Skip to content
LMC Feed-Models, Papers, Benchmarks. Zero Fluff.Live

Mistral AI

18 个模型 · 平均评分 46

Last updated: 48m ago
模型18
平均评分46
最佳排名#154
顶级模型Mistral Medium 3.5

服务商优势

平均综合评分
46
最佳模型
Mistral Medium 3.5
Score 72
价格范围 (每百万输出)
$0.03 - $7.50
平均上下文窗口
161K
模型总数
18
功能覆盖
5/7
项功能

模型时间线

Apr 2026Mistral Medium 3.5Score 72
Mar 2026Mistral Small 4Score 40
Dec 2025Ministral 3 14B 2512Score 40
Dec 2025Ministral 3 8B 2512Score 40
Dec 2025Ministral 3 3B 2512Score 40
Dec 2025Mