Official commercial channel
Gemini 3.1 Pro API
Google’s current reasoning flagship: a 1M window, priced in tiers by request length.
Context
1M
Max output
64K
Serving precision
Vendor-managed
Licence
Proprietary
Gemini 3.1 Pro is Google’s current tier for flagship reasoning, with a 1-million-token context window, roughly 64K maximum output and text plus image input. Pricing steps with context length: $2 per million input and $12 per million output up to 200K, rising to $4 and $18 above that. We provide access through an official commercial channel — no account pools, no reverse-engineered client protocols.
Proprietary, provided through the vendor’s official commercial licensing. Contract, invoice, audit trail — not an account pool and not a reverse-engineered client protocol.
Specifications
- Model ID
- gemini-3.1-pro
- Published by
- How it is served
- Official commercial channel
- Licence
- Proprietary
- Context
- 1M
- Max output
- 64K
- Serving precision
- Vendor-managed
- Input modalities
- Text · Image
- Released
- 2026-02-19
- Price
- See pricing
What it is good at
- Competitive on short requests
- Below 200K tokens the rate is $2 input and $12 output per million. The overwhelming majority of production requests land in that band, so real invoices tend to come in well under a top-tier estimate.
- A million-token window with image input
- Long documents and images in the same request, which suits analysis over material full of charts and scans.
- The other pole of the ecosystem
- Part of the value of carrying Gemini is lateral comparison: run the same prompts against Claude, GPT and the open flagships on one key and the capability-versus-cost differences surface immediately.
Where it fits
- Analysis of long material containing charts and scans
- Cost-sensitive pipelines whose requests mostly stay under 200K tokens
- Cross-vendor comparison and supplier redundancy
- Workloads that need consistency with the Google ecosystem
Three steps
Change one line of base_url, put the model ID below in the `model` field, and leave the rest of your code alone.
model: "gemini-3.1-pro"
curl https://api.nexaix.net/v1/chat/completions \
-H "Authorization: Bearer $NEXAIX_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3.1-pro",
"stream": true,
"messages": [
{ "role": "user", "content": "Summarise this contract." }
]
}'import os
from openai import OpenAI
client = OpenAI(
# the only new line
base_url="https://api.nexaix.net/v1",
api_key=os.environ["NEXAIX_API_KEY"],
)
# everything below is your existing code
stream = client.chat.completions.create(
model="gemini-3.1-pro",
stream=True,
messages=[
{"role": "user", "content": "Summarise this contract."}
],
)
for chunk in stream:
print(chunk.choices[0].delta.content or "", end="")import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.nexaix.net/v1", // the only new line
apiKey: process.env.NEXAIX_API_KEY,
});
// everything below is your existing code
const stream = await client.chat.completions.create({
model: "gemini-3.1-pro",
stream: true,
messages: [
{ role: "user", content: "Summarise this contract." }
],
});
for await (const chunk of stream) {
process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}About this model
How exactly does the pricing tier work?
It steps by the context length of the request: below 200K tokens you pay the lower rate, above it the whole request moves to the higher one. Estimate from your real input-length distribution rather than an average.
Why is there no Gemini 3.5 Pro?
Because it does not exist. Google’s Pro tier this generation is 3.1 Pro; 3.5 and 3.6 are Flash models — the faster, cheaper speed tier. Anyone marketing a "Gemini 3.5 Pro" is working from an unreliable source.
Do you log conversations?
No. Nothing is written to disk on our side; we keep only the request ID, token counts, timestamp and status code needed for billing and debugging. The upstream vendor’s data handling is governed by their own terms.
Also worth a look
- Official commercial channelClaude Opus 5A million tokens that never degrade, five-level effort control, zero data retention.claude-opus-5
- Official commercial channelClaude Fable 5Anthropic’s tier above Opus — the one you pick when the ceiling is the point.claude-fable-5
- Official commercial channelGPT-5.6 SolThe frontier tier of the GPT-5.6 family: 1.05M context, built for long-horizon agents.gpt-5.6-sol
Run your own eval on it
Take the trial credit, run the evals you already have against this model, and then decide whether it gets your production traffic.
