Gemini 3.6 Flash is the Flash model Google released in July 2026 and called its workhorse for coding, knowledge work and multimodal tasks. Google says it produces polished results with fewer unnecessary edits and uses noticeably fewer output tokens than Gemini 3.5 Flash. In Jantaris it has a context window of 1,048,576 tokens, and each reply is capped at 8,192 output tokens. That cap makes it a good fit for requests with long inputs and compact answers, such as questions over big documents.
Best for
- Question answering over long documents and knowledge bases
- Code snippets, fixes and web or app front-end tasks
- Summaries and structured notes from long inputs
- High-volume text jobs where output length stays short
How to call it
Use the OpenAI SDK with the Jantaris endpoint, no Google SDK needed:
from openai import OpenAI
client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
model="gemini-3.6-flash",
messages=[{"role": "user", "content": "List the risks in this lease agreement."}],
max_tokens=2000,
)
print(resp.choices[0].message.content)FAQ
What does Gemini 3.6 Flash cost per 1M tokens?
In Jantaris it costs $0.79 / $3.95 per 1M tokens. Google's current list price for the 3.6, 3.7 and 3.8 Flash models is an introductory rate that runs until December 31, 2026, so check the models page after that date.
Why is my Gemini 3.6 Flash answer cut off?
Replies are limited to 8,192 output tokens per request. If you need longer output, split the task into parts or use gemini-3.8-flash, which allows much longer replies.
Gemini 3.6 Flash or Gemini 3.8 Flash?
For new projects Google itself recommends its newest Flash model. Keep 3.6 Flash if your prompts are already tuned to it and short answers are enough, and compare both on the same requests before switching.
How to call Gemini 3.6 Flash via API
from openai import OpenAI
client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
model="gemini-3.6-flash",
messages=[{"role": "user", "content": "Write a haiku about Baltic amber."}],
)
print(resp.choices[0].message.content)curl https://jantaris.com/v1/chat/completions \
-H "Authorization: Bearer $JANTARIS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gemini-3.6-flash", "messages": [{"role": "user", "content": "Write a haiku about Baltic amber."}]}'import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://jantaris.com/v1", apiKey: process.env.JANTARIS_API_KEY });
const resp = await client.chat.completions.create({
model: "gemini-3.6-flash",
messages: [{ role: "user", content: "Write a haiku about Baltic amber." }],
});
console.log(resp.choices[0].message.content);