Page 1 of 2

How to Configure LLM for OpenClaw in TOS

Posted: 01 Apr 2026, 16:54
by hurhan001
Important Note: Before doing this, please confirm that you have installed the latest version of the OpenClaw application in the App Center.
The required TOS compatible version is 7.0.0746


How to Configure Large Language Models?
I. Popular Large Language Models (Using OpenAI as an Example)
1.Go to the official OpenAI console page [ https://platform.openai.com/ ]
and log in to your account. After logging in, open the [ API key ] page and create an API key.
Image
2.After creating it, you will receive the [ API key ] ( You can purchase the plan through the official link: https://platform.openai.com/settings/or ... g/overview).
Image
Image


II. Configure Large Language Models
1.Open the model configuration page
On the TNAS device, use the super administrator account to open the [ OpenClaw ] app, and directly click [ Configure Large Language Model ] to enter the model configuration page.
Image

2. Select a model service provider
Choose the corresponding model provider from the service provider list.
For example:
● OpenAI
● Anthropic
● Gemini
If your desired service provider is not listed, or if you need to access the model service through a proxy server, please select [ Custom Provider ].
Image
Image

3. Fill in the model information
Using OpenAI as an example:
● API Key: Enter the API Key you just created
● Model: Select the model you wish to use (e.g., gpt-5.4)
After filling in the information, click [ Save Configuration ].
Once the system prompts that the save is successful, you are ready to start using it.
Image


Ⅲ.Verifying Successful Model Configuration
Open the [ OpenClaw ] application and type ‘ hello ’ in the dialog box.
1. If OpenClaw responds properly, the large language model configuration is successful.
Image
2. If the response content looks like the image below, it means the configuration failed. You can retry the operations in step two to ensure they are executed correctly.
Image
3. If the error message appears as shown, it means there are insufficient tokens in your plan. Please visit the OpenAI official website [https://platform.openai.com/settings/or ... g/overview] to purchase more tokens.
Image

Re: How to Configure LLM for OpenClaw in TOS

Posted: 22 Jul 2026, 00:48
by Supr4
I dont see google gemini as option to configure. How?

Re: How to Configure LLM for OpenClaw in TOS

Posted: 22 Jul 2026, 12:09
by TMzethar
Supr4 wrote: ↑22 Jul 2026, 00:48 I dont see google gemini as option to configure. How?
Please select "Custom Provider" and enter the API information you obtained from Gemini.

Re: How to Configure LLM for OpenClaw in TOS

Posted: 21 Sep 2026, 16:46
by TheIntegrator
I'm getting

Connection test failed!
The failure may be caused by the following reasons
1. Network connection interrupted
2. Incorrect API Base URL or API Key input
3. Incorrect large language model ID input
4. The plan has expired, the call quota has been used up, or the usage limit has been reached
5. API protocol configuration error

When I try to add Deepseek. All fields appear to be fllled correctly

Re: How to Configure LLM for OpenClaw in TOS

Posted: 21 Sep 2026, 18:21
by EriChan
TheIntegrator wrote: ↑21 Sep 2026, 16:46
To accurately identify the issue, please contact the online support team via the following link for assistance in troubleshooting:
https://salesiq.zohopublic.com/signatur ... 97d0be8892

Re: How to Configure LLM for OpenClaw in TOS

Posted: 25 Sep 2026, 23:27
by rfbjr
Help. I have openclaw installed and ollama installed on a VMS (on different NAS).
Has anyone got a setup like this working correctly?
I picked ollama since it's completely free and I wanted to try something before committing to a paid LLM.
Since it's ollama it's a custom setup, mainly having fun with the json file.

Re: How to Configure LLM for OpenClaw in TOS

Posted: 25 Sep 2026, 23:58
by TMroy
You should give DeepSeek V4.1 Flash a try. The cloud LLM is way stronger than your local setup. And it's so cheap that even a nobody like me can afford it easily.

Re: How to Configure LLM for OpenClaw in TOS

Posted: 26 Sep 2026, 00:25
by rfbjr
Thank you I will check it out.

Re: How to Configure LLM for OpenClaw in TOS

Posted: 26 Sep 2026, 01:53
by rfbjr
Up and running already. Thanks for suggestion.
Two things.
The button on the bottom, I think should just be "OK", "Back" seemed confusing, meaning the config/changes save correctly.

Image

Also in my case the openclaw console/gui has problems displaying correctly on Chrome. Granted I have Malwarebytes installed but totally disabled on the NAS GUI. If I open the webgui for the NAS on Firefox and launch the openclaw console, it displays correctly. Granted Malwarebytes is not installed on Firefox.

Re: How to Configure LLM for OpenClaw in TOS

Posted: 26 Sep 2026, 02:23
by rfbjr
Okay I'm up and running. OpenClaw is installed right now on only one of my devices. Can that instance or whatever you want to call it report or query on my other devices? If not, then OpenClaw needs to be installed and would each need it's own ApiKey?