Lab build · October 24, 2026

AI Model Showdown: Squid 3D Video

We gave Claude Opus 5.5 and GPT-6 Astra the exact same prompt: bring our squid to life as a 20-second 3D film you can then explore. Three runs each, then one best and final. Here are all eight squids, exactly as the models wrote them. Pick any two and watch them side by side, or open one full screen and drag to fly around it.

▶ 20-second filmThen the camera is yours: drag to orbit, scroll or pinch to zoom

Your turn: drag to orbit, scroll to zoom

Open alone

Your turn: drag to orbit, scroll to zoom

Open alone

Two 3D scenes at once is heavy work for a phone. If yours struggles, switch to One at a time.

The awards

Best Cinematography: Claude Opus 5.5

For the camera work and the world around the squid: rocky trench walls, drifting jellyfish and marine snow, in almost every run.

Best Special Effects: GPT-6 Astra

For wet, glossy skin and glow that reads as real light, best seen in its final version with shafts of light falling from above.

What we learned

One answer is a sample. The same prompt on the same model gave very different films every time. If the result matters, run it more than once.
Check what comes back. Two of the eight squids failed in ways you can spot in seconds: a blank page and a squid that never lights up.

Feedback works. Showing each model its own three attempts produced Astra's best squid by a wide margin.
More time is not more quality. Claude worked at least twice as long as Astra in every round. Picking the right model for the job is a skill, and it is one we teach.

How long each model worked

Round	Claude Opus 5.5 (High)	GPT-6 Astra (High)

"Worked" is the thinking and writing time each chat app reported for the reply.

The prompt, word for word

Every first-round run used this prompt, unchanged, in a fresh chat. Try it in your own AI tool.

Do not use web search, connectors or any other tools. Write the complete HTML file directly in your reply. Build a cinematic, interactive 3D scene as one self-contained HTML file. SUBJECT A giant squid gliding through a deep-sea trench at night. Eight arms and two long feeding tentacles that move independently and fluidly, a mantle that pulses as it swims by jet propulsion, large reflective eyes, and bioluminescent spots along its body that ripple in waves of cyan and teal light. Around it: drifting marine snow, faint jellyfish, rocky trench walls fading into darkness, and thin shafts of light from far above. LOOK Photographic and atmospheric: physically based materials, wet glossy skin, volumetric haze, soft bloom on the glowing spots, deep navy darkness. Every object is a closed solid from every angle; no parts pass through each other. MOTION The first 20 seconds play as a film with no input: start on a close-up of one glowing eye in the dark, follow the arms as they unfurl, then pull back to reveal the whole squid swimming past the camera, ending on a clean hero shot of it in full glow. After 20 seconds, the visitor can orbit and zoom with the mouse. TECHNICAL - Use Three.js loaded from cdnjs or jsDelivr. No image files: generate every texture in code. - Drive all motion from elapsed time (performance.now), not from frame counts. - Fill the whole browser window at any size, including tall phone shapes. - No on-screen text. DONE MEANS The arms and tentacles move smoothly and naturally with no stiff or broken joints, the bioluminescence visibly ripples, nothing clips or floats, and the 20-second camera move is smooth.

The best-and-final prompt

For the last round, each model got its own three earlier files as attachments, with this message:

Attached are three separate attempts you made in three separate sessions in response to the Squid Prompt.md. Using those three attempts as context, produce the best and final version you can to satisfy all the goals of the original Squid Prompt.md.

About this build

Recorded October 24, 2026. Each model ran on its High setting in its own chat app (Claude Opus 5.5 in Claude, GPT-6 Astra in ChatGPT), in fresh chats, with the prompt above. Each model wrote a complete web page, and the pages on this site are those files as delivered, with no edits to the code. Two of Claude's answers were too long for a single reply and arrived in two parts with instructions on where to join them; we joined them exactly as instructed. Astra's first run stopped on a JavaScript syntax error. We pressed ChatGPT's own fix button, which sent Astra the error, and Astra returned a fixed page about 40 seconds later; both the page as delivered and the fixed page are here. For the best and final round, each model got its own three earlier attempts as attachments and the best-and-final message shown above.

The 3D engine (Three.js) loads from a public CDN, so the scenes need an internet connection and a reasonably modern browser. Every texture, light and creature is generated in code. Picking the right AI model for the job, and checking what it gives you, is part of every training engagement we run, for companies, teams and individuals. More builds in the Lab.

Sunlight shining down through deep blue water

Find light in the deep waters of AI

Let's talk! A short call will help you learn more about our training and consulting solutions. We can also discuss whether our existing licensed tools can help, or whether you need a custom build.

Book a free call