Best AI model for…
Best AI model for Blackhole
Black Hole — gravitational lensing visualisation.
🥇
Best model for Blackhole
Fusion
"Photon back-tracing through a curved-space metric (claims as much in the code) for actual gravitational lensing — disk's far side lifted over and under the shadow. Loading screen says "computing space-time metric…" Range-slider parameter panel for spin/disk tilt/exposure. Most ambitious blackhole attempt by code structure."
See Fusion's full track record →
The prompt — what I asked each model
Every model on this page got the same fixed prompt inside the Agent Operating System: Black Hole — gravitational lensing visualisation.
Single HTML file out. No iteration. No "best of N." No examples in the system prompt. Whatever each model produced on the first run is what's on this page. 23 models have attempted it so far — Fusion, Kimi K3, Opus 4.8, Claude Opus 5, DeepSeek V4 Pro, DeepSeek V4 Flash, Claude Fable 5, Fugu Ultra, Fugu Mini, Gemini 3.6 Flash, GLM-5.2, GPT-5.6 Sol, Grok, Inkling, Kimi K2.7, Kimi K2.7 · Fast, Kimi K2.7 · No-Think, Kimi K2.7 · Quality, MiniMax M3, Hermes MoA, Qwen 3.8, Qwen 3.7, Claude Sonnet 5.
What counts as winning here
This is the sim category on Goldie Bench. The question isn't "did the model write code that compiles" — the question is "did the model ship a thing you'd actually use." For Blackhole that means three things, in order:
- Does it run? Drop the .html file in a browser. If it opens to a broken page, it scores zero on the first axis.
- Did it hit the brief? The prompt asks for a specific thing. A model that ships a different thing — however polished — gets docked on the brief axis.
- Does it look good? Visual polish, motion, interactivity, attention to detail. This is where the difference between gold and silver usually lives.
Final score is my honest 0–10 across all three axes, averaged. Across the 18 models I've scored on this task so far, the average score is 7.45/10.
Every model's attempt — ranked by my 0–10 score
Models ranked by medal (highest score = 🥇 gold, second = 🥈 silver, third = 🥉 bronze). Click any tile to play that model's actual one-shot HTML.
▶ LIVEFusion 🥇
OpenRouter
Photon back-tracing through a curved-space metric (claims as much in the code) for actual gravitational lensing — disk's far side lifted over and under the shadow. Loading screen says "computing space-time metric…" Range-slider parameter panel for spin/disk tilt/exposure. Most ambitious blackhole attempt by code structure.
▶ LIVEKimi K3 🥇
Moonshot AI
Gorgeous real geodesic raytracing with a properly lensed accretion disk wrapping over the top, convincing Doppler asymmetry, photon-ring glow, and a rich nebula starfield backdrop; polished typography and clean UI push it to the top of the field. Minor stat overlap at bottom-left ('photon sphe' clipped by hint pill) is the only blemish.
▶ LIVEOpus 4.8 🥇
Anthropic
Opus nailed it — a pure-black event horizon, a bright photon ring, and the disk bent up and over the top exactly like the film's lensing. GLM came in strong with a clean ring and a starfield warping past the hole. Kimi's disk is fine, but the background is a soft grey blur instead of stars. This one's Opus's.
▶ LIVEClaude Opus 5 🥇
Anthropic
Stunning Schwarzschild ray-traced lensing with a convincing event horizon, warped accretion disk arcing over/under the shadow, Doppler beaming, and a rich lensed starfield — plus a polished, functional control panel with quality/inclination/distance sliders. The only faint quibble is a slightly hazy background and 27fps, but visually it matches the top of the field.
▶ LIVEDeepSeek V4 Pro
DeepSeek
▶ LIVEDeepSeek V4 Flash
DeepSeek
▶ LIVEClaude Fable 5
Anthropic
Strong Interstellar-style render with a clean event-horizon shadow, tilted accretion disk wrapping over/under the black hole, visible Doppler beaming brightening one side, and a subtle lensed arc — all polished with tasteful nebula/starfield and typography. Minor nit is the slightly soft disk detail, but it clearly matches or beats the field's best.
▶ LIVEFugu Ultra
Sakana AI
Ultra v2 — gravitational-lensing black hole. Smoke-test PASS (2.2% pixel diff).
▶ LIVEFugu Mini
Sakana AI
Gravitational-lensing black hole. Smoke-test PASS — pixel change confirmed after camera-orbit input.
