curl -fsSL https://ollama.com/install.sh | sh
Can you provide more details on your issue? Logs and commands you run would be much appreciated.
Vibe Coding in VS Code
OLLAMA INSTALL
OPENCLAW
“ollama”: {
“baseUrl”: “http://192.168.2.94:11434/v1”,
“apiKey”: “ollama-local”,
“api”: “openai-completions”,
“models”: [
{
“id”: “glm-4.7-flash:bf16”,
“name”: “GLM-4.7 Flash BF16”,
“reasoning”: false,
“input”: [
“text”
],
“cost”: {
“input”: 0,
“output”: 0,
“cacheRead”: 0,
“cacheWrite”: 0
},
“contextWindow”: 131072,
“maxTokens”: 8192
}
]
}
Respectfully, using openclaw, generally, is way beyond the scope of this forum and you are unlikely to get much handholding. Setting up Ollama is straightforward with this playbook: Open WebUI with Ollama | DGX Spark
There is nothing specific to Spark about openclaw. Try doing a Google search for the proper syntax for openclaw.json.
1 Like
Surprisingly there is a dedicated guide for this.
OP - this is all you should need.
Anything beyond that you’d be on your own.
2 Likes
Wow! Did not know.
You could/should run llama.cpp or vLLM and set the
“baseUrl”: “http://localhost:1234/v1” to your local IP … it’s more pure, more honest ;)
@hongde
Have you enabled the following settings? When I use ollama glm-4.7, it has memory capability. You might also try updating to the latest version of OpenClaw to see if that helps.
