11 replies

Ollama disappeared from onboarding options after reinstall

✅Solved

I think I screwed something up and just need a kick in the right direction.

I have Ollama with ⁨

glm-4.7-flash

glm-4.7-flash

⁩. Running ⁨

ollama run glm-4.7-flash

ollama run glm-4.7-flash

⁩ works as expected, and I can get output from chats, so I know it's working right.

The first time I tried to install Openclaw with ⁨

curl -fsSL https://openclaw.bot/install.sh | bash

curl -fsSL https://openclaw.bot/install.sh | bash

⁩ Ollama simply showed up in the model list after I skipped the Model/Auth provider step. However, the installation failed due to issues with my NPM installation and a bug with the Google Chat plugin.

I uninstalled Openclaw and stopped the service.
⁨

$ systemctl stop --user openclaw-gateway
$ npm -g uninstall openclaw clawdhub
removed 695 packages in 849ms

$ systemctl stop --user openclaw-gateway
$ npm -g uninstall openclaw clawdhub
removed 695 packages in 849ms

⁩
Just for good measure, I restarted the system as well.

Then, I attempted to reinstall from scratch. Ollama no longer appears in the list of model providers.
⁨

◇  Model/auth provider
│  Skip for now
│
◆  Filter models by provider
│  ● All providers
│  ○ amazon-bedrock
│  ○ anthropic
│  ○ azure-openai-responses
│  ○ cerebras
│  ○ github-copilot
│  ○ google
│  ○ google-antigravity
│  ○ google-gemini-cli
│  ○ google-vertex
│  ○ groq
│  ○ huggingface
│  ○ kimi-coding
│  ○ minimax
│  ○ minimax-cn
│  ○ mistral
│  ○ openai
│  ○ openai-codex
│  ○ opencode
│  ○ openrouter
│  ○ vercel-ai-gateway
│  ○ xai
│  ○ zai

◇  Model/auth provider
│  Skip for now
│
◆  Filter models by provider
│  ● All providers
│  ○ amazon-bedrock
│  ○ anthropic
│  ○ azure-openai-responses
│  ○ cerebras
│  ○ github-copilot
│  ○ google
│  ○ google-antigravity
│  ○ google-gemini-cli
│  ○ google-vertex
│  ○ groq
│  ○ huggingface
│  ○ kimi-coding
│  ○ minimax
│  ○ minimax-cn
│  ○ mistral
│  ○ openai
│  ○ openai-codex
│  ○ opencode
│  ○ openrouter
│  ○ vercel-ai-gateway
│  ○ xai
│  ○ zai

⁩

Attempting to set the model name manually gives an error
⁨

Model not found: ollama/glm-4.7-flash. Update agents.defaults.model or run /models list.
No auth configured for provider "ollama". The agent may fail until credentials are added.

Model not found: ollama/glm-4.7-flash. Update agents.defaults.model or run /models list.
No auth configured for provider "ollama". The agent may fail until credentials are added.

⁩
And, of course, hatching fails. ⁨

ollama ps

ollama ps

⁩ shows no chats running during the hatching process.

Surely, somewhere, something got misconfigured or reconfigured wrong. What's a good way to fix this, or at least a way to completely clean Openclaw off my system and start over?

Solution

If anyone comes across this in the future, the solution was to make Ollama generate some default settings like this:
⁨⁨

ollama launch clawdbot --config

ollama launch clawdbot --config

⁩⁩
This will make a folder in ⁨⁨

.clawdbot

.clawdbot

⁩⁩ that has your Ollama information already filled in, so you don't have to do it manually.

When you run the openclaw installer with no ⁨⁨

.openclaw

.openclaw

⁩⁩ folder, it will migrate the folder Ollama just made to this folder. Afterwards it should work just fine. You should delete ⁨⁨

.clawdbot

.clawdbot

⁩⁩ afterwards or the doctor will complain.

Jump to solution

FotC�

Friends of the Crustacean 🦞🤝•2w ago•

11 replies

Orangestar

Ollama disappeared from onboarding options after reinstall

✅Solved

I think I screwed something up and just need a kick in the right direction.

I have Ollama with ⁨

glm-4.7-flash

glm-4.7-flash

⁩. Running ⁨

ollama run glm-4.7-flash

ollama run glm-4.7-flash

⁩ works as expected, and I can get output from chats, so I know it's working right.

The first time I tried to install Openclaw with ⁨

curl -fsSL https://openclaw.bot/install.sh | bash

curl -fsSL https://openclaw.bot/install.sh | bash

$ systemctl stop --user openclaw-gateway
$ npm -g uninstall openclaw clawdhub
removed 695 packages in 849ms

$ systemctl stop --user openclaw-gateway
$ npm -g uninstall openclaw clawdhub
removed 695 packages in 849ms

⁩
Just for good measure, I restarted the system as well.

Then, I attempted to reinstall from scratch. Ollama no longer appears in the list of model providers.
⁨

◇  Model/auth provider
│  Skip for now
│
◆  Filter models by provider
│  ● All providers
│  ○ amazon-bedrock
│  ○ anthropic
│  ○ azure-openai-responses
│  ○ cerebras
│  ○ github-copilot
│  ○ google
│  ○ google-antigravity
│  ○ google-gemini-cli
│  ○ google-vertex
│  ○ groq
│  ○ huggingface
│  ○ kimi-coding
│  ○ minimax
│  ○ minimax-cn
│  ○ mistral
│  ○ openai
│  ○ openai-codex
│  ○ opencode
│  ○ openrouter
│  ○ vercel-ai-gateway
│  ○ xai
│  ○ zai

◇  Model/auth provider
│  Skip for now
│
◆  Filter models by provider
│  ● All providers
│  ○ amazon-bedrock
│  ○ anthropic
│  ○ azure-openai-responses
│  ○ cerebras
│  ○ github-copilot
│  ○ google
│  ○ google-antigravity
│  ○ google-gemini-cli
│  ○ google-vertex
│  ○ groq
│  ○ huggingface
│  ○ kimi-coding
│  ○ minimax
│  ○ minimax-cn
│  ○ mistral
│  ○ openai
│  ○ openai-codex
│  ○ opencode
│  ○ openrouter
│  ○ vercel-ai-gateway
│  ○ xai
│  ○ zai

⁩

Attempting to set the model name manually gives an error
⁨

Model not found: ollama/glm-4.7-flash. Update agents.defaults.model or run /models list.
No auth configured for provider "ollama". The agent may fail until credentials are added.

Model not found: ollama/glm-4.7-flash. Update agents.defaults.model or run /models list.
No auth configured for provider "ollama". The agent may fail until credentials are added.

⁩
And, of course, hatching fails. ⁨

ollama ps

ollama ps

Solution

If anyone comes across this in the future, the solution was to make Ollama generate some default settings like this:
⁨⁨

ollama launch clawdbot --config

ollama launch clawdbot --config

⁩⁩
This will make a folder in ⁨⁨

.clawdbot

.clawdbot

⁩⁩ that has your Ollama information already filled in, so you don't have to do it manually.

When you run the openclaw installer with no ⁨⁨

.openclaw

.openclaw

⁩⁩ folder, it will migrate the folder Ollama just made to this folder. Afterwards it should work just fine. You should delete ⁨⁨

.clawdbot

.clawdbot

⁩⁩ afterwards or the doctor will complain.

Jump to solution

Ollama disappeared from onboarding options after reinstall

Ollama disappeared from onboarding options after reinstall

Similar Threads

Similar Threads

Similar Threads