DeepSeek API

BeatAPI provides free and paid DeepSeek API routes through the same API key. The free route supports Responses and Chat Completions; paid models use the formats documented in the Text API.

The old deepseek-v4-flash-vision-exp route is absent from the current BeatAPI catalogue as checked on September 30, 2026. Do not treat it as an automatic alias for V4.1 Flash. For text tasks, explicitly select deepseek-v4.1-flash; for image tasks, confirm a replacement model supports image input. Migration notice.

Free DeepSeek API

BeatAPI provides free DeepSeek V4.1 Flash API access through deepseek-v4.1-flash-free. Input and output cost $0. No top-up is required. Create an API key in the default auto group and select the free model ID explicitly.

• Before your first top-up: 1 successful request per minute per account.
• After any top-up: up to 10 successful requests per minute per account.
• Free calls are counted separately from the paid rate limit. These are frequency limits, not concurrent-request counts. Exceeding the limit returns HTTP 429.
• Free availability may change and the route may be withdrawn. Check current availability before relying on it.

Routes and first request

Base URL: https://api.beatapi.io/v1. Authenticate with Authorization: Bearer <BEATAPI_API_KEY>.

Use POST /v1/responses or POST /v1/chat/completions. The published free-model contract lists these two formats.

curl https://api.beatapi.io/v1/responses \
  -H "Authorization: Bearer $BEATAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4.1-flash-free","input":"Explain this system clearly.","max_output_tokens":256}'

For Chat Completions, send {"model":"deepseek-v4.1-flash-free","messages":[{"role":"user","content":"Explain this system clearly."}],"max_tokens":256} to https://api.beatapi.io/v1/chat/completions with the same headers.

From free to paid

Any top-up increases the free-model limit to up to 10 successful requests/min. It does not change your model or turn free calls into paid calls. To use the paid route, explicitly change model to deepseek-v4.1-flash; keep the key and base URL. Paid calls use the shared paid account allowance and bill actual token usage at the published prices.

Free DeepSeek API setup and pricing.

OpenAI Chat Completions

POST https://api.beatapi.io/v1/chat/completions

import OpenAI from "openai";
const client = new OpenAI({
  apiKey: process.env.BEATAPI_API_KEY,
  baseURL: "https://api.beatapi.io/v1"
});
const completion = await client.chat.completions.create({
  model: "deepseek-v4-pro-0813",
  max_tokens: 2000,
  messages: [
    { role: "user", content: "Design a resilient webhook retry strategy for a payments API." }
  ]
});
console.log(completion.choices[0].message.content);

curl --request POST \
  --url https://api.beatapi.io/v1/chat/completions \
  --header 'Authorization: Bearer <BEATAPI_API_KEY>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "deepseek-v4-pro-0813",
  "messages": [
    {
      "role": "user",
      "content": "Design a resilient webhook retry strategy for a payments API."
    }
  ]
}'

High-cache (HC) models

deepseek-v4-pro-0813-hc and deepseek-v4.1-flash-hc are the same two models, served on a high-cache line. When requests share a prompt prefix — a long system prompt, a tool list, a conversation that keeps growing — the prefix is read from cache from the second request on and billed at the cached-input rate; in our tests 95% or more of a repeated prefix was billed that way. The regular line costs less per token but does not guarantee a cache hit. Pick an HC model when most of each prompt is a prefix you send again and again; for prompts with little repetition the regular model is cheaper. Requests are identical to the regular models’ — only the model ID changes.

Rates by time of day

The three regular paid models — deepseek-v4-pro-0813, deepseek-v4.1-flash and deepseek-v4-flash-0731 — bill a single rate at every hour of every day. The weekday peak window DeepSeek publishes is not applied to them: the price on the pricing page is what a request costs at 10:00 on a Tuesday and at 03:00 on a Sunday alike.

The two HC models do have a peak window. Every rate doubles on weekdays 09:00–12:00 and 14:00–18:00 Beijing time (GMT+8) — input, cached input and output alike. Outside those hours, and all day on Saturday and Sunday, they bill the off-peak rate, which is the one the pricing page shows.

The three regular models reason by default and report the thinking budget under usage.completion_tokens_details.reasoning_tokens, which is counted in completion_tokens and billed as output.

Other SDK formats

The paid models also answer on /v1/responses, /v1/messages and the Gemini-compatible endpoint. The request and response shapes are identical for every text model, so they are documented once on the Text API page rather than repeated here.

Authentication and availability

One BeatAPI key reaches every model on this page — there is no per-vendor key, no per-vendor base URL, and no separate signup. Send Authorization: Bearer <BEATAPI_API_KEY> against https://api.beatapi.io.

Free versions

The two Flash models each have a free version with its own model ID, called on /v1/chat/completions and /v1/responses. deepseek-v4-pro-0813 and the HC models have none.