WorldBox agentic sandbox doubles GPT-6 Astra's MazeBench score to 61 gems
Lianhuiq · x · 2026-10-16
EnactraAI launched WorldBox, an agentic sandbox for spatial reasoning that reportedly doubles GPT-6 Astra's MazeBench score from 23 to 61 gems. MazeBench, a brutal long-horizon spatial reasoning test of 256 interconnected rooms, saw even Opus 5.5 burn 2.2B tokens for just 13 of 100 gems. WorldBox gives agents a sandbox for long-term memory, inferring complex mechanics, and simulating reasoning. Academics hailed it as evidence that world models take reasoning to the next level.
Related event: WorldBox Sandbox Doubles GPT-6 Astra's MazeBench Score to 61(3 posts)→
More from Models
- OpenAI ships GPT-Live-1.1 voice update with new accents and voices in six languages — AndrewSchmidtFC · 2026-10-16
- GPT-4o API dies January 14 as OpenAI refuses to release weights, pushing users to Qwen — mimi10v3 · 2026-10-16
- Gemini 4 Argon goes live in Antigravity with its own weekly quota for Pro subscribers — rseroter · 2026-10-16
- Opus 5.5 value may end the habit of using the top model for everything — adonis_singh · 2026-10-16
- Musk says Grok Bot is growing faster than anything he's seen, with no free tier — elonmusk · 2026-10-16
- Peter Yang: Opus 5.5 Still Struggles to Hold Its Ground When Pushed Back — petergyang · 2026-10-16