跳到正文
智迹闻AuraTracer

发布于

Mistral Medium 3.5

Mistral 面向长流程编程、研究分析和跨工具任务的模型。

型号官方发布介绍

开始使用

NVIDIA 托管原型入口

Mistral 发布页列出该型号的 NVIDIA 托管原型入口;可否在线交互、账户、可用地区和价格以平台为准。

API 接入

Eden AI API文档
目录部署标识:mistral/mistral-medium-3-5。具体区域、权限与价格以服务商为准。

LLM Gateway API文档
目录部署标识:mistral/mistral-medium-3-5。具体区域、权限与价格以服务商为准。

公开成绩与计算依据

综合参考分根据已公开的评测资料整理。查看发布日期依据

综合参考分 20.8 · 16项基准

推理求解 15.1

GPQA Diamond · 0.748484848484849 比例

Artificial Analysis · 独立评测 · Diamond

Mistral Medium 3.5 · Current AA standalone protocol. Status verified in aa-method-current.md section gpqa-diamond

Terminal-Bench-Science · 0 比例

Artificial Analysis · 独立评测 · 0.1.0

Mistral Medium 3.5 · Artificial Analysis current published protocol; terminal-bench-science

HLE · 0.047 比例

Artificial Analysis · 独立评测 · Diamond

Mistral Medium 3.5 · Artificial Analysis current published protocol; hle-diamond

CritPt · 0.005 比例

Artificial Analysis · 独立评测 · 2 Hard

Mistral Medium 3.5 · Artificial Analysis current published protocol; critpt-v2-hard

工具与任务执行 19.3

tau3-Banking · 0.150515463917526 比例

Artificial Analysis · 独立评测 · 1.0.1

Mistral Medium 3.5 · Current AA standalone protocol. Status verified in aa-method-current.md section tau3-banking

GDPval-AA · 768.87 分

Artificial Analysis · 独立评测 · 2.1

Mistral Medium 3.5 · Artificial Analysis current published protocol; gdpval-aa-normalized

Terminal-Bench · 0 比例

Artificial Analysis · 独立评测 · 4.0

Mistral Medium 3.5 · Artificial Analysis current published protocol; terminalbench-4-0

AutomationBench · 0.06299648656993127 比例

Artificial Analysis · 独立评测 · current published task suite

Mistral Medium 3.5 · Artificial Analysis current published protocol; automationbench-aa

AA-Briefcase · 523.26 分

Artificial Analysis · 独立评测 · 1.1

Mistral Medium 3.5 · Artificial Analysis current published protocol; aa-briefcase

EnterpriseOps-Gym · 0.3372127723067741 比例

Artificial Analysis · 独立评测 · current published task suite

Mistral Medium 3.5 · Artificial Analysis current published protocol; enterprise-ops-gym-aa

AA-AnalystAgent · 0.125 比例

Artificial Analysis · 独立评测 · current AA pass^5

Mistral Medium 3.5 · Artificial Analysis current published protocol; aa-analyst-agent

编程 20.4

FrontierCode · 0.0176893939393939 比例

Artificial Analysis · 独立评测 · 2.0

Mistral Medium 3.5 · Artificial Analysis current published protocol; frontiercode-2-0

语言与信息处理 36.1

IFBench · 0.687755102040816 比例

Artificial Analysis · 独立评测 · current published task suite

Mistral Medium 3.5 · Artificial Analysis current published protocol; ifbench

GDP.pdf · 0.028 比例

Artificial Analysis · 独立评测 · current published task suite

Mistral Medium 3.5 · Artificial Analysis current published protocol; gdp-pdf

AA-Omniscience · 0.23033333333333333 比例

Artificial Analysis · 独立评测 · 1.1

Mistral Medium 3.5 · Artificial Analysis current published protocol; omniscience-accuracy

AA-Omniscience · 0.19792117799913378 比例

Artificial Analysis · 独立评测 · 1.1

Mistral Medium 3.5 · Artificial Analysis current published protocol; omniscience-non-hallucination

多模态理解 8.8

MMMU-Pro · 0.648554913294798 比例

Artificial Analysis · 独立评测 · Standard 10

Mistral Medium 3.5 · Artificial Analysis current published protocol; mmmu-pro

查看原始成绩与计算数据

资料来源

模型资料来源 · 目录许可