跳过主要内容
评测来源/APEX-Agents 1.1
参与排名专业工作

APEX-Agents 1.1

Mercor · APEX-Agents 1.1 的公开模型评测结果。

官方评测
来源说明

该来源测量模型在「专业工作」领域的具体交付表现,原始成绩会和统一能力读数分开展示。

限制: 不同版本、推理设置和测试集可能导致分数变化,请结合数据时间阅读。

使用方式

参与排名的来源会影响能力分;参考、交叉参考和观察来源仅用于对比与核验。

本来源的模型成绩

Gemini 4 Carbon90.477.12026-11-11
Claude Fable 5.287.674.12026-11-10
Gemini 4 Argon87.473.92026-11-05
Claude Sonnet 5.583.769.92026-11-05
Claude Opus 5.581.367.42026-11-05
DeepSeek V4.1 Flash 101779.465.32026-11-05
GPT-6.1 Terra79.365.22026-11-05
Grok 4.677.363.12026-11-05
Claude Fable 5.177.162.82026-11-05
Claude Opus 57762.72026-11-05
Muse Wonder7762.72026-11-05
GPT-6 Astra75.160.72026-11-05
Claude Fable 574.560.02026-11-05
Qwen3.8 Max74.359.82026-11-05
GPT-6.1 Sol73.258.62026-11-05
Gemini 3.8 Flash72.958.32026-11-05
GPT-5.6 Terra71.857.12026-11-05
Grok 4.570.555.72026-11-05
Muse Spark 1.370.455.62026-11-05
GPT-5.570.355.52026-11-05
GPT-6 Sol68.753.82026-11-05
Grok 4.768.753.82026-11-05
Claude Haiku 5.568.353.42026-11-05
GPT-5.468.253.32026-11-05
MiniMax M3.1 Flash68.153.22026-11-05
DeepSeek V4 Flash67.552.52026-11-05
Claude Sonnet 567.452.42026-11-05
GLM-5.3 Flash66.851.82026-11-05
GLM-5.3 Flash66.851.82026-10-18
GLM-5.3 Flash66.851.82026-10-17
GLM 5.366.251.12026-11-05
MiMo V2.6 Pro65.350.12026-11-05
GPT-5.6 Sol65.250.02026-11-05
Claude Opus 4.864.449.22026-11-05
Mistral Large 464.349.12026-11-05
Gemini 3.6 Flash63.448.12026-11-05
Claude Opus 4.763.147.82026-11-05
DeepSeek V4 Pro 081362.346.92026-11-05
Kimi K362.246.82026-11-05
Claude Opus 4.661.846.42026-11-05
GLM-5.261.145.62026-11-05
Muse Spark 1.259.844.22026-11-05
GPT-6 Luna5943.42026-11-05
GPT-5.6 Luna57.842.12026-11-05
Claude Sonnet 4.657.641.92026-11-05
GLM-5.156.841.02026-11-05
MiniMax-M353.537.52026-11-05
Kimi K2.7 Code53.437.32026-11-05
Gemini 3.1 Pro Preview52.236.12026-11-05
DeepSeek V4.1 Flash50.534.22026-11-05
Inkling48.832.42026-11-05
Muse Spark 1.146.229.62026-11-05
Gemini 3.5 Flash Lite46.129.52026-11-05
Qwen3.5 397B-A17B40.323.32026-11-05
DeepSeek V3.235.117.72026-11-05
gpt-oss-120b8.30.02026-11-05

常见问题