Use NeverMetered with OpenClaw

The open-source, always-on personal AI agent, using a custom OpenAI-compatible provider.

OpenAI Chat

Model

qwen3.8-27b

128K context on the free plan65,536 max output tokens on the free planReplace YOUR_API_KEY with your key.

1
Create an API key
Create a key in API keys and name it after the tool, for example “OpenClaw”. A dedicated key per tool lets you see its usage separately and give it its own parallel-stream cap. The key is shown once; the snippets below use YOUR_API_KEY where it goes.
2
Install OpenClaw
Already have OpenClaw? Skip this step. Open a terminal first: on macOS press ⌘ Space, type Terminal and press Return; on Windows open the Start menu, type PowerShell and press Enter. It needs Node 24.16 or newer.
```
curl -fsSL https://openclaw.ai/install.sh | bash
```

Add the gateway to OpenClaw

One paste. It adds the gateway as a provider in ~/.openclaw/openclaw.json (merged into your config, nothing else changes), keeps your key in that file's env section, and makes the model OpenClaw's default. To go back later, run openclaw models set with your previous model.

openclaw config patch --stdin <<'EOF'
{
  "env": {
    "vars": {
      "NEVERMETERED_API_KEY": "YOUR_API_KEY"
    }
  },
  "models": {
    "providers": {
      "nevermetered": {
        "baseUrl": "https://api.nevermetered.com/v1",
        "apiKey": "${NEVERMETERED_API_KEY}",
        "api": "openai-completions",
        "models": [
          {
            "id": "qwen3.8-27b",
            "name": "qwen3.8-27b",
            "input": [
              "text"
            ],
            "contextWindow": 131072,
            "maxTokens": 65536
          }
        ]
      }
    }
  }
}
EOF
[ $? -eq 0 ] && openclaw models set "nevermetered/qwen3.8-27b"

4
If OpenClaw's Gateway service is running (optional)
A running Gateway reads its settings when it starts, so restart it once to pick up the new provider. The check below doesn't need it.
```
openclaw gateway restart
```
5
Verify it works
Run a single message. --local runs the agent in this terminal, so OpenClaw's Gateway doesn't have to be running. The request shows up in Requests within a few seconds, with its input, cached and output tokens.
```
openclaw agent --local -m "Reply with the word ready"
```

Tips

contextWindow and maxTokens in the model entry are your plan's limits. Without them OpenClaw assumes 128K of context and 8,192 output tokens.
Your key is stored under env.vars.NEVERMETERED_API_KEY in ~/.openclaw/openclaw.json and the provider refers to it as ${NEVERMETERED_API_KEY}, so the provider block itself has no key in it. Run the paste again with a new key to replace it.
openclaw models list --provider nevermetered shows the model with a default tag once it's selected.
Each OpenClaw message is one request, and OpenClaw sends its tool list with it, so requests are larger than a plain chat message.

Troubleshooting

authentication failed, HTTP 401: The key was wrong or has been revoked. Paste the block again with a current key from API keys. After an authentication failure OpenClaw pauses the provider for about a minute, so wait a moment before retrying.
Model override "nevermetered/…" is not allowed … modelPolicy.allow: Your OpenClaw config pins the models it may use. Run openclaw models set "nevermetered/qwen3.8-27b" to select the model, or add nevermetered/* to agents.defaults.modelPolicy.allow in ~/.openclaw/openclaw.json.
OpenClaw still answers with your old model: Run openclaw models set "nevermetered/qwen3.8-27b" and check that openclaw models list --provider nevermetered shows default. If OpenClaw's Gateway service is running, restart it with openclaw gateway restart.
Config validation failed: The paste was cut off or edited. Copy the whole block again: openclaw config patch checks the settings before writing anything, so a failed paste leaves your config unchanged.
404 model_not_found: Use a model id exactly as GET /v1/models lists it, for example qwen3.8-27b. Plans can restrict which models a key may call.
400 context_length_exceeded: The conversation outgrew the 128K context window (input + output). Let the tool compact or start a fresh session, or get a larger window on your plan.
429 concurrency_limit_exceeded: All your parallel streams are busy and your queue allowance is full. Lower the tool's parallelism (subagents, batch size), or get more streams on your plan.
Answers stop mid-sentence: The response hit your plan's output cap of 65,536 tokens (finish reason length or max_tokens). Ask for shorter steps, or upgrade for a higher cap.
403 email_verification_required: Free requests need a verified email. Click the link in your verification email (you can resend it from the dashboard), or upgrade.
429 daily_quota_exceeded: You've used today's free requests; agents send one request per step, so they go quickly. The allowance resets at 00:00 UTC, and paid plans have no daily limit.
503 / 504 request_queue_timeout: The service stayed busy longer than the queue timeout and nothing was generated. Retry with backoff, and give the tool a generous request timeout.
503 service_unavailable: No GPU capacity is online at the moment. It's temporary and your setup is fine: retry in a little while, and check the status page.

Use NeverMetered with OpenClaw

Create an API key

Install OpenClaw

Add the gateway to OpenClaw

If OpenClaw's Gateway service is running (optional)

Verify it works

Troubleshooting