DeepSeek: DeepSeek V4.1 Flash

Pricing

What you pay per 1 million tokens. Cached prompt tokens are billed at the cache-read price when the upstream reports them; otherwise at the input price.

Worked example: a request with 10,000 prompt tokens and 1,000 completion tokens costs $0.00525. While it runs, the request holds the maximum it could cost (based on max_tokens), then settles at the actual cost. How pricing works

Capabilities

Only capabilities we have verified end to end through the API are listed.

Features

StreamingTool callingStructured outputsReasoning

Input

TextImage

Output

Text

Context window

1,048,576 tokens

Max output

393,216 tokens

Performance on AI Models Router

Measured from real requests through our API in the last 7 days, all customers. A figure appears once it rests on at least 50 requests; until then we show how many there are, never an estimate.

Updated every 10 minutes · last computed 19:55 UTC

Usage · last 30 days

API

1. 1Get your API key

   Create a key on the API keys page and export it:

   Shell

   export AMR_API_KEY=sk-amr-v1-...

2. 2Make your first request

   Any OpenAI-compatible client works. The model slug is deepseek/deepseek-v4.1-flash.

   PythonTypeScriptcURL

   from openai import OpenAI
   
   client = OpenAI(
       base_url="https://api.aimodelsrouter.ai/v1",
       api_key="sk-amr-v1-...",  # or os.environ["AMR_API_KEY"]
   )
   
   completion = client.chat.completions.create(
       model="deepseek/deepseek-v4.1-flash",
       messages=[{"role": "user", "content": "What is the meaning of life?"}],
   )
   print(completion.choices[0].message.content)

3. 3Stream the response

   Set stream: true. The final chunk carries usage and cost.

   PythonTypeScriptcURL

   stream = client.chat.completions.create(
       model="deepseek/deepseek-v4.1-flash",
       messages=[{"role": "user", "content": "Write a haiku about routers."}],
       stream=True,
       stream_options={"include_usage": True},
   )
   for chunk in stream:
       if chunk.choices and chunk.choices[0].delta.content:
           print(chunk.choices[0].delta.content, end="", flush=True)
       if chunk.usage:  # final chunk: one choice with an empty delta, usage present
           print("\ncost (USD):", chunk.usage.cost)