DeepSeek V4.1 Flash is a mixture-of-experts model DeepSeek released in September 2026, the smallest model in its new architecture family. It uses a causal encoder-decoder design that activates about 8B parameters per token while reading the prompt and 16B while writing, which DeepSeek says makes it cheap to run on input-heavy agentic workloads. In Jantaris it has a 1,000,000-token context window and up to 384,000 output tokens. DeepSeek reports that it beats its larger V4 Pro model in its own tests, and the weights are open on Hugging Face.
Best for
- Coding agents that read large codebases before each change
- Bulk processing of long documents on a tight budget
- Generating very long outputs in one call
- Tests and prototypes where you run thousands of requests
How to call it
The OpenAI SDK works with the Jantaris base URL:
from openai import OpenAI
client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
model="deepseek-v4.1-flash",
messages=[{"role": "user", "content": "Refactor this module to remove the global state."}],
)
print(resp.choices[0].message.content)FAQ
How much does DeepSeek V4.1 Flash cost?
$0.54 / $1.75 per 1M tokens in Jantaris. It is among the lowest rates in the chat catalog, which is the main reason people pick it for agents that send long prompts on every step.
Do I need a DeepSeek account to use it?
No. You call it with your Jantaris key and pay from the same balance you use for every other model, in USD with no subscription. Jantaris does not use your prompts or outputs to train models.
DeepSeek V4.1 Flash or DeepSeek V4 Pro?
DeepSeek itself says V4.1 Flash is ahead of V4 Pro on performance, cost and speed in tests by several parties. Start with Flash and keep DeepSeek V4 Pro for prompts where you have measured a clear quality difference.
How to call DeepSeek V4.1 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="deepseek-v4.1-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": "deepseek-v4.1-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: "deepseek-v4.1-flash",
messages: [{ role: "user", content: "Write a haiku about Baltic amber." }],
});
console.log(resp.choices[0].message.content);