GPT-6.1 Sol · 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. · 成绩与配置
GPT-6.1 Sol 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. 记录。本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录),附 5 条来源定位。
GPT-6.1 Sol · Factual error rate on difficult prompts 成绩与配置
指标与成绩
配置与评测条件
出处
Responses with at least one factual error14.7%原文 14.7% · 单位 %
公布
GPT-6.1 Sol low
设置
GPT-6.1 Sol; reasoning effort low as directly labeled in the official chart.
未知项
未另记
工具
工具条件未记录
推理预算
low
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.
Responses with at least one factual error12%原文 12.0% · 单位 %
公布
GPT-6.1 Sol medium
设置
GPT-6.1 Sol; reasoning effort medium as directly labeled in the official chart.
未知项
未另记
工具
工具条件未记录
推理预算
medium
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.
Responses with at least one factual error6.6%原文 6.6% · 单位 %
公布
GPT-6.1 Sol high
设置
GPT-6.1 Sol; reasoning effort high as directly labeled in the official chart.
未知项
未另记
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.
Responses with at least one factual error5.5%原文 5.5% · 单位 %
公布
GPT-6.1 Sol xhigh
设置
GPT-6.1 Sol; reasoning effort xhigh as directly labeled in the official chart.
未知项
未另记
工具
工具条件未记录
推理预算
xhigh
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.
Responses with at least one factual error4.4%原文 4.4% · 单位 %
公布
GPT-6.1 Sol max
设置
GPT-6.1 Sol; reasoning effort max as directly labeled in the official chart.
未知项
未另记
工具
工具条件未记录
推理预算
max
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.
GPT-6.1 Sol · 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. · Scores and configurations
GPT-6.1 Sol record for 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.. 5 score records for this release, including 5 exact values (may include historical records). Includes 5 source locations.
GPT-6.1 Sol · Factual error rate on difficult prompts scores and configurations
Metric and score
Configuration and evaluation conditions
Sources
Responses with at least one factual error14.7%Reported 14.7% · Unit %
Published
GPT-6.1 Sol low
Settings
GPT-6.1 Sol; reasoning effort low as directly labeled in the official chart.
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
low
Framework
not recorded
Protocol
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.
Responses with at least one factual error12%Reported 12.0% · Unit %
Published
GPT-6.1 Sol medium
Settings
GPT-6.1 Sol; reasoning effort medium as directly labeled in the official chart.
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
medium
Framework
not recorded
Protocol
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.
Responses with at least one factual error6.6%Reported 6.6% · Unit %
Published
GPT-6.1 Sol high
Settings
GPT-6.1 Sol; reasoning effort high as directly labeled in the official chart.
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
high
Framework
not recorded
Protocol
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.
Responses with at least one factual error5.5%Reported 5.5% · Unit %
Published
GPT-6.1 Sol xhigh
Settings
GPT-6.1 Sol; reasoning effort xhigh as directly labeled in the official chart.
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
xhigh
Framework
not recorded
Protocol
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.
Responses with at least one factual error4.4%Reported 4.4% · Unit %
Published
GPT-6.1 Sol max
Settings
GPT-6.1 Sol; reasoning effort max as directly labeled in the official chart.
Unknowns
none recorded
Tools
Tool access not recorded
Reasoning budget
max
Framework
not recorded
Protocol
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.