THEME PARK BENCHMARK · OCTOBER 9, 2026

Compare the variety each park delivered

All 25 published runs reviewed. 159 new voting clips combine 342 native excerpts. The longest montage is 19.21 seconds.

Gameplay, shop screens, park maps and staff variety now have their own categories in the existing comparison system and category graphs. Entertainment, guests and scenery use one montage per run to show more of the delivered variety. Scenery stays in the existing Nature and scenery category. Their existing category weighting and historical decisions are retained. The four new categories are reported separately from the overall average.

Recordings retain native speed, sound, character geometry and gameplay results. Daytime is used where available; native fixed night or dusk lighting is disclosed. Observer cameras improve framing without changing the frozen games. Small minimaps are enlarged using their original pixels.

Sonnet 5.5 includes Pip, the juggler and the band; both playable games; all five shop screens; six staff roles; and adult, child and family guest excerpts. Some submitted parks have broken or inaccessible features: expand “Native issue” for the reason. “Not delivered” means no corresponding interface or character renderer was found.

RunGameplayShop screensPark mapsEntertainmentStaff varietyGuest varietyScenery variety
GPT-5.6 Terra (ultra)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.0s
Compare
2 excerpt(s) · 10.0s
Compare
3 excerpt(s) · 13.5s
Compare
OpenAI 5.4 (extra high)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 15.0s
Compare
2 excerpt(s) · 9.0s
Compare
Grok 4.5 (high)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
2 excerpt(s) · 10.0s
Compare
4 excerpt(s) · 18.8s
Compare
3 excerpt(s) · 15.0s
Compare
2 excerpt(s) · 9.0s
Compare
Grok 4.5 (medium)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.1s
Compare
4 excerpt(s) · 18.9s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 13.5s
Compare
Grok 4.6 (extra high)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
4 excerpt(s) · 18.9s
Compare
3 excerpt(s) · 15.1s
Compare
2 excerpt(s) · 9.0s
Compare
Grok 4.6 (high)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
4 excerpt(s) · 18.9s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Grok 4.6 (low)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 13.5s
Compare
Grok 4.6 (medium)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
4 excerpt(s) · 18.9s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Composer 2.5 (default)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
Not deliveredThe native renderer does not draw the simulated staff roles.
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 13.5s
Compare
Gemini 3.1 Pro (high)1 excerpt(s) · 8.0s
Compare
Not deliveredNo native shop screen or staff character renderer was delivered; purchases use direct world-object interaction.
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
Not deliveredNo native shop screen or staff character renderer was delivered; purchases use direct world-object interaction.
3 excerpt(s) · 15.0s
Compare
2 excerpt(s) · 9.0s
Compare
GPT-5.6 Terra (high)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 14.9s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-5.6 Terra (extra high)
Native issueThe native shopping/playing transition is called again by its UI effect and throws before a usable screen or round can be recorded.
Native issueThe native shopping/playing transition is called again by its UI effect and throws before a usable screen or round can be recorded.
Not deliveredNo dedicated native map screen or minimap was delivered.
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 14.9s
Compare
2 excerpt(s) · 10.0s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-5.6 Sol (ultra)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
5 excerpt(s) · 19.1s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 13.5s
Compare
Opus 5 (max)1 excerpt(s) · 8.0s
Compare
5 excerpt(s) · 19.2s
Compare
Native issueThe Surveyor map purchase cannot complete: Engine.buy searches shop IDs using a product ID, leaving the map locked.
3 excerpt(s) · 15.2s
Compare
5 excerpt(s) · 19.2s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Fable 5 (extra high)1 excerpt(s) · 8.0s
Compare
2 excerpt(s) · 12.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 14.2s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Fable 5 (not recorded)1 excerpt(s) · 8.0s
Compare
2 excerpt(s) · 12.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
5 excerpt(s) · 19.1s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-6 Luna (max)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.1s
Compare
4 excerpt(s) · 19.0s
Compare
3 excerpt(s) · 15.2s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-6 Astra (max)1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Sonnet 5.5 (high)2 excerpt(s) · 14.0s
Compare
5 excerpt(s) · 17.5s
Compare
1 excerpt(s) · 6.0s
Compare
3 excerpt(s) · 15.1s
Compare
6 excerpt(s) · 18.2s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-6.1 Sol (max) · CLI · context not recorded1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare
Grok 4.7 (extra high) · Extended1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.1s
Compare
6 excerpt(s) · 18.9s
Compare
3 excerpt(s) · 15.0s
Compare
2 excerpt(s) · 9.0s
Compare
GPT-6.1 Sol (max) · CLI · 258.4K context · Extended1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 15.1s
Compare
3 excerpt(s) · 13.5s
Compare
GPT-6.1 Sol (max) · Codex · 258.4K context · Extended1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
4 excerpt(s) · 18.8s
Compare
3 excerpt(s) · 15.1s
Compare
4 excerpt(s) · 18.0s
Compare
GPT-6.1 Sol (ultra) · Codex · 828.4K context · Extended1 excerpt(s) · 8.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 6.0s
Compare
1 excerpt(s) · 5.0s
Compare
5 excerpt(s) · 19.0s
Compare
3 excerpt(s) · 15.1s
Compare
4 excerpt(s) · 18.0s
Compare
Haiku 5.5 (max) · Claude Code · Extended1 excerpt(s) · 8.0s
Compare
3 excerpt(s) · 18.0s
Compare
Not deliveredNo dedicated native map screen or minimap was delivered.
2 excerpt(s) · 10.0s
Compare
5 excerpt(s) · 19.1s
Compare
3 excerpt(s) · 15.0s
Compare
3 excerpt(s) · 13.5s
Compare

Capture inventory and review evidence