GPT-6.2 Astra and Sol post highest error rates ever on VoxelBench 2
legit_api · x · 2026-11-18
Third-party evaluator voxelbench reports that OpenAI's new GPT-6.2 Astra and GPT-6.2 Sol models recorded the highest error rates of any model so far on VoxelBench 2: 19 for Astra and a striking 62 for Sol.
Most failures are dumb mistakes rather than capability gaps — syntax errors or forgetting to define something for something. A screenshot shows Sol's full 350-generation run still in progress (286 completed, 62 failed).
More from Models
- Astra 6.2 plays detective: finds URL on a projector screen in hours of B-roll — gabrielchua · 2026-11-18
- Qwen 3.8 27B vs Flash-Next vs DeepSeek V4.1 Flash: one-shot landing page build-off — jesdga95 · 2026-11-18
- GPT-6.2 Astra plays RimWorld for 250 in-game days and builds a decent colony — jxnlco · 2026-11-18
- Polymarket claims GPT-6 Cyber uncovered 35 security flaws across Apple iPhones and Macs — Polymarket · 2026-11-18
- OpenAI ships GPT-6.2 Astra: 22-point ExploitBench jump, tighter cyber safeguards — charliermarsh · 2026-11-18
- Halocline inference engine hits 148-182 tok/s on Qwen-Flash-Next with KLD 0.082 — Distinct-Pie2389 · 2026-11-18