Interactive 3D gummy jellyfish, two-model build
- Views
- 14.7K
- Likes
- 119
- Replies
- 11
- Saves
- 60
An interactive 3D gummy jellyfish built with WebGPU that can be grabbed, stretched, and squished in the browser, generated from a detailed prompt and compared against a version made by ChatGPT-6 Astra.
Full prompt3,007 chars
Create “Jellyfish Jelly” — a beautiful, interactive 3D gummy jellyfish that users can grab, stretch, and squish directly in their browser. Deliver a complete single-file HTML experience using genuine WebGPU. ART DIRECTION Make it look like a premium translucent gummy sculpture: A rounded, softly scalloped jellyfish bell. Six flowing tentacles and a few delicate internal structures visible through the bell. Glossy, translucent material with thickness-dependent color absorption, refraction, subtle subsurface lighting, and gentle rim highlights. Rich color in thicker areas and more transparency at the edges. A warm off-white background, soft studio lighting, and a subtle shadow beneath the floating jellyfish. Keep the composition elegant and uncluttered. Avoid opaque plastic, excessive bloom, blown-out highlights, and visible mesh seams. GEOMETRY AND PHYSICS Generate all geometry procedurally. No external models or image files. Use stable position-based dynamics or a mass-spring system with approximate volume preservation. The bell should feel soft but substantial. Tentacles should be lighter, more flexible, and respond with slightly delayed motion. Their roots must remain attached to the bell. Allow users to grab the bell or any tentacle at the clicked location. Pulling should deform the nearby area first, then elastically move the surrounding geometry. On release, the jellyfish should wobble and gradually recover without snapping back, teleporting, or oscillating forever. Add a subtle idle swimming motion: the bell slowly contracts and expands while the tentacles trail behind. Reduce this animation while the jellyfish is held. Do not fake softness by scaling or rotating the entire model. Clamp extreme stretching and use fixed simulation steps. INTERACTION Left-click or touch to grab and stretch. Right-drag or drag empty space to orbit the camera. Support a limited zoom range. Keep camera gestures separate from object interaction. Include Give it a nudge, Reset, Pause, and Reset view. Add Firmness, Internal damping, and Pulse rate sliders. Include ¼ speed and Show mesh toggles. Provide three palettes: Moon, Coral, and Deep. Switching colors must not reset the simulation. INTERFACE Use a minimal editorial layout with thin borders, generous whitespace, clean sans-serif controls, and a large italic serif heading: “Jellyfish.” Caption: “A little current. A little glow. A very soft drifter.” Place a compact “THE SPECIMEN” control panel on the right and a WebGPU status indicator at the top. QUALITY AND PERFORMANCE Reuse geometry and buffers instead of rebuilding meshes during dragging. Keep internal details attached to the deforming body. Handle transparency carefully, without disappearing tentacles, flickering surfaces, or harsh black outlines. Test repeated grabs, strong pulls, releases, palette changes, pause, reset, and mobile interaction. Show a clear fallback message if WebGPU is unavailable. Deliver the full working HTML, not a mockup or code fragment.
Source: the creator's reply under the post. Kept in its original language, unedited.
Want more prompts like this one?
Get the 50 most-viewed Claude 5.5 video prompts on one page: the full original text, the creator and the post it came from. We'll only email you when there's a new prompt worth trying. Unsubscribe anytime.
Tools the creator mentioned
Original post: https://x.com/vib3coded/status/2104651061336117614
All works are open data at GitHub · awesome-opus-5.5-video.
Model attribution is as stated by the creator and was not independently reproduced. View counts are a snapshot from 2026-11-19. The work belongs to its creator; DM @GoSailGlobal on X for corrections or removal.
More comparisons
0:42AI self-portrait challenge: Sonnet 5.5 vs GPT-6 Astra
7.3M views
0:17Interactive gummy orange slice comparison
1.2M views
0:13Gummy bear soft-body physics scene
632.6K views
0:13Glitter sticker effect comparison
480.8K views
0:18Jelly candy simulation comparison
448.7K views
0:18Watermelon rubber band test: Opus 5.5 vs GPT-6 Astra
421.5K views
Related articles
OpenAI Academy vs Anthropic: A Complete Comparison of Two Leading AI Companies' Free Learning Paths
Both of the top AI companies offer official free learning platforms — together that's 25+ courses and hundreds of hours of content, enough to take you from AI newcomer to resume-worthy engineer. This post maps out all 15 OpenAI Academy courses along a 10-week roadmap and provides a full side-by-side comparison with the Anthropic learning path.
Anthropic Claude Certified Architect Complete Roadmap: 8 Free Courses + $99 Exam
Anthropic launched the Claude Certified Architect certification on March 12, 2026, with 13 self-paced courses — all completely free, no registration required, no paywalls. The exam itself costs just $99. This post walks through a 12-week roadmap covering the 8 key courses in Weeks 1–9, with direct links to each.
OpenAI Agents SDK Major Upgrade: Harness and Sandbox Now Built Into the SDK — The Opposite Approach from Anthropic
OpenAI has shipped a major upgrade to its Agents SDK, with the headline changes being harness and sandbox support built directly into the SDK. Anthropic's Claude Agent SDK takes exactly the opposite approach — one gives you everything pre-assembled, the other hands you a toolbox.
Google Skills Complete Roadmap: AI Learning Gamified — Earn Skill Badges to Show Off on LinkedIn
Google has rebranded Cloud Skills Boost as Google Skills and reimagined AI learning as a series of 'dungeon runs' — each path has 5–8 activities, and you earn a Skill Badge to display on LinkedIn when you finish. This post covers the 3 paths beginners should take, plus direct links to each.