My Local LLM Setup

calendar Feb 28, 2026 · 1 min read · ollama llm open-web-ui claude code  ·
Share on: linkedin copy

Install Ollama

Execute following in terminal

1curl -fsSL https://ollama.com/install.sh | sh

Once ollama is installed pull a model using

1ollama run ministral-3:8b

Ollama Model

Install OpenWebUi

1docker run -d -p 3000:8080 --network host \n
2 -v open-webui:/app/backend/data \n 
3 --name open-webui ghcr.io/open-webui/open-webui:main

Open Web Ui
Navigate to http://localhost:8080 enter default credentials and you have completed setup.

Install claude-code

1curl -fsSL https://claude.ai/install.sh | bash

Once claude code is installed setup below file and import in .zshrc

1touch /home/ashish/.claude/claude-config.env

Paste the following content

1export ANTHROPIC_BASE_URL="http://gpu.silentsudo.in:11434"
2export ANTHROPIC_AUTH_TOKEN="ollama"
3export ANTHROPIC_API_KEY="ollama" # Required field, but Ollama ignores the actual value
4export ANTHROPIC_MODEL="ministral-3:8b" # Name of the ollama model you can use to see list `ollama list`
5export CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC=1 # disable analytics

then execute source .zshrc

Verify Installation and Connection with Ollama

Claude Code

comments powered by Disqus