oneshotlm
Compare all models on this prompt →

Reaction-diffusion · minimax-m2.5

Run a Gray-Scott reaction-diffusion simulation on a canvas that grows organic Turing patterns in real time, with the mouse seeding new growth.

⛨ sandboxed Open in new tab ↗
transcript · 6 messages
user

Run a Gray-Scott reaction-diffusion simulation on a canvas that grows organic Turing patterns in real time, with the mouse seeding new growth.

assistant
thinking
The user wants me to create a Gray-Scott reaction-diffusion simulation in a single HTML file. This is a classic mathematical model that produces Turing patterns (like spots, stripes, spirals) that look organic and biological.

The Gray-Scott model uses two chemicals, A and B, with the following equations:
- ∂A/∂t = Dₐ∇²A - AB² + f(1-A)
- ∂B/∂t = Dᵦ∇²B + AB² - (k+f)B

Where:
- Dₐ and Dᵦ are diffusion rates
- f is the feed rate
- k is the kill rate
- ∇² is the Laplacian (diffusion)

I'll create a canvas-based simulation where:
1. The simulation runs in real-time using WebGL or Canvas 2D
2. Mouse interaction seeds new chemical B (the "growth" chemical)
3. Beautiful organic patterns emerge

Let me build this as a single self-contained HTML file with:
- CSS for styling (fullscreen canvas, dark theme)
- JavaScript for the Gray-Scott simulation using Canvas 2D (simpler than WebGL but still works well)
- Mouse event handling to seed new growth

I'll implement it with the Laplacian convolution approach for the reaction-diffusion.
write /tmp/ws/index.html (12198 bytes)
toolResult

Successfully wrote 12198 bytes to /tmp/ws/index.html

assistant
thinking
The file has been created. Let me verify it exists and check that everything looks correct.
bash {"command":"ls -la /tmp/ws/index.html"}
toolResult

-rw-r--r--. 1 node node 12216 Aug 1 17:58 /tmp/ws/index.html

assistant
thinking
The file has been created successfully. It's a self-contained Gray-Scott reaction-diffusion simulation. Let me provide a summary to the user.

Done! I've created a **Gray-Scott reaction-diffusion simulation** in `/tmp/ws/index.html`. ### Features: - **Real-time simulation** using the Gray-Scott model with two chemicals (A and B) - **Organic Turing patterns** emerge from the reaction-diffusion equations - **Mouse interaction** - click and drag anywhere to seed new growth (adds chemical B) - **5 pattern presets**: - **Coral** - classic coral-like branching - **Mitosis** - cell-dividing spots - **Spirals** - rotating spiral waves - **Worms** - worm-like tendrils - **Chaos** - chaotic spotted patterns - **Clear button** to reset and start fresh - **Mobile-friendly** with touch support - **Deep teal/organic color palette** that highlights chemical concentration ### How it works: The simulation solves the Gray-Scott equations: - ∂A/∂t = Dₐ∇²A - AB² + f(1-A) - ∂B/∂t = Dᵦ∇²B + AB² - (k+f)B Where the feed rate (f) and kill rate (k) parameters determine the pattern type. Click/drag to add chemical B and watch the patterns grow and evolve!

TAKE IT TO CODE

A single self-contained HTML file — copy it and open in any browser, or reuse the prompt with your own pi-agent run.

RUN

modelminimax-m2.5
vendorMiniMax
harnesspi-agent
generated2026-08-01

EVALUATION

4/5 working

Gray-Scott reaction-diffusion patterns evolve organically over time, with click/drag seeding visible growth spots.

Vision grade · anthropic/claude-sonnet-5

MEASUREMENTS

Raw readings from the capture, not a verdict — the grade above is the only judgement. Percentages are how much of the screen changed; the measure under-reports sparse motion on a dark background.

idle motion2.6%
on click23.6%
on drag28.5%
on wheel20.4%
on enter + space16.5%
on arrow keys11.5%
on w/a/s/d8.8%
frame spread38.4 / 255
console errors0
js errors none

METRICS

runtime121.4s
tokens in12k
tokens out3.8k
cached11k
cost / run$0.0040