Local AI for Local Climates: Building an Offline Garden Planner with Gemma 3 & Ollama

Description and use

An offline garden planner uses a local Gemma 3 model via Ollama to provide season‑specific planting advice for Indian cities, avoiding generic frost‑based guidance.

Problem

Planting advice online assumes climates that do not match Indian monsoon, winter, and summer seasons, leading to irrelevant or incorrect guidance for local gardeners.

Outcome

The app delivers accurate, season‑specific planting advice without internet or API costs, preserving user privacy and enabling offline use in garden settings.

Workflow

• 1. User enters city name and clicks "What should I plant?"
• 2. App calculates current season (monsoon, winter, or summer) from today’s date
• 3. App sends a prompt to the local Gemma 3 model via Ollama, including city and season context
• 4. Model returns a short list of recommended plants
• 5. App renders the list, ensuring proper formatting (bold, lists) and no duplicate entries

Evidence

I built a small app that gives planting advice for Indian cities, using an open model that runs entirely on my own computer.

Source

The app picks the season (monsoon, winter or summer) from today's date, then asks Gemma (gemma3:4b), running locally through Ollama, to write the advice.

Source

My first answer had three problems: raw ** symbols showing on the page, advice about winter rain and frost that doesn't fit Gujarat, and the same vegetable listed twice under two names (radish and mooli).

Source

I fixed it by telling the model that India's October to February is dry and mild, limiting it to 5 plants with no repeats, and rendering bold text and lists properly on the page.

Source

I used Cursor's agent for the first version and DevRelay to connect to the Hacktoberfest challenges.

Source