MazeBench: GPT-6.2 Sol narrows gap with Astra but still scores less at same price
patience_cave · x · 2026-11-20
Blogger patiencecave reports on his self-built MazeBench 3D open-world benchmark that GPT-6.2 Sol made strong progress over its predecessor and is approaching GPT-6.2 Astra's capabilities, though it still scores slightly worse than Astra at the same price. Accompanying maps show gpt-6.1-sol vs gpt-6.2-sol exploration traces, with 6.2 covering far more area. Self-built benchmark, unverified.
More from Models
- GPT-6.2 Astra matches predecessor's MazeBench score at $225 vs $840 — patience_cave · 2026-11-20
- Gemini 4 Neon leads MazeBench on uncached token efficiency by a wide margin — patience_cave · 2026-11-20
- MazeBench: GPT-6.2 Astra tops at 16% while costing 25% of its predecessor — patience_cave · 2026-11-20
- AssemblyAI STT 3.7 Tops Voice Arena Streaming ASR Leaderboard at 2.45% WER — AssemblyAI · 2026-11-20
- Transparency Isn't Objectivity: Community Debates Bias in Model Reviews — RichardsonDx · 2026-11-19
- Fireworks launches Specialized Intelligence Index scoring models on domain benchmarks, cost and duration — sophiamyang · 2026-11-19