Voxelcraft
Voxel Craft — Minecraft-style sandbox, place + break blocks, day/night cycle.
What I asked each model — the Voxelcraft prompt
Every model on this page got this exact prompt inside the Agent Operating System: Voxel Craft — Minecraft-style sandbox, place + break blocks, day/night cycle.
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, Fusion, Gemini 3.6 Flash, GLM-5.2, GPT-5.6 Sol, Grok, Inkling, Kimi K3, LongCat-2.0, MiniMax M3, Hermes MoA, Muse Spark 1.2, Opus 4.8, Claude Opus 5, Qwen 3.7, Claude Sonnet 5, Kimi K2.7 · Fast, Kimi K2.7 · No-Think, Kimi K2.7 · Quality, DeepSeek V4 Pro, DeepSeek V4 Flash, Kimi K2.7.
Why this task matters. Voxelcraft is a textbook test of game-class capability — the kind of build that exposes whether a model is doing pattern-matching or actual reasoning. A model that ships this in one shot is usually safe to wire into your agent loop for harder tasks of the same shape.
How each model handled Voxelcraft
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: AAA rebuild is an authentic textured Minecraft: procedurally-textured grass/dirt/stone/wood blocks (not flat colors), a blocky tree, flowers and stone structures, a first-person hand with a placement crosshair, a day clock, and a full HUD with a heart row, a hunger row and a textured 5-slot block hotbar. Break, place and move verified — a real little world.
What I saw: Ultra v2 (gap-fill) — Minecraft-style voxel sandbox. Smoke-test PASS.
What I saw: Minecraft-style voxel sandbox in 23KB. Block-based world, fly around, click-to-break, right-click-to-place, day/night cycle, block-picker hotbar. Closest to a real Minecraft any Fusion build managed.
What I saw: Strong voxel world with terrain, a tree, HP/time/kills HUD, an 8-block hotbar and — crucially — a visible zombie-style enemy (plus a red one at right) delivering the combat the brief rewards; place/break plus day/night cycle all present, edging past typical walking-sim entries.
What I saw: 44KB · plays clean · three, webgl
What I saw: Renders a clean, colorful voxel world with trees, water, crosshair, hotbar and day/night HUD — clearly on-brief and polished; strong pointer-lock FPS setup with place/break and fly, though the flat pastel lighting and simple terrain keep it just shy of the top build.
What I saw: Minecraft-style voxel sandbox, fly + break + place, day/night cycle, hotbar. 22KB.
What I saw: Renders a clean, colorful voxel terrain with nice shadows, gradient title, and polished HUD; but the flat pastel palette looks more candy than Minecraft, placement only works on y=0 ground plane, and the pointermove-based 'orbit' fights every click since dragging is never gated — solid but generic and mechanically shaky.
What I saw: Strong: gorgeous procedural terrain with textured grass/dirt/stone/water/trees, real shadows, day/night clock HUD, crosshair and control hints all rendering cleanly and on-brief. Weak: the hotbar area at bottom looks faded/half-rendered in the shot, but the world itself is clearly a top-tier voxel sandbox that matches the field's best.
What I saw: One-shot 9KB Minecraft-style voxel world — 16x16 grass/dirt/stone cubes, voxel trees, day/night sky, raycast break+place, real WASD+mouse. verify-move: walks+looks, 0 errors. Built the full world one-shot but the initial camera yaw faced away (sky-only) — a one-line framing patch (yaw 0, pitch -0.5) pointed it at the terrain.
The winner on Voxelcraft
Claude Fable 5 took gold on this task.
What I saw: AAA rebuild is an authentic textured Minecraft: procedurally-textured grass/dirt/stone/wood blocks (not flat colors), a blocky tree, flowers and stone structures, a first-person hand with a placement crosshair, a day clock, and a full HUD with a heart row, a hunger row and a textured 5-slot block hotbar. Break, place and move verified — a real little world.
See Claude Fable 5's full model card: /models/fable-5.
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 Voxelcraft — 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 game benchmarks: all tasks in the Game category · See the best AI model for Voxelcraft · 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.