Get the Agent OS + join 4,000+ founders inside the AI Profit Boardroom → Join AIPB ($59/mo)
Game

Nordiccrypt

Nordic Crypt — torch-lit Nordic dungeon crawler, ancient ruin to explore, first-person.

CategoryGame
Models tested25
Scored18/25
Avg score7.31/10
WinnerFusion

What I asked each model — the Nordiccrypt prompt

Every model on this page got this exact prompt inside the Agent Operating System: Nordic Crypt — torch-lit Nordic dungeon crawler, ancient ruin to explore, first-person.

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. 25 frontier models have attempted it so far: Claude Fable 5, Fugu Ultra, Fugu Ultra 1.1, Fugu Mini, Fusion, Gemini 3.6 Flash, GLM-5.2, GPT-5.6 Sol, Inkling, Kimi K3, MiniMax M3, Hermes MoA, Muse Spark 1.2, Opus 4.8, Claude Opus 5, Qwen 3.8, 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, Grok, Kimi K2.7.

Why this task matters. Nordiccrypt 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 Nordiccrypt

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.

Claude Fable 5 Anthropic
• 7.8/10

What I saw: Strong atmospheric first-person crypt with convincing stone-block walls, warm torch glow, vignette, and clean rune-styled title; the huge foreground flame sprite reads slightly like an over-bright blob and the scene feels a touch empty/generic corridor rather than a rich explorable ruin, keeping it just below the top tier.

▶ Play Claude Fable 5's attempt →
Fugu Ultra Sakana AI
🥈 9.0/10 · winner · dungeon depth

What I saw: Ultra v2 (gap-fill) — 61.5KB Nordic dungeon crawler with bloom + boss room. Smoke-test PASS with 22.8% pixel diff — highly reactive.

▶ Play Fugu Ultra's attempt →
Fugu Ultra 1.1 Sakana AI
• 5.2/10

What I saw: Strong HUD, first-person weapon/shield and visible enemies (7) with combat scaffolding are present, but the scene reads bright and flat rather than torch-lit crypt — lighting/atmosphere totally missing the Nordic dungeon mood, and floating stretched geometry looks buggy.

▶ Play Fugu Ultra 1.1's attempt →
Fugu Mini Sakana AI
• 7.0/10

What I saw: Mini gap-fill — 31KB Nordic dungeon. Smoke-test MAYBE (0.5% diff) — pointer-lock FPS; flagged for manual verification.

▶ Play Fugu Mini's attempt →
Fusion OpenRouter
🥇 9.5/10 · winner · best dungeon

What I saw: RETRY @ 24K tokens — now complete: 29KB with rAF + 7 input handlers + closed tags. Torch-lit ancient ruin, PointerLockControls, bloom, chasing enemies, boss room. The original truncated attempt has been replaced with a working build.

▶ Play Fusion's attempt →
• 7.3/10

What I saw: Strong atmospheric first-person crypt with polished HUD, textured brick walls, torch lighting, and a nicely modeled viewmodel axe; but the screenshot shows no visible enemies/combat and the framing is dominated by a flat wall, so it reads as a walking sim rather than proving the Draugr/rune loop advertised in the counters.

▶ Play Gemini 3.6 Flash's attempt →
GLM-5.2 Zhipu / Z.ai
• 8.0/10

What I saw: 30KB · plays clean · three, webgl

▶ Play GLM-5.2's attempt →
GPT-5.6 Sol OpenAI
• 7.8/10

What I saw: Strong architectural depth with pillars, wooden crossbeams, glowing runes, torches and a distant shrine/portal that reads clearly as an ancient Nordic ruin, plus polished HUD framing. But the lighting is too bright/washed-out for a 'torch-lit dungeon' — it lacks the dark, atmospheric moodiness the best entries nailed, so it's shippable but not the winner.

▶ Play GPT-5.6 Sol's attempt →
Inkling Thinking Machines
• 5.2/10

What I saw: Renders with a strong Cinzel title and clean HUD, but the scene is blown out in flat orange — torch point lights over-saturate everything so pillars/floor lose all stone texture and the 'crypt' reads as a bright yellow tunnel rather than a dark torch-lit ruin. Camera also spawns clipped inside a pillar; atmosphere and mood miss the brief badly.

▶ Play Inkling's attempt →
Kimi K3 Moonshot AI
• 8.6/10 · atmospheric torch-lit crypt

What I saw: Strong: genuine first-person 3D dungeon with convincing torch glow, warm falloff lighting on stone brickwork, scattered rubble props, and a polished serif Nordic UI with rune-collection goal that fully nails the brief. Minor weakness: the corridor geometry is a bit boxy and props read as simple low-poly, but the atmosphere and rendering are clearly task-winning quality.

▶ Play Kimi K3's attempt →

The winner on Nordiccrypt

Fusion took gold on this task. winner · best dungeon.

What I saw: RETRY @ 24K tokens — now complete: 29KB with rAF + 7 input handlers + closed tags. Torch-lit ancient ruin, PointerLockControls, bloom, chasing enemies, boss room. The original truncated attempt has been replaced with a working build.

See Fusion's full model card: /models/fusion. Direct head-to-head against the runner-up: Fusion vs Fugu Ultra.

Every attempt — live, playable

Side by side. Click any tile to run that model's actual one-shot HTML in a new tab.

How I scored Nordiccrypt — 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.

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