Nightloom: A Dream-Logging Application that Visualizes Your Dreams from Fragments

Description and use

Nightloom turns scattered dream fragments into a coherent story and watercolor image, running entirely locally on modest hardware.

Problem

Users forget vivid dream details quickly and lack a local tool to transform fragmented memories into coherent stories and images.

Outcome

Nightloom enables users to transform fragmented dream memories into coherent stories and watercolor images locally, even on machines with only 4–6 GB free RAM.

Workflow

• 1. Install Ollama and pull Gemma 4 (or a smaller tag).
• 2. Start the Python server with python3 dream.py (optionally set DREAM_MODEL=gemma4:e2b).
• 3. Open http://127.0.0.1:8765 in a browser.
• 4. Enter dream fragments in any order.
• 5. The server sends fragments to Gemma 4 via Ollama to generate a story.
• 6. The server unloads Gemma from RAM and runs stable‑diffusion.cpp to paint a watercolor scene.
• 7. The story and image are displayed and saved to dreams.jsonl.

Evidence

Nightloom turns those fading scraps into something lasting.

Source

Everything runs 100% locally on your computer, so your beloved dreams stay with you.

Source

Because memory efficiency was a core design constraint, it works seamlessly even on modest machines with just 4–6 GB of free RAM.

Source

The page talks to a small Python server on 127.0.0.1, which talks to Ollama on localhost and, for images, to a local stable-diffusion.cpp program.

Source

To run smoothly on devices with only 4–6 GB of free RAM, dream.py automatically unloads Gemma from RAM before running stable-diffusion.cpp, freeing up system resources for image rendering.

Source