Something not connecting in default model

[gilded wind]

After the most recent update, prompts are not responding. I restarted the gateway, but as soon as I send a prompt, openclaw tries to switch off the default openai OAuth connection I use, and switches to a backup model, but does not reply.
michael@AInchor:~/csu-practice/week1$ openclaw models status

🦞 OpenClaw 2026.5.12 (f066dd2)
The only crab in your contacts you actually want to hear from. 🦞

Config : ~/.openclaw/openclaw.json
Agent dir : ~/.openclaw/agents/main/agent
Default : openai/gpt-5.4
Fallbacks (1) : ollama/gemma3:4b
Image model : -
Image fallbacks (0): -
Aliases (1) : gpt -> openai/gpt-5.4
Configured models (6): ollama/gemma3:4b, lmstudio/google/gemma-4-e4b, ollama/llama3.2:3b, ollama/qwen3:4b, ollama/granite4:3b, openai/gpt-5.4

Auth overview
Auth store : ~/.openclaw/agents/main/agent/auth-profiles.json
Shell env : off
Providers w/ OAuth/tokens (1): openai-codex (1)

• lmstudio effective=models.json:sk-lm-0B...Abgclc1O | models.json=sk-lm-0B...Abgclc1O | source=models.json: ~/.openclaw/agents/main/agent/models.json
• ollama effective=profiles:~/.openclaw/agents/main/agent/auth-profiles.json | profiles=1 (oauth=0, token=0, api_key=1) | ollama:default=marker(ollama-local) | models.json=OL...EY | source=models.json: ~/.openclaw/agents/main/agent/models.json | synthetic=plugin-owned | source=models.providers.ollama (synthetic local key)
• openai-codex effective=profiles:~/.openclaw/agents/main/agent/auth-profiles.json | profiles=1 (oauth=1, token=0, api_key=0) | openai-codex:[e-mail]=OAuth ([e-mail])

OAuth/token status

• openai-codex usage: 5h 99% left ⏱4h 59m · Week 98% left ⏱4d 15h
  • openai-codex:[e-mail] ([e-mail]) ok expires in 6d

[limpid ledge]

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

[wanton crown]

This status output has a pretty strong clue: your default is openai/gpt-5.4, but the only OAuth profile shown is for openai-codex.

Those are different providers in OpenClaw:

• openai/* = direct OpenAI Platform API key
• openai-codex/* = Codex/ChatGPT OAuth subscription route

So OpenClaw is likely trying openai/gpt-5.4, finding no matching OpenAI API-key auth, then falling through to your ollama/gemma3:4b fallback. I’d try:

openclaw models set openai-codex/gpt-5.4
openclaw models status
openclaw models status --probe --probe-provider openai-codex
openclaw gateway restart

If you intended to use direct OpenAI API access instead, keep openai/gpt-5.4 but add/configure an OpenAI API key. For Codex OAuth, the prefix should be openai-codex/.

[wanton crown]

Also run this once because recent versions explicitly check for stale Codex OAuth provider overrides:

openclaw doctor

Docs:
https://docs.openclaw.ai/providers/openai
https://docs.openclaw.ai/cli/models
https://docs.openclaw.ai/gateway/doctor

If it still hangs after changing the prefix, paste the relevant openclaw logs --follow lines from one failed prompt. The specific fallback reason will matter there.

[gilded wind]

That got it back, I'll mark this as answered, can you also tell me the cli to set the default thinking level to medium?

[wanton crown]

Yep. For the global default:

openclaw config set agents.defaults.thinkingDefault medium
openclaw gateway restart

For just the current chat/session, use:

/think medium

Docs: https://docs.openclaw.ai/tools/thinking

And thank you for marking it answered — that helps Answer Overflow index the fix for the next person.