VoxelBench 实测:GPT-6.2 两款模型错误率创历史新高
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).
「模型」频道最新
- Domingos 吐槽 iWSP 论文:直接用程序,何必写进 LLM 权重 — pmddomingos · 2026-11-18
- GPT-6.2 Astra 首过「巴赫基准」:首次写出经过装饰的 4-3 挂留 — deanwball · 2026-11-18
- 西语博主上手 GPT-6.2:跑分领先还意外便宜 — ramos_casals · 2026-11-18
- Chutes 联手阿里云上线 GLM-5.3 与 DeepSeek V4.1,限时 65 折 — markjeffrey · 2026-11-18
- Meta 发布 Muse Spark Open:816B 参数 MoE 编码模型,上下文 100 万 token — PyTorch · 2026-11-18
- 博主实测 GPT-6.2 双模型:Sol 适合日用,Astra 贵且毛病多 — soumitrashukla9 · 2026-11-18