GPT-6.1 Terra · Factual error rate on difficult prompts(Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified.) 7.7% · 主文证据(ev-gpt61-terra-chart-factuality-gpt61-terra-high)

GPT-6.1 Terra 的 Factual error rate on difficult prompts · Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified. 官方出处。GPT-6.1 Terra high · 7.7%。见《Introducing GPT-6.1 Terra | OpenAI》Official chart #factuality; exact point aria label; model GPT-6.1 Terra; effort High; embedded Vega data index 22。证据 ev-gpt61-terra-chart-factuality-gpt61-terra-high。

原文与定位

GPT-6.1 Terra · Factual error rate on difficult prompts(Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified.),GPT-6.1 Terra high配置,Responses with at least one factual error 7.7%(具体成绩公布时间未确认)。见《Introducing GPT-6.1 Terra | OpenAI》Official chart #factuality; exact point aria label; model GPT-6.1 Terra; effort High; embedded Vega data index 22。 查看成绩条件

Model: GPT-6.1 Terra; Effort: High; Cost: $0.036; Answers with any factual error: 7.7%
官方材料
Introducing GPT-6.1 Terra | OpenAI
定位
主文 · Official chart #factuality; exact point aria label; model GPT-6.1 Terra; effort High; embedded Vega data index 22
材料版本
Official GPT-6.1 Terra article; body and all six interactive charts reviewed; exact chart points cross-checked against official SVG aria-labels and embedded Vega values (143/143). The chart's launch-day inclusion is not verified.
版本公布
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
Wayback CDX has 2026-10-26 captures with HTTP 200. The 19:49:19Z replay was actually read (HTTP 200) and preserves the Terminal-Bench 4.0 max score in article prose; its interactive charts are empty. The earlier 18:05:14Z replay attempt returned HTTP 429.
引用文本

GPT-6.1 Terra · Factual error rate on difficult prompts(Internal error-inducing prompts in the GPT-6.1 Terra comparison article; exact dataset version, sample size and user-flagged provenance unspecified.),GPT-6.1 Terra high配置,Responses with at least one factual error 7.7%(具体成绩公布时间未确认)。官方出处 Introducing GPT-6.1 Terra | OpenAI(主文,Official GPT-6.1 Terra article; body and all six interactive charts reviewed; exact chart points cross-checked against official SVG aria-labels and embedded Vega values (143/143). The chart's launch-day inclusion is not verified.)Official chart #factuality; exact point aria label; model GPT-6.1 Terra; effort High; embedded Vega data index 22,https://openai.com/index/introducing-gpt-6-1-terra/。Benchlog https://benchlog.lekshome.cc/evidence/gpt-6-1-terra/ev-gpt61-terra-chart-factuality-gpt61-terra-high/

关联成绩与条件

GPT-6.1 Terra · Factual error rate on difficult prompts 成绩与配置
指标与成绩配置与评测条件证据关系
Responses with at least one factual error7.7%原文 7.7% · 单位 %
公布 未确认
GPT-6.1 Terra high
设置
GPT-6.1 Terra; reasoning effort high, directly labeled in the official chart.
未知项
Sampling and harness settings are not stated for this chart.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
Factual error rate on difficult prompts (lower is better); y metric Answers with any factual error; x axis Cost (USD). Lower is better; article describes an internal benchmark of error-inducing prompts without identifying the user-flagged dataset.
条件
GPT-6.1 Terra; reasoning effort High; chart cost $0.036.

直接支持该成绩

在排名中查看此成绩