DeepSeek · Text

OpenClaw x DeepSeek

DeepSeek models are large language models tuned for reasoning, code, and agentic tool use. Through RunAPI they share a single API key with pay-as-you-go token billing, and are callable from both the OpenAI and Anthropic SDKs.

4 variants from $0.30 / 1M tokens Commercial OK
Get API Key View OpenClaw Docs

Prerequisite: npx runapi mcp install

Prompt

Prompt models

Use deepseek-flash when it matches the task: deepseek-flash.

You have access to RunAPI chat models for DeepSeek.

Available DeepSeek models:
- deepseek-flash: deepseek-flash
  chat endpoints: /v1/chat/completions, /v1/messages
- deepseek-v4-flash: Fast, low-cost reasoning; toggle thinking per request
  chat endpoints: /v1/chat/completions, /v1/messages
- deepseek-v4-flash-vision-exp: 1M context; image input for vision tasks; JPEG, PNG, GIF, and WebP
  chat endpoints: /v1/chat/completions, /v1/messages
- deepseek-v4-pro: Flagship reasoning and analysis for complex agentic work
  chat endpoints: /v1/messages, /v1/chat/completions

Use the model ID and chat endpoint that match the user's request.
Example request: Refactor this Python module for readability, explain each change, then add unit tests for the edge cases.
Verify the active model with: openclaw models status;openclaw models set runapi/<model-id>

Public Versions and Endpoints

Model ID Endpoints Price Catalog
deepseek-flash
/v1/chat/completions /v1/messages
$0.30 / 1M tokens Model detail
deepseek-v4-flash
/v1/chat/completions /v1/messages
$0.30 / 1M tokens Model detail
deepseek-v4-flash-vision-exp
/v1/chat/completions /v1/messages
$0.30 / 1M tokens Model detail
deepseek-v4-pro
/v1/messages /v1/chat/completions
$1.32 / 1M tokens Model detail

Verify

Poll until the task reaches a terminal status

Select <model-id> to generate verification commands.

Configuration

Guide endpoint: <endpoint>

Select <model-id> to generate a request with the endpoint's public input contract.
How it works

Get Started in 3 Steps

  1. Choose a model ID

    Select a public catalog model ID and review its endpoint and current starting price.

  2. Configure RunAPI

    Add the provider configuration for this agent.

  3. Verify the result

    Run the agent status and model-selection commands shown below.

What to Build with OpenClaw + DeepSeek

  • Budget-friendly model for high-volume apps

    Use DeepSeek for apps that process many requests a day, keeping token costs low without changing your OpenAI SDK code.

  • Reasoning tasks with visible chain of thought

    Enable thinking mode for math, logic, debugging, or planning, and show the model's reasoning in your interface.

  • Code generation and debugging

    Use DeepSeek for code generation, algorithm implementation, and debugging at a low cost per token.

Why Use DeepSeek Through RunAPI + OpenClaw

  • 4 variants, one API key

    Use one RunAPI connection to choose among the live model variants without changing your integration.

  • Clear usage pricing

    See current catalog pricing before you send a request, with no subscription or minimum spend required.

  • Direct responses

    Synchronous calls return the result in the same response, so your agent can use it immediately without task polling.

OpenClaw + DeepSeek Questions

Is DeepSeek a drop-in replacement for OpenAI models in OpenClaw?

Yes. DeepSeek supports the OpenAI chat completions format. Point OpenClaw at RunAPI, change the model to a DeepSeek version from this page, and keep the rest of your configuration.

How do thinking tokens affect cost?

When reasoning is enabled, DeepSeek generates extra thinking tokens that are billed as output tokens. They improve accuracy on hard tasks but increase total usage.

How do the DeepSeek versions differ?

Faster versions favor latency and cost, while larger versions favor quality on complex reasoning and coding. All share the same endpoint and parameters.

Which model ID should I use?

Choose a public model ID from the version table. Each ID exposes the endpoints shown for that version.

Does this guide configure a chat model?

Yes. This Model Line supports a client-facing LLM protocol.

Start using DeepSeek with OpenClaw

Get API Key Read Quickstart

Building with a team?

We're here to help with enterprise setup, integrations, and technical questions.

Contact Us