Boids
Boids — flocking-birds emergent behaviour simulation.
What I asked each model — the Boids prompt
Every model on this page got this exact prompt inside the Agent Operating System: Boids — flocking-birds emergent behaviour simulation.
Single HTML file out. No iteration. No examples in the system prompt. Whatever each model produced on the first run is what's on this page. 23 frontier models have attempted it so far: Claude Fable 5, Fugu Ultra, Fugu Mini, Fusion, Gemini 3.6 Flash, GLM-5.2, GPT-5.6 Sol, Grok, Inkling, Kimi K3, MiniMax M3, Hermes MoA, Opus 4.8, Claude Opus 5, Qwen 3.8, Qwen 3.7, Claude Sonnet 5, DeepSeek V4 Pro, DeepSeek V4 Flash, Kimi K2.7, Kimi K2.7 · Fast, Kimi K2.7 · No-Think, Kimi K2.7 · Quality.
Why this task matters. Boids is a textbook test of sim-class capability — the kind of build that exposes whether a model is doing pattern-matching or actual reasoning. Shipping this cleanly is the floor for what I expect from a frontier model — every model on the leaderboard should at least attempt it.
How each model handled Boids
Ranked by my 0–10 score from the source comparison guides on agentos.guide. Click any to play the actual one-shot HTML the model produced.
What I saw: Strong 3D boids sim: cone-shaped birds with cyan-magenta gradient clearly flocking within a subtle wireframe bounds box, plus attract/scatter marker and clean title UI. Docks slightly for the somewhat scattered/loose clustering visible and generic starfield, but it's polished and clearly on-brief.
What I saw: Ultra v2 — boids flock, agents respond to mouse. Smoke-test PASS (9.5% pixel diff — strong motion).
What I saw: Boids flocking sim — separation/alignment/cohesion. Smoke-test PASS, agents visibly respond to mouse disturbance.
What I saw: Canvas2D boids with separation/alignment/cohesion. Hundreds of agents, click to disturb the flock, HUD shows agent count + FPS.
What I saw: Strong polished UI with full flocking controls (separation/alignment/cohesion sliders, modes, themes) and a visible multi-colored 3D boid swarm, but the flock reads as a tiny clustered blob rather than expansive emergent flocking, and FPS:6 indicates serious performance issues undermining the sim.
What I saw: Beautiful nebula backdrop with colorful trailing boids, clean glass control panel, and full interactive controls (alignment/cohesion/separation/speed, pointer attract/scatter ring) — very polished and clearly on-brief. Slight concern: the FPS readout shows '1 FPS', which if real would indicate performance trouble, keeping it just shy of the top.
What I saw: Hundreds of glowing fish with flowing trails and the lone red predator they flee. The prettiest of the pure-visual set.
What I saw: Renders cleanly with a polished neon title, grid, starfield and working 3D boid flocking, but the sparse scattered cones and one oversized foreground boid look under-clustered and slightly awkward rather than showing tight emergent flocks; solid but not a task winner.
What I saw: Renders a polished 3D boids sim with genuine emergent flocking clusters, colorful instanced birds with wings, a glowing repel/attract orb, dusk gradient sky and a clean control panel — strong visual and simulation depth that matches the top of the field. Minor weakness: 44fps at 280 boids hints at some performance cost, but it's clearly a task winner.
The winner on Boids
Kimi K3 took gold on this task. 3D flocking birds.
What I saw: Renders a polished 3D boids sim with genuine emergent flocking clusters, colorful instanced birds with wings, a glowing repel/attract orb, dusk gradient sky and a clean control panel — strong visual and simulation depth that matches the top of the field. Minor weakness: 44fps at 280 boids hints at some performance cost, but it's clearly a task winner.
See Kimi K3's full model card: /models/kimik3.
Every attempt — live, playable
Side by side. Click any tile to run that model's actual one-shot HTML in a new tab.
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVE
▶ LIVEHow I scored Boids — methodology
Three axes, 0–10 each, averaged. Runs: drop the .html in a browser; if it opens to a broken page, it scores zero. Hits the brief: did the model ship the thing the prompt asked for, or a different thing it found easier. Looks good: visual polish, motion, interactivity — where most of the gap between gold and silver lives.
My scores trace back to the source comparison guides on agentos.guide. See the full methodology page for data provenance, including which source guide each cell's score came from.
Related
More sim benchmarks: all tasks in the Sim category · See the best AI model for Boids · Back to the leaderboard
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.