FAQ

Everything you need to know about the product and billing.

Gemini 3.1 Pro performs competitively on coding and reasoning benchmarks. Its key advantage is pricing: at $2/$12 per 1M tokens, it is priced well below Claude Opus 4.6 ($15/$75) and GPT 5.4. See our detailed benchmark comparisons for task-specific performance data.
Test coding accuracy, instruction following, tool-use reliability, and response consistency on your own prompts. EvoLink also exposes a dedicated CustomTools variant for tool-heavy workflows. Both routes use the same 1M context window and listed pricing.
gemini-3.1-pro-customtools is a separately exposed route on EvoLink optimized for requests that rely on function calling, tools arrays, and multi-step agent loops. Same pricing as the base route. Use the base gemini-3.1-pro-preview for general reasoning.
Standard: $2.00 input / $12.00 output per 1M tokens. Over 200K token prompts: $4.00 input / $18.00 output. Cache reads: $0.20 per 1M tokens. Check the EvoLink dashboard and official Gemini pricing docs for current grounding and tool charges.
Use "gemini-3.1-pro-preview" as the model value. For tool-heavy workloads, use "gemini-3.1-pro-customtools". Both route to api.evolink.ai/v1/chat/completions with Bearer token auth.
Yes. Thinking mode enables chain-of-thought reasoning for complex tasks. Reasoning tokens are tracked separately in completion_tokens_details.reasoning_tokens, so you can monitor reasoning depth and cost.
1,000,000 input tokens (1M context) with up to 65,536 max output tokens. Supports caching (implicit + explicit) and Batch API for long-context pipelines.
Yes. EvoLink's endpoint is fully OpenAI SDK compatible. Use /v1/chat/completions with the standard model + messages format. Switch from any OpenAI model by changing the model string and base URL — no other code changes needed.
Wait 5-10 minutes and retry. For production reliability, use the recommended preview route for your workload and keep a fallback model configured.

Where Gemini 3.1 Pro fits in the Gemini family

Gemini 3.1 Pro is EvoLink's flagship Gemini reasoning route. Use CustomTools for tool-heavy agent workflows, Flash for lower-cost tasks, or compare against GPT 5.4 and Claude Opus 4.6 for competitive context. Explore the Gemini family or read comparison guides to pick the right model.

Gemini API FamilyGemini 3.1 Pro CustomToolsGemini 3 Flash PreviewGemini 3.1 Flash Lite