Which config file does DeepSeek Harness use?

Use ~/.dsh/cordis.patch.yml for the recommended WaveSpeed AI setup. The configuration points DeepSeek Harness to endpoint /v1/chat/completions.

Which protocol should DeepSeek Harness use?

DeepSeek Harness should use OpenAI-compatible chat completions. This page uses endpoint /v1/chat/completions to match that protocol.

How do I verify DeepSeek Harness is connected?

Run DeepSeek Harness locally and open WaveSpeed Logs. Successful requests should appear with the selected model and usage details.

Agent Setup

Connect DeepSeek Harness to WaveSpeed LLM API

Copy the recommended ~/.dsh/cordis.patch.yml setup for DeepSeek Harness, route requests to /v1/chat/completions, and verify local agent calls in WaveSpeed Logs.

API Key Test Model

DeepSeek's agent harness; add WaveSpeed as a provider route.

Config file

~/.dsh/cordis.patch.yml

Protocol

OpenAI-compatible chat completions

Endpoint

/v1/chat/completions

Manual setup

Create a WaveSpeed access key

Export the required key locally

Paste config into~/.dsh/cordis.patch.yml

Run dsh --profile headless and check Logs

Model

OpenAI-compatible chat completions

Usage appears in Dashboard and Logs

Switch models without rewriting agent code

One-command setup

Configures DeepSeek Harness, checks that it works, and backs up your previous settings. Add --undo to restore them.

1
Install the WaveSpeed CLI
npm install -g @wavespeed/cli
2
Sign in
wavespeed login
3
Configure DeepSeek Harness
wavespeed setup dsh

No Node.js? See other install options

Manual setup

Edit ~/.dsh/cordis.patch.yml

# File: ~/.dsh/cordis.patch.yml
- id: llm-pi-ai
  config:
    providers:
      wavespeed:
        apiKeyEnv: WAVESPEED_API_KEY
        api: openai-completions
        baseURL: https://llm.wavespeed.ai/v1
        models:
          - id: deepseek/deepseek-v4.1-flash-1017
- id: agent-default-model
  config:
    provider: wavespeed
    model: deepseek/deepseek-v4.1-flash-1017

Do not replace the variable name

WAVESPEED_API_KEY is the environment variable name that the agent reads. Keep this name in the config file, and put your real API key in your shell instead.

Temporary for this terminal

Works until you close this terminal window.

export WAVESPEED_API_KEY="wsk_live_..."

Keep it after restart

Writes the key into your shell profile so new terminals can read it.

echo 'export WAVESPEED_API_KEY="wsk_live_..."' >> ~/.zshrc
source ~/.zshrc

Use DeepSeek Harness with WaveSpeed LLM API

This guide shows how to connect DeepSeek Harness to WaveSpeed AI with ~/.dsh/cordis.patch.yml, OpenAI-compatible chat completions, /v1/chat/completions, base URL https://llm.wavespeed.ai/v1, and default model deepseek/deepseek-v4.1-flash-1017. Copy the configuration above, export your API key, run the agent, then verify requests in Logs.

Connect DeepSeek Harness to WaveSpeed LLM API

Manual setup

Use DeepSeek Harness with WaveSpeed LLM API

What this DeepSeek Harness setup configures

Troubleshooting

Can this DeepSeek Harness guide show up in search?