MODEL ARCADE
Twenty-nine autonomous coding outputs published as unmodified benchmark artifacts. Select a model and game to inspect the result. Entries marked failed are preserved as benchmark evidence, not certified playable results.
SPACE INVADERS — 11 MODELS
Claude Opus 5.5
Anthropic via OpenRouter · high reasoning · 29/29 deterministic checks plus real-input verification
DeepSeek V4.1 Flash
DeepSeek via OpenRouter · high reasoning · independently browser-verified
GPT-6 Astra High
OpenAI via OpenRouter · high reasoning · independently browser-verified
Mellum2 12B-A2.5B
Local MXFP4 MoE · failed autonomous repair · score remained 0
GPT-5.6 Sol
OpenAI endpoint · 32K allowance
Claude Fable 5
Anthropic endpoint · 32K allowance
DeepSeek V4 Flash 0731
Morph BF16 · 32K allowance
Gemini 3.7 Flash
Google endpoint · 32K allowance
GLM-5.3
Z.AI FP8 endpoint
Qwen3.8 Max
Alibaba · completed after resume
Qwen3.8 27B
AkashML BF16
GALAGA — 11 MODELS
Claude Opus 5.5
Anthropic via OpenRouter · high reasoning · entry, formation, dives, fire and terminal states verified
DeepSeek V4.1 Flash
DeepSeek via OpenRouter · high reasoning · entry, formation and dives browser-verified
MiMo 2.6 Flash
Xiaomi via OpenRouter · high reasoning · 19/19 browser checks passed
GPT-6 Astra High
OpenAI via OpenRouter · high reasoning · browser interaction verified
GPT-5.6 Sol
OpenAI · $0.1785
Claude Fable 5
Anthropic · browser crash self-fixed
DeepSeek V4 Flash 0731
Morph BF16 · extended natural completion
Gemini 3.7 Flash
Google · largest visual artifact
GLM-5.3
Z.AI FP8 · extensive verification
Qwen3.8 Max
Alibaba · real Chrome/end-state tests
Qwen3.8 27B
AkashML BF16 · required 128K continuation
3D DESCENT — 2 MODELS
Claude Opus 5.5 — Neon Descent
Self-contained WebGL · six-axis flight · keys, combat, reactor escape, failure and victory independently verified
GPT-6 Astra High — Deep Vector
Self-contained WebGL · six-axis flight · reactor escape · 83/83 independent checks
PIXEL ROGUELITE — 2 MODELS
Claude Opus 5.5 — Pixel Panic
Exact matched prompt · three procedural floors · elemental chains · Food-Court King · 37/37 static plus 68/68 browser checks
GPT-6 Astra High — Pixel Panic
Procedural mall mayhem · elemental chains · Food-Court King · 41 verified test groups
ORIGINAL FREE-CHOICE — 1 MODEL
Claude Opus 5.5 — Lanternwake
Original lantern-weaving action game · four escalating phases · God Mode · 68/68 checks plus three complete bot victories
TURN-BASED JRPG — 1 MODEL
Claude Opus 5.5 — Emberwake
Original 16-bit-inspired RPG chapter · four classes · exploration, magic and progression · three-phase boss · independently verified
THIRD-PERSON 3D ADVENTURE — 1 MODEL
Claude Opus 5.5 — Heliolabe
Original cinematic WebGL adventure · climbing, rope and zipline traversal · cover combat, mirror puzzle, set pieces and boss · 218/218 checks
No external game assets or runtime network dependencies. Publicly served through Cloudflare Pages.