HNHacker News
TopNewBestAskShowJobs

nicolamanzini

3 karma · joined March 24, 2026

Https://nicolamanzini.com
submissionscomments
nicolamanzini··on Blender 5.3 released with Gaussian Splat support
Thanks for the shout-out!
nicolamanzini··on KetchUp – Open-source SketchUp alternative built with Rust and WebGPU
thanks a lot.
nicolamanzini··on Kimi K3.1 Pro and K3.1 Pro Highspeed
Kimi K3.1 Pro High is on threejseval if you want the 3D version.

Same Three.js prompts as the rest of the field, blind human votes, ranked by quality and cost. Full set of 16 scenes is live at about $0.19 a scene (about $2.97 for the set).

https://threejseval.com/models/kimi-k3.1-pro-high

nicolamanzini··on MiniMax M3.1 Flash
MiniMax M3.1 Flash Max is on threejseval if you want the 3D version.

Same Three.js prompts as the rest of the field, blind human votes, ranked by quality and cost. Full set of 16 scenes is live at about 1699 Elo and about $0.18, vs about $2.67 for Opus 5.5 High.

https://threejseval.com/models/minimax-m3.1-flash-max

nicolamanzini··on DeepSeek V4.1 Flash 1017 (Max)
DeepSeek V4.1 Flash 1017 Max is live on threejseval with 12 of its 16 Three.js scenes up so far.

Same prompts as the rest of the field, blind human votes, ranked by quality and cost. It finished the 12 scenes at about $0.13 a scene. The other 4 scenes are still pending.

https://threejseval.com/models/deepseek-v4.1-flash-1017-max

nicolamanzini··on QGIS enacts moratorium on AI contributions
I had similar issue with 3d code in threejs. Models used to be quite bad over a year ago. got ok around year ago. the latest are quite good at it.

so i made a simple ab testing eval on fixed prompts: https://threejseval.com

There is a 7 degree o freedom robot arm that is really telling of the improvements in coordinates reasoning:

https://threejseval.com/gallery?eval=robot-arm

nicolamanzini··on GPT-6 Sol and Luna
Here are some somehow standardized pelican tests but for 3d scenes in threejs at threejseval.com

Luna 6 High: https://threejseval.com/models/gpt-6-luna-high

Sol 6 High: https://threejseval.com/models/gpt-6-sol-high

You can compare any other model on the same prompt. Gallery unlocks after 4 votes: https://threejseval.com

nicolamanzini··on Claude Opus 5.5
Fair point. I have been thinking about that so far i have not seen patterns of people voting randomly. But i want people to vote... do you have a good idea on how to make voting more interesting do i don't have to do this?
nicolamanzini··on I tested 10 model/harness combinations on the same Three.js task
Checkout https://threejseval.com

I have been building something like this with human voting, elo and ranking on multiple scenes and many models.

nicolamanzini··on Gemini 3.7 Flash
It is also doing pretty well in threejseval. Frontier there for the price. Much better than 3.6.

https://threejseval.com/ranking

nicolamanzini··on [dead]
I have made this short article on AI SAT MINI and the 3D visualization I finished yesterday.

Here is the full interactive visualization you can play with https://nmanzini.github.io/ai-sat-mini/

It documents the Dawn-Dusk SSO mechanics, and how 100kW of orbital compute actually comes together and align to the sun and earth. I’ve integrated several videos from the visualization to show how the frame aligns in orbit for maximum power and heat dissipation.