Use It for Chat, Not Agents Yet
On Kie.ai this model supports chat requests only and has not been adapted for agent use; connecting it to an agent may cause request errors or unexpected results.
gemini-3.1-pro-openai
Automatically append previous messages to maintain multi-turn context. May increase token usage.
Grounding with Google Search
Source:Google Search
Thinking process
Reasoning effort
How can I help you today?
README
Complete guide to using gemini-3.1-pro
Gemini 3.1 Pro API: Specs, API Key and Pricing on Kie.ai
Gemini 3.1 Pro is Google's Gemini 3.1 Pro Preview model (gemini-3.1-pro-preview), built by Google for better thinking and more grounded answers. It reads up to 1,048,576 input tokens of text, images, video, audio and PDF and replies in text. On Kie.ai you call it with an OpenAI-compatible chat completions request and your Kie.ai API key.
Gemini 3.1 Pro API Parameters on Kie.ai
Kie.ai serves the model as gemini-3.1-pro for chat requests in the OpenAI chat completions format.
| Parameter | Value |
|---|---|
| Model | gemini-3.1-pro (Google model code: gemini-3.1-pro-preview) |
| Endpoint | POST /gemini-3.1-pro/v1/chat/completions, OpenAI chat completions format |
| Input token limit | 1,048,576 tokens |
| Output token limit | 65,536 tokens |
| Inputs | Text, images, video, audio and PDF; every media file is passed as an image_url part with its URL |
| Output | Text |
| Reasoning effort | low or high; default high |
| Thought summaries | include_thoughts, default off; works only with streaming |
| Tools | Function calling and Google Search grounding |
| Streaming | Off by default; set stream to true for server-sent events |
| Pricing | Shown in the pricing card above |
What Gemini 3.1 Pro Does Well
Google describes Gemini 3.1 Pro Preview as refining the Gemini 3 Pro series with better thinking, improved token efficiency and more grounded, factually consistent answers.
Long, Mixed-Media Inputs
Long documents and media files can be read together in one request.
Reasoning You Can Dial
Set reasoning effort to high for harder problems or low for faster replies, and stream thought summaries when you want to see how it worked.
Answers Grounded in Search
Turn on the Google Search tool when an answer needs current information instead of only what the model already knows.
Structured Replies
Ask for structured output when your code needs fields it can parse rather than free text.
Gemini 3.1 Pro vs GPT-6 Astra: What the Leaderboards Show
Searchers often compare Gemini 3.1 Pro with GPT-6 Astra. The two leaderboards disagree: LMArena's text leaderboard scores Gemini 3.1 Pro Preview 1487 and GPT-6 Astra 1475 (as of October 8, 2026), while the Artificial Analysis Intelligence Index gives Gemini 3.1 Pro Preview 30 and GPT-6 Astra 53 (as of October 2026). The two measure different things: LMArena ranks models by human preference votes, while the Artificial Analysis index aggregates benchmark results. Test both on your own prompts; both models are available on Kie.ai.
| Gemini 3.1 Pro | GPT-6 Astra | |
|---|---|---|
| LMArena text score (Oct 8, 2026) | 1487 | 1475 |
| Artificial Analysis Intelligence Index (Oct 2026) | 30 | 53 |
| Available on Kie.ai | Yes | Yes |
Typical Uses for Gemini 3.1 Pro on Kie.ai
On Kie.ai the model handles chat requests. These are the kinds of tasks that fit.
Long-Document Questions
Questions answered across long reports, contracts or research papers in a single request.
Video, Audio and PDF Analysis
Summaries and answers drawn from recordings, slide decks and PDF documents passed by URL.
Research With Current Information
Answers that need fresh facts, with the Google Search tool switched on.
Code Help in Chat
Writing, explaining and reviewing code inside a chat conversation.
How to Use the Gemini 3.1 Pro API on Kie.ai
Step 1. Get Your API Key
Step 2. Test Your Prompt in the Playground
Step 3. Send a Request
Tips Before You Ship
Send Every File as image_url
Images, video, audio and documents all use the same format: a content part of type image_url whose url points to the file.
Thought Summaries Need Streaming
include_thoughts only takes effect with stream set to true; the summaries arrive in reasoning_content before the answer.
Why Use Gemini 3.1 Pro on Kie.ai
Same Key Across the Kie.ai Market
All models in the Kie.ai market use the same authentication method: an Authorization: Bearer YOUR_API_KEY header with the key from your Kie.ai API Key Management page.
Free Credits to Start
New Kie.ai users receive 80 free credits for testing. Your account has one credit balance that all generation services draw from, and you can check it with the Get Remaining Credits API.
Nothing to Host or Install
Kie.ai provides this model as an API with a playground; there is no app to download or install. There are no GPUs or model weights to manage.
Frequently Asked Questions About Gemini 3.1 Pro API on Kie.ai
Pricing is shown in the pricing card above.
On Kie.ai you do not need a Google API key: create a key on the Kie.ai API Key Management page and send it in the Authorization header. If you want Google's own key for Google's API, Google issues it in Google AI Studio.
Not on Kie.ai: each request uses credits, as shown in the pricing card above. New Kie.ai users receive 80 free credits for testing.
LMArena's text leaderboard puts Gemini 3.1 Pro Preview higher, and the Artificial Analysis Intelligence Index puts GPT-6 Astra higher; the comparison table above has the scores.
Gemini 3.1 Pro is Google's Gemini 3.1 Pro Preview model for reasoning over long, multimodal inputs. On Kie.ai it is the chat model gemini-3.1-pro; its parameters and uses are covered above.
Google lists an input limit of 1,048,576 tokens and an output limit of 65,536 tokens. The parameter table above has the other limits.
Yes. Add the Google Search tool to a request when the answer needs current information; the parameter table above lists the supported tools.
Text, images, video, audio and PDF. On Kie.ai each file is sent as an image_url part with the file's URL, whatever its type.