DeepSeek V4 Flash

deepseek

deepseek/deepseek-v4-flash

Open in Console

DeepSeek V4 Flash reasoning model. Low-cost official DeepSeek Anthropic-compatible route.

Context

200K

Max output

66K

Tools

MINIMAL

Reasoning

Supported

Pricing

Lane	Per 1M tokens
Input	$0.14
Output	$0.28
Cache read	$0.00

Billed per token. No minimums, no per-request fees. Caching applies on supported providers; misses fall back to standard input/output rates.

Reasoning effort

offlowmedium · defaulthigh

Routed providers

anthropic

Requests are automatically routed to the highest-priority healthy upstream. Failures fall back to the next route transparently.

Quickstart

from openai import OpenAI

client = OpenAI(
    base_url="https://api.vecbase.com/v1",
    api_key="sk-vbc-...",
)

response = client.chat.completions.create(
    model="deepseek/deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Ready to ship?

Create a project, mint an API key, and call this model from any OpenAI-compatible client.

Get started free