Neonsnake
Neonsnake — auto-discovered task.
What I asked each model — the Neonsnake prompt
Every model on this page got this exact prompt inside the Agent Operating System: Neonsnake — auto-discovered task.
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. 9 frontier models have attempted it so far: Gemini 3.6 Flash, GPT-5.6 Sol, Inkling, Kimi K3, Muse Spark 1.2, Claude Opus 5, Qwen 3.8, DeepSeek V4 Pro, DeepSeek V4 Flash.
Why this task matters. Neonsnake is a textbook test of other-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 Neonsnake
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: Polished neon start menu with a visible 3D grid boundary and clean cyberpunk styling behind the blur, but the screenshot shows only the pre-game overlay — no actual snake, gameplay, or scoring visible, so it reads as a strong shell rather than a proven playable build.
What I saw: Clean, atmospheric render with a well-lit segmented snake, glowing food orb, orbital rings and crisp HUD that clearly beats the field's best on polish; slightly undersold by the muted grid/rings and the neon glow reading softer than the branding promises.
What I saw: Strong glowing title typography and clean grid/UI, but the neon snake reads as a small dim segmented tube with weak emissive glow and minimal presence—the centerpiece lacks the vivid neon punch and visual scale the brief demands, leaving it below the field's best.
What I saw: Strong 3D take on the brief: crisp neon walls with glowing corner nodes, tilted perspective grid, glowing snake with eyes and a radiant food orb, plus polished HUD, autopilot AI, and pause overlay — visibly beats a flat 2D field. Minor nit is the slightly empty arena interior, but the render is clean, on-brand, and clearly a task winner.
What I saw: The 3D perspective grid and neon 'LINK SEVERED' modal render with polish, but the game state shows immediate game-over (Score 0, Length 3, Tasks 0/12) with no snake visible — it evidently auto-ran into a wall on load, and most HUD text is barely legible/washed out. Strong aesthetic direction undercut by a broken-looking initial state.
What I saw: Strong 3D neon arena with polished glow, additive bloom, HUD, autopilot AI (BFS pathfinding), and clean neon frame/ring aesthetics; snake head and food read clearly with nice lighting. Loses a touch because the snake body barely reads as segmented at score 0 and the floor/border geometry looks slightly loose, but overall a shippable, distinctive take on the brief.
What I saw: Strong neon aesthetic with polished HUD, CRT vignette, and clean paused panel, but the screenshot shows an empty playfield with no visible snake or food, so the actual gameplay isn't demonstrated. Presentation is above-average yet it reads as a shell rather than a proven, running game.
Demo on the bench. Not scored yet — play it and form your own opinion.
Demo on the bench. Not scored yet — play it and form your own opinion.
The winner on Neonsnake
Kimi K3 took gold on this task. 3D neon arena.
What I saw: Strong 3D take on the brief: crisp neon walls with glowing corner nodes, tilted perspective grid, glowing snake with eyes and a radiant food orb, plus polished HUD, autopilot AI, and pause overlay — visibly beats a flat 2D field. Minor nit is the slightly empty arena interior, but the render is clean, on-brand, and clearly a task winner.
See Kimi K3's full model card: /models/kimik3. Direct head-to-head against the runner-up: Kimi K3 vs Claude Opus 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
▶ LIVEHow I scored Neonsnake — 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 other benchmarks: all tasks in the Other category · See the best AI model for Neonsnake · 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.