AA 实测:GPT-6.1 Terra 幻觉率降 50%,准确率还升 2 分
ArtificialAnlys · x · 2026-10-27
Artificial Analysis data shows GPT-6.1 Terra drops its hallucination rate by 50 points on the AA-Omniscience benchmark while raising accuracy by 2 points — answering more questions correctly while admitting ignorance more often.
The charts also compare rivals: Gemini 4 Argon has the lowest hallucination rate (15%), Claude Opus 5.5 the highest accuracy (67%), while GPT-6 Luna Preview hallucinates a staggering 88% of the time.
「模型」频道最新
- KOL 吐槽 OpenAI 6.1 Sol 被评 B 级:发布策略搞砸了 — Angaisb_ · 2026-10-27
- php2shell 漏洞实测：仅 2/5 前沿模型能完成利用移植，耗时 3.5-6 小时 — xeophon · 2026-10-27
- Qwen Code 发布 v0.26.2 预览版：新增会话权限规则与实验性共享 Agent Loop — qwen-code-review-bot · 2026-10-27
- GPT-6.1 Terra 事实错误率从 22.3% 降至 4.3%，降幅近 81% — haider1 · 2026-10-27
- GPT-6.1 Terra 发布:早期用户称写作与 computer use 表现亮眼 — BenBajarin · 2026-10-27
- GPT-6.1 Terra 跑分曝光:成本低至 Sol 一半,追不平 Astra — Angaisb_ · 2026-10-27