GPT-6.1 Sol

OpenAI·7 个数据来源·更新于 2026-11-10

上下文窗口1.05M厂商官网
最大输出128K厂商官网
输入 / 百万 Token$2厂商官网
输出 / 百万 Token$10厂商官网

评测表现

只在同一来源、同一口径下比较。切换来源可查看不同测试结果。

6 个评测来源

当前配置:各基准设置见下表

采集于 2026-11-10查看原始页面

综合分数和同榜图表来自 LLM Stats 榜单(采集于 2026-11-10);下方逐项成绩来自模型页。榜单出处

LLM Stats · LLM Stats Score
53.1原榜第 10 名

同榜模型表现

LLM Stats Score · 同一来源与榜单口径,分数越高越好

7 个配置
所选配置其他配置
查看完整对比数据
模型与配置LLM Stats Score每任务成本
Kimi K3.1 Pro55.1未公布
Claude Mythos Preview53.9未公布
Gemini 4 Argon53.8未公布
GPT-6.1 Sol · 当前模型53.1未公布
Claude Opus 553未公布
Muse Wonder Full52.8未公布
Claude Fable 552.4未公布

不同来源的分数不能直接相加或横向比较;评测任务成本不是 API Token 单价。

逐项基准成绩

18 项已收录成绩 · 保留原始单位和测试条件

APEX-Agents来源汇总 · 越高越好
60%

Third-party Mercor APEX-Agents. Pass@1 overall 60.0% at Max. MC 58.4% / IB 58.1% Max; Pass@1 ±5.6. Single overall row only (no separate MC/IB bench IDs).

成绩出处
AutomationBench v1.0.6厂商自报 · 越高越好
36.1%

AutomationBench 1.0.6. OpenAI GPT-6.1 Sol launch interactive chart. Highest published score across the reported effort sweep: max effort (0.361). Other efforts: low 0.247, medium 0.317, high 0.332, xhigh 0.355. Evaluations used OpenAI research environments or API and may differ from production ChatGPT.

成绩出处
DeepSWE 1.1厂商自报 · 越高越好
75.22%

DeepSWE v1.1. OpenAI GPT-6.1 Sol launch interactive chart. Highest published score across the reported effort sweep: high effort (0.7522). Other efforts: low 0.6438, medium 0.7301, xhigh 0.719, max 0.719. Evaluations used OpenAI research environments or API and may differ from production ChatGPT.

成绩出处
FrontierCode 1.1来源汇总 · 越高越好
50.2%

Third-party Cognition / Epoch. FrontierCode 1.1 Main at medium reasoning effort (50.2%). Extended stored separately as frontiercode-1.1-extended.

成绩出处
FrontierCode 1.1 (Extended)来源汇总 · 越高越好
60.4%

Third-party Cognition / Devin. FrontierCode 1.1 Extended (60.4%). Main stored separately as frontiercode-1.1 (medium 50.2%).

成绩出处
FrontierMath Tier 4 (v2)来源汇总 · 越高越好
100%

Third-party Epoch AI Benchmarking Hub. FrontierMath Tier 4 (v2) at max reasoning effort (100%).

成绩出处
GDP.pdf厂商自报 · 越高越好
32%

gdp.pdf. OpenAI GPT-6.1 Sol launch interactive chart. Highest published score across the reported effort sweep: high effort (0.32). Other efforts: low 0.27, medium 0.30, xhigh 0.318, max 0.31. Evaluations used OpenAI research environments or API and may differ from production ChatGPT.

成绩出处
HealthBench厂商自报 · 越高越好
58.5%

HealthBench, length-adjusted score (unadjusted 0.567). GPT-6.1 Sol system card addendum Table 7.

成绩出处
HealthBench Consensus厂商自报 · 越高越好
96%

HealthBench Consensus, length-adjusted score (unadjusted 0.959). GPT-6.1 Sol system card addendum Table 7.

成绩出处
HealthBench Hard厂商自报 · 越高越好
36.2%

HealthBench Hard, length-adjusted score (unadjusted 0.334). GPT-6.1 Sol system card addendum Table 7.

成绩出处
HealthBench Professional厂商自报 · 越高越好
64.2%

HealthBench Professional, length-adjusted score (unadjusted 0.672). GPT-6.1 Sol system card addendum Table 7.

成绩出处
Humanity's Last Exam (no tools, text-only)来源汇总 · 越高越好
52.9%

Third-party Artificial Analysis AA HLE text-only (2,158 questions, no tools) at max reasoning effort (52.9%).

成绩出处
LifeSci Diamond (2026-11)来源汇总 · 越高越好
14.5%

Strict pass rate on 55 regulatory-affairs tasks. Plain API via Vercel AI Gateway with a web_search tool (Exa), provider default settings, one answer per task, graded by GPT-6.1 Sol at medium reasoning effort (Arca, 4 Nov 2026 edition).

成绩出处
MentalHealthBench厂商自报 · 越高越好
57.9%

MentalHealthBench overall at max reasoning effort (0.579). Acuity splits at max effort: non-acute 0.573, high acuity 0.591, emergent 0.580. GPT-6.1 Sol system card addendum Table 9.

成绩出处
MMMU-Pro来源汇总 · 越高越好
85.95%

Third-party Artificial Analysis (AA). MMMU-Pro at max reasoning effort (85.95%).