▶ LIVEGemini 3.6 Flash
Google
Polished glassmorphism UI and an ambitious raymarching lensing shader, but the render is badly broken — a white/blown-out background with noisy artifacts and a black dome instead of a proper event horizon with a lensed accretion disk, plus a 2 FPS counter indicating serious performance failure.
▶ LIVEGLM-5.2
Zhipu / Z.ai
Opus nailed it — a pure-black event horizon, a bright photon ring, and the disk bent up and over the top exactly like the film's lensing. GLM came in strong with a clean ring and a starfield warping past the hole. Kimi's disk is fine, but the background is a soft grey blur instead of stars. This one's Opus's.
▶ LIVEGPT-5.6 Sol
OpenAI
Gorgeous shader render — the tilted accretion disk with fine banding, the photon-ring glow above/below the event horizon and the Doppler warm/cool gradient read as genuine gravitational lensing, all wrapped in a clean, polished HUD. Slightly weak on a distinct top-arc lensed disk crossing over the shadow, but overall a task-topping build.
▶ LIVEGrok
xAI
A striking black event horizon ringed by a white-hot accretion disk over a real starfield. The sharper sentence asked for a brighter ring and a defensive render — and got both.
▶ LIVEInkling
Thinking Machines
Only the title overlay and hint text render — the entire Three.js scene (black hole, accretion disk, stars, lensing) is absent from the screenshot, so the core brief is completely missing. The gradient headline is polished but the visualisation failed to render.
▶ LIVEKimi K2.7
Moonshot AI
Opus nailed it — a pure-black event horizon, a bright photon ring, and the disk bent up and over the top exactly like the film's lensing. GLM came in strong with a clean ring and a starfield warping past the hole. Kimi's disk is fine, but the background is a soft grey blur instead of stars. This one's Opus's.
▶ LIVEKimi K2.7 · Fast
Moonshot AI
▶ LIVEKimi K2.7 · No-Think
Moonshot AI
▶ LIVEKimi K2.7 · Quality
Moonshot AI
▶ LIVEMiniMax M3
MiniMax
Minimal gravitational-lens shader.
▶ LIVEHermes MoA
Hermes · Mixture of Agents
Solid geodesic ray-marcher with real per-step bending, a thin disk crossed via plane-intersection, doppler-ish beaming, photon-ring glow, and polished orbit/zoom/slider controls — cleaner and more complete than Grok/GLM/Qwen, but the disk lensing doesn't visibly fold up-and-over the shadow the way Fusion and Opus 4.8 (both 9.0) achieve, so it lands just shy of the top tier.
▶ LIVEQwen 3.8
Alibaba
Strong Interstellar-style render: bright photon ring hugging the black silhouette with the accretion disk arcing over the top and secondary lower image reads convincingly as lensing. Slightly let down by the scattered clumpy orbiting particles that look noisy against the clean disk, keeping it just shy of the top mark.
▶ LIVEQwen 3.7
Alibaba
5KB · plays clean · webgl, rAF
▶ LIVEClaude Sonnet 5
Anthropic
The lensing shader and event horizon shadow are present with a visible photon ring, but the camera has zoomed far too close, blowing out the disk into overexposed white banding and blocky artifacts on the right that read as broken/glitchy rather than a clean gravitational-lensing visual. The core concept works but the framing and clipping make it look flawed rather than shippable.
How I tested this — the methodology in 60 seconds
Every comparison on Goldie Bench follows the same recipe:
- I pick a creative coding prompt that a frontier model should be able to ship in one shot.
- I dispatch the exact same prompt to each model from the kanban inside the Agent Operating System.
- I save whatever .html file the model produced on the first run. No iteration. No coaching.
- I score each result 0–10 on my three axes (runs / hits the brief / looks good).
- I publish the scores publicly in the source comparison guides on agentos.guide — and that's what feeds this page.
See the methodology page for full data provenance, including which source guides each cell's score came from.
Related tasks and comparisons
More tasks in the Sim category · All attempts on Blackhole · Back to the leaderboard
The same stack Julian uses
Run this stack yourself.
Every demo on this bench was built inside the Agent Operating System — one prompt, one shot, single HTML file out. The Agent OS, the prompts, the templates, the weekly walkthroughs and 4,000+ founders shipping with it every day all live inside the AI Profit Boardroom.
4,000+founders
258documented wins
38countries
$59/momonthly