DEV Community

Rıdvan Tülünay (TulunaY)
Rıdvan Tülünay (TulunaY)

Posted on • Originally published at livchart.com

Set Up Local AI Dashboards With Ollama in Under 5 Minutes

#ai #llm #tooling #tutorial

Running AI models locally used to be complex. With Ollama, it's a few terminal commands.

In this guide, I'll walk through installing Ollama, downloading a business-capable AI model, and connecting it to LivChart for AI-powered dashboard generation — all running locally, no cloud dependency.

Step 1: Install Ollama

macOS:

brew install ollama
Enter fullscreen mode Exit fullscreen mode

Linux:

curl -fsSL https://ollama.com/install.sh | sh
Enter fullscreen mode Exit fullscreen mode

Windows: Download from ollama.com

After installation, start the Ollama server:

ollama serve
Enter fullscreen mode Exit fullscreen mode

Step 2: Download a Model

For business analytics, I recommend starting with Qwen2.5 7B — it handles multilingual prompts well (including Turkish) and performs reliably for chart generation.

ollama pull qwen2.5:7b
Enter fullscreen mode Exit fullscreen mode

This downloads approximately 4.7 GB. Other good options:

Model Size Best For RAM Required
Qwen2.5 7B 4.7 GB Multilingual analytics 8 GB
Llama 3.1 8B 4.9 GB High-accuracy charts 16 GB
Gemma 4 E2B 1.6 GB Fast interactive use 8 GB
Mistral 7B 4.1 GB Lightweight deployment 8 GB

Step 3: Test Your Model 

ollama run qwen2.5:7b
Enter fullscreen mode Exit fullscreen mode

Try a business prompt:

Show me a bar chart of monthly revenue by region for Q1 2026
Enter fullscreen mode Exit fullscreen mode

If the model responds with structured output, it's working.

Step 4: Connect to LivChart

  1. Open LivChart and go to Settings → AI Configuration
  2. Set the AI provider to Ollama
  3. Set the endpoint: http://localhost:11434 (default Ollama port)
  4. Select your model: qwen2.5:7b
  5. Click Test Connection

If the connection succeeds, you're ready to use AI-powered chart generation.

Step 5: Create Your First AI Dashboard

  1. Import your data (Excel, CSV, or SQL database)
  2. Open the AI Chart Wizard
  3. Describe what you want: "Show me quarterly revenue comparison by product category"
  4. The AI generates the chart
  5. Refine if needed, then add to your dashboard

Common Problems and Solutions

"Connection refused" — Make sure Ollama is running (ollama serve). Check that port 11434 is not blocked by firewall.

"Model not found" — Run ollama list to see downloaded models. If empty, pull a model first.

"Slow responses" — Try a smaller model (Gemma 4 E2B) or enable GPU acceleration. On Apple Silicon, Ollama uses Metal by default.

"Incorrect charts" — Try rephrasing your prompt with more specific terms. Include chart type explicitly: "Create a line chart showing..."

Hardware Recommendations

Setup RAM GPU Models
Entry-level 16 GB Not required Gemma 4 E2B, Mistral 7B
Mid-range 32 GB Recommended Qwen2.5 7B, Llama 3.1 8B
Enterprise 64 GB+ Multi-GPU Large models, concurrent users

Why Local AI for Dashboards?

  • Privacy: Business data never leaves your machine
  • Speed: No network latency, models respond locally
  • Compliance: KVKK and GDPR requirements are simpler when data stays internal
  • Cost: No per-request API charges after initial hardware investment
  • Offline: Works without internet connection

The setup takes under 5 minutes. The benefits last much longer.

Try it yourself — download LivChart and connect Ollama for local AI-powered analytics.

Top comments (0)