All models

DeepSeek

DeepSeek V4.1 Flash

A Flash model that reads text and images, built to make long, input-heavy agent workloads cheaper to run.

Start buildingOpen dashboard

Context window
1M tokens
Accepts
Text and images
Input per 1M tokens
$0.50
Output per 1M tokens
$1.50

Send your first request

Set LYCEUM_API_KEY to your API key, then run this request. Usage is billed to your account.

Base URL
https://api.lyceum.technology/openai/v1
Model ID
deepseek/deepseek-v4.1-flash
Request fields
model, messages, stream, max_tokens, tools, tool_choice

Documented request fields for this endpoint. See the documentation for model-specific controls and limits.

app.py · Install openai and set LYCEUM_API_KEY

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["LYCEUM_API_KEY"],
    base_url="https://api.lyceum.technology/openai/v1",
)

response = client.chat.completions.create(
    model="deepseek/deepseek-v4.1-flash",
    messages=[{"role": "user", "content": "Hello!"}],
)

print(response.choices[0].message.content)

What it costs and how fast it runs

You pay only for the tokens you use. Speed figures come from our own status checks.

Token pricing

DeepSeek V4.1 Flash: token prices in US dollars per million tokens
USD per million tokens
Input$0.50
Cached input$0.13
Output$1.50

No base fee. Caching applies only where listed. Check the model's processing region before sending data.

Speed on Lyceum

DeepSeek V4.1 Flash: median speed in our status checks over the last 7 days
Last 7 days
Time to first token254 ms
Output speed85 tokens/s

Median over the last 7 days from our status checks: short requests. Measured 6 October 2026, 09:41 UTC. Reasoning tokens count towards both figures. Live status and availability

The model in detail

Checked against the sources below. Anything they don't state is left out.

Capabilities

Input and output
Text and images in, text out
Context window
1M tokens
Tool calling
Supported

Architecture and licence

Total parameters
552B backbone, plus 196B of Engram memory
Active parameters
8B per token in prefill, 16B in decode
Architecture
Mixture of experts
Licence
MIT
Released
10 September 2026

Availability

API model ID
deepseek/deepseek-v4.1-flash
Provider
DeepSeek
Processing region
EU-hosted
Sources

Build with DeepSeek V4.1 Flash