How to add a DeepSeek API key to DeepSeek Harness

The default path, and the one DeepSeek tests first-party. DSH has a dedicated DeepSeek card in Settings → Models — you paste one key and you are done. No base URL, no model list, no protocol picker.

Before you start

You need a partially funded DeepSeek API account, an API key, and Node.js installed. If DSH itself is not running yet, do the install guide first.

Step 1 — create an API key

Sign in to DeepSeek API and create a key in its console. Copy it once — most providers never show it again.

Open DeepSeek API →

Step 2 — open Settings → Models

Start the harness if it is not running:

npx @deepseek-ai/dsh web

That serves the Web UI at http://127.0.0.1:3080 and opens it. Then go to Settings → Models. The DeepSeek card has a single API-key field.

Step 3 — paste the key and save

The key is written to $DSH_HOME/.credentials.yaml; settings keep only a reference to it. The page shows a redacted descriptor after saving, never the secret itself, and the field is write-only by design.

Step 4 — pick a model

Configured providers appear in the model picker. Selecting a model also makes it the default for new sessions. A session that has already sent a request keeps the model recorded in its own log — switching the default does not retroactively change it.

Step 5 — verify

Send any message in a new session. Model changes take effect on the next request without restarting the server, so there is nothing to reload.

Or: configure it in YAML

The Web UI writes the same document you can edit directly. The profile for a standard dsh web launch is web, so the file is $DSH_HOME/profiles/web/cordis.patch.yml. Adapters re-read it on the next request — no restart needed.

# $DSH_HOME/profiles/web/cordis.patch.yml
# The DeepSeek card writes this for you. Edit by hand only if you need to
# adjust a single model — for example:
- id: llm-deepseek
  config:
    modelOverrides:
      deepseek-chat:
        maxOutputTokens: 8192

For a signed-in DeepSeek account DSH offers no model editor at all, so modelOverrides is the only way to tune one model. Unset fields keep the listed values.

Export the variable referenced above in the shell that starts DSH:

export DSH_API_KEY="sk-..."

Troubleshooting

• 404 on every request — the base URL is wrong. It must end at the version segment (/v1), never at /chat/completions.
• 401 Unauthorized — key revoked, or pasted with trailing whitespace. Note that DSH's env-var detection does not validate the key.
• CLIENT_UPGRADE_REQUIRED — the model needs a newer harness. The error names the version to install.
• Model missing from the picker — a saved default pointing at a deleted provider makes the composer show Select model and block input until you pick another.
• Empty or truncated reply — usually the model id resolved to a different model than you expected. Add the id explicitly rather than relying on Fetch available models.

Cost and caveats

Pay per token, no subscription. On the order of $0.27 per million input tokens and $1.10 per million output tokens for deepseek-chat.

• A funded account is required — a few dollars covers weeks of normal use.
• Create keys in the platform console, not in DSH. DSH only stores the value.
• Keys are write-only in the UI: after saving, DSH shows a redacted descriptor, never the literal secret.

DSH also supports signing in with a DeepSeek account instead of a key. That path lists models from the account service and offers no per-model editor — use modelOverrides in the profile patch if you need to adjust a single model.

Plugins that pair well with this provider

Other providers

Source for every field above: the official DSH provider docs. If DSH changes a field name, that document wins over this page.