Gemini 3.6 Flash

Gemini 3.6 Flash is built for the pattern "a lot in, a little out": you send a long contract, manual or code file, and it returns a compact answer. Google released it in July 2026 as its everyday model for code, knowledge work and multimodal tasks. Google says it delivers cleaner results with fewer needless edits and spends noticeably fewer output tokens than Gemini 3.5 Flash.

No Google SDK is needed; the OpenAI SDK pointed at Kadraris works.

What you can use it for

• Answering questions about long documents and internal wikis
• Small code fixes, snippets and front-end tweaks for web or app UIs
• Turning long inputs into short summaries or structured notes
• Large batches of text tasks with brief outputs

Settings and limits

• Context window: 1,048,576 tokens.
• Output cap: 8,192 tokens per reply. Ask for short formats such as bullets or JSON.
• Set max_tokens to what you actually need, which keeps cost predictable.
• For long outputs, split the job into parts or move to a model with a higher limit.

Example request

from openai import OpenAI

client = OpenAI(
    base_url="https://kadraris.com/v1",
    api_key="sk-kd-...",
)
resp = client.chat.completions.create(
    model="gemini-3.6-flash",
    max_tokens=1500,
    messages=[{"role": "user",
               "content": "Which clauses in this contract "
                          "limit our liability?"}],
)
print(resp.choices[0].message.content)

Attach the document text after the question and ask for a numbered list to keep the answer short.

Questions

How is Gemini 3.6 Flash priced?

$0.79 / $3.95 per 1M tokens. Google's list price for the 3.6, 3.7 and 3.8 Flash models is an introductory rate that ends on December 31, 2026, so check this page again after that. Failed calls cost $0.

My answer stops mid-sentence. Why?

You hit the 8,192-token output cap. Split the request, or use a model with longer replies such as gemini-3.8-flash or DeepSeek V4 Flash.

Gemini 3.6 Flash in your code

from openai import OpenAI

client = OpenAI(base_url="https://kadraris.com/v1", api_key="sk-kd-...")
resp = client.chat.completions.create(
    model="gemini-3.6-flash",
    messages=[{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}],
)
print(resp.choices[0].message.content)

curl https://kadraris.com/v1/chat/completions \
  -H "Authorization: Bearer $KADRARIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gemini-3.6-flash", "messages": [{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}]}'

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://kadraris.com/v1", apiKey: process.env.KADRARIS_API_KEY });
const resp = await client.chat.completions.create({
  model: "gemini-3.6-flash",
  messages: [{ role: "user", content: "Suggest three opening shots for a short film about a cafe." }],
});
console.log(resp.choices[0].message.content);