成绩出处
OSWorld 2.0厂商自报 · 越高越好
71.42%

OSWorld 2.0 offline set, partial reward (v2026.08.08), not strict completion or the full task set. OpenAI GPT-6.1 Sol launch interactive chart. Highest published score across the reported effort sweep: max effort (0.7142). Other efforts: low 0.5896, medium 0.6684, high 0.6956, xhigh 0.6938. Evaluations used OpenAI research environments or API and may differ from production ChatGPT.

成绩出处
SimpleQA Verified来源汇总 · 越高越好
73.9%

Third-party Epoch AI Benchmarking Hub. SimpleQA Verified at max reasoning effort (73.9%).

成绩出处
Terminal-Bench-Science 0.1厂商自报 · 越高越好
57.02%

Terminal-Bench Science 0.1. OpenAI GPT-6.1 Sol launch interactive chart. Highest published score across the reported effort sweep: max effort (0.5702). Other efforts: low 0.4371, medium 0.4756, high 0.5114, xhigh 0.5371. Evaluations used OpenAI research environments or API and may differ from production ChatGPT.

成绩出处

技术规格

只展示有可靠来源的资料;优先采用厂商官网与官方模型仓库。

发布日期
2026-09-29LLM Stats
知识截止
2026-04-30厂商官网
上下文窗口
1.05M Token厂商官网
最大输入
1.05M TokenLLM Stats
最大输出
128K Token厂商官网
输入模态
文本、图像厂商官网
输出模态
文本厂商官网
开放权重
否Artificial Analysis
推理模式
支持Artificial Analysis
许可证
ProprietaryLLM Stats

价格与用量

美元 / 百万 Token。不同服务商或测试配置的报价分别列出。

模型 API 价格,单位为美元每百万 Token
来源 / 服务商输入输出缓存读取
厂商官网 · Explore models and compare capabilities$2$10$0.1
OpenAI$2$10$0.1

估算一次用量

按表格第一条报价计算,普通文本、无缓存。

预计费用$3
不含工具、图片、推理额外计费或阶梯价格;以服务商实际账单为准。

每个数字,都有出处

同名模型通过开发机构和明确版本关联。预览版、日期版本和不同规模模型分别建档。

厂商官网

厂商官方文档 · 基础规格与标准 API 报价

1 条记录最近采集 2026-11-10

Artificial Analysis

公开评测数据;保留来源原始口径

10 条记录最近采集 2026-11-10

LLM Stats

公开模型页与公开榜单;自报成绩单独标识

2 条记录最近采集 2026-11-10

BenchLM

公开评测数据;保留来源原始口径

1 条记录最近采集 2026-11-10

Arena

Arena 官方公开数据集 · CC BY 4.0

4 条记录最近采集 2026-11-10

LiveBench

公开评测数据;保留来源原始口径

1 条记录最近采集 2026-11-10

OpenRouter

公开模型目录与 API 报价

1 条记录最近采集 2026-11-10
查看全部配置与采集记录
来源 / 配置采集时间原始数据
Artificial AnalysisGPT-6.1 Sol (max)SHA-256 c969b09538b1254d…2026-11-10T14:08:35Z查看来源
Artificial AnalysisGPT-6.1 Sol (xhigh)SHA-256 c969b09538b1254d…2026-11-10T14:08:35Z查看来源
Artificial AnalysisGPT-6.1 Sol (high)SHA-256 c969b09538b1254d…2026-11-10T14:08:35Z查看来源
Artificial AnalysisGPT-6.1 Sol (medium)SHA-256 c969b09538b1254d…2026-11-10T14:08:35Z查看来源
Artificial AnalysisGPT-6.1 Sol (low)SHA-256 c969b09538b1254d…2026-11-10T14:08:35Z查看来源
Artificial AnalysisGPT-6.1 Sol (Max)SHA-256 d88cf30c17a50945…2026-11-10T13:51:48Z查看来源
Artificial AnalysisGPT-6.1 Sol (High)SHA-256 d88cf30c17a50945…2026-11-10T13:51:48Z查看来源
Artificial AnalysisGPT-6.1 Sol (Low)SHA-256 d88cf30c17a50945…2026-11-10T13:51:48Z查看来源
Artificial AnalysisGPT-6.1 Sol (Medium)SHA-256 d88cf30c17a50945…2026-11-10T13:51:48Z查看来源
Artificial AnalysisGPT-6.1 Sol (Xhigh)SHA-256 d88cf30c17a50945…2026-11-10T13:51:48Z查看来源
LLM StatsGPT-6.1 Sol2026-11-10T14:08:28Z查看来源
LLM StatsGPT-6.1 SolSHA-256 8c8c3ecddc3a9ff1…2026-11-10T13:52:09Z查看来源
BenchLMGPT-6.1 Sol2026-11-10查看来源
ArenaGPT-6.1 Sol (Max)2026-11-10T14:07:52Z查看来源
ArenaGPT-6.1 Sol (Max)2026-11-10T14:07:52Z查看来源
ArenaGPT-6.1 Sol (Max)2026-11-10T14:07:52Z查看来源
ArenaGPT-6.1 Sol (Max)2026-11-10T14:07:52Z查看来源
LiveBenchGPT-6.1 Sol2026-11-10T14:08:34Z查看来源
OpenRouterGPT-6.1 SolSHA-256 63cbc45064d84060…2026-11-10T13:51:49Z查看来源
厂商官网Explore models and compare capabilitiesSHA-256 c7a891dfc83f874b…2026-11-10T13:59:32Z查看来源
继续浏览模型排行榜