OpenClaw

Checked against OpenClaw's docs on 2026-10-11

OpenClaw connects as a custom provider: one entry under models.providers in ~/.openclaw/openclaw.json, with the key in ~/.openclaw/.env.

Base URL: https://api.creeksidecompute.ai/v1
API: OpenAI Chat Completions
Model id: qwen/qwen3.8-27b, or qwen/qwen3.6-35b-a3b for speed, or meta/muse-glimmer-30b, which always reasons first
Context window: 131072 tokens, up to 65536 of them output
Key: Make one at creeksidecompute.ai/account

Set it up

Make a key. Sign in at creeksidecompute.ai/account with your email and make a key. Each request is paid from prepaid credit: top up by card from $10, or redeem a promotional code.
Store the key. Add this line to ~/.openclaw/.env:
~/.openclaw/.env
```
CREEKSIDE_API_KEY=<your key>
```

Add the provider to ~/.openclaw/openclaw.json, merged with what the file holds already:

~/.openclaw/openclaw.json

{
  models: {
    providers: {
      creekside: {
        baseUrl: "https://api.creeksidecompute.ai/v1",
        apiKey: "${CREEKSIDE_API_KEY}",
        api: "openai-completions",
        models: [{
          id: "qwen/qwen3.8-27b",
          name: "Qwen3.8 27B",
          reasoning: true,
          input: ["text"],
          contextWindow: 131072,
          maxTokens: 32768,
          compat: { thinkingFormat: "qwen-chat-template", supportsUsageInStreaming: true },
        }, {
          id: "qwen/qwen3.6-35b-a3b",
          name: "Qwen3.6 35B A3B",
          reasoning: true,
          input: ["text"],
          contextWindow: 131072,
          maxTokens: 32768,
          compat: { thinkingFormat: "qwen-chat-template", supportsUsageInStreaming: true },
        }, {
          id: "meta/muse-glimmer-30b",
          name: "Muse Glimmer 30B",
          reasoning: true,
          input: ["text", "image"],
          contextWindow: 131072,
          maxTokens: 32768,
          compat: { thinkingFormat: "qwen-chat-template", supportsUsageInStreaming: true },
        }],
      },
    },
  },
  agents: { defaults: { model: { primary: "creekside/qwen/qwen3.8-27b" } } },
}

name is required. contextWindow gives OpenClaw the real window; without it, OpenClaw assumes 200,000 tokens. The compat line is what lets OpenClaw switch thinking.

Check it and pick the model: openclaw config validate, then openclaw models set creekside/qwen/qwen3.8-27b. In a chat: /model creekside/qwen/qwen3.8-27b. For speed, use creekside/qwen/qwen3.6-35b-a3b instead, and for Muse Glimmer 30B, creekside/meta/muse-glimmer-30b. Keep the creekside/ prefix: on its own, qwen/… names a different provider.

Thinking

With reasoning: true, OpenClaw starts at medium thinking, which is on. /think off turns it off, and /think low, medium or high turn it on; agents.defaults.thinkingDefault sets the level for new sessions. The compat line makes OpenClaw send the switch as chat_template_kwargs.enable_thinking; without it, OpenClaw 2026.10.1 sends nothing and thinking stays off. /reasoning on shows the thinking text. Thinking tokens are billed as output.

Muse Glimmer 30B always reasons before it answers, so the switch sets how long, not whether: with /think off it reasons briefly, and with thinking on, at any level, it reasons at its own default, high.

What does not work

Images, on the Qwen models. They are listed for text only, so OpenClaw does not send attachments to them. Muse Glimmer 30B reads images, so its entry lists "image".
The onboarding defaults. openclaw onboard writes a 128,000-token window, a 4,096-token output limit and thinking off for a custom provider. Change them to the values above.
Cost display. Without a cost block, OpenClaw shows no spend. Your bill uses the prices in GET /v1/models.

Troubleshooting

401 invalid API key: The key is wrong, or OpenClaw did not get it. Check the line in ~/.openclaw/.env, then restart OpenClaw. To test the key alone, curl -s https://api.creeksidecompute.ai/v1/models -H "Authorization: Bearer $CREEKSIDE_API_KEY" returns the model list when the key is good.
404 model … is not served: The model id is wrong. Use qwen/qwen3.8-27b exactly, in lower case, with the qwen/ prefix. GET /v1/models lists every id we serve. In model references, write creekside/qwen/qwen3.8-27b: the provider, then the id.
400 maximum context length is 131072 tokens: The conversation is longer than the model's window. Keep contextWindow: 131072 so OpenClaw compacts in time, then start a new session.
402 or 429: A 402 means your prepaid credit does not cover the request: add credit at creeksidecompute.ai/account. A 429 means the key's rate, or the number of requests your account may run at once, is used up; agents that work in parallel reach that sooner. Wait and retry. The API reference lists the limits.

Other tools

Sources, read on 2026-10-11: custom providers, configuration, thinking.