x402.press DeepSeek
simple-inference

DeepSeek

DeepSeek AI inference without an API key or credit card. Pay per request in USDC or EURC on any of 7 chains. OpenAI-compatible chat completions endpoint.

inferencedeepseekllmaix402openai-compatible

What is DeepSeek?

OpenAI-compatible AI inference via x402 micropayments. Proxies requests to DeepSeek's API, supporting two models — deepseek-v4-flash (fast, cost-effective) and deepseek-v4-pro (higher capability for complex reasoning). No API keys, no accounts, no subscriptions. A drop-in replacement for OpenAI client libraries: point your existing OpenAI SDK at https://x402.press/deepseek/v1 and it works, with all OpenAI-compatible parameters (temperature, max_tokens, top_p, etc.) forwarded to the underlying model. The price depends on the model requested and the size of the prompt (a small prompt costs less than a large one), and is higher in DeepSeek's peak hours (01:00-04:00 and 06:00-10:00 UTC, Monday-Friday). The amount to pay is always returned with the payment request, so clients never have to compute it. A metered (`upto`) accept is also offered: authorize the prompt plus your `max_tokens` as a ceiling and settle the measured input and output tokens, never above that ceiling. Nothing in the metered path caps context or output.

Pay per call in USDC, EURC across Base, Arbitrum, Polygon, Avalanche, Ethereum, Robinhood and Arc via the x402 micropayment protocol — no subscriptions, no API keys, no wallets to configure for the caller.

Operations

inference

MethodEndpointDescriptionTier
POST/v1/chat/completionsOpenAI-compatible chat completions. Supports deepseek-v4-flash and deepseek-v4-pro. Additional OpenAI parameters (temperature, max_tokens, top_p, frequency_penalty, presence_penalty, stop, etc.) are forwarded. The quoted price follows the model requested and the prompt size, and is higher inside DeepSeek's peak windows (01:00-04:00 and 06:00-10:00 UTC, Monday-Friday).deepseek

Data sources

Examples

chat_completions

Ask a question via the OpenAI-compatible endpoint

{
  "model": "deepseek-v4-flash",
  "messages": [
    {
      "role": "user",
      "content": "What is the capital of France?"
    }
  ]
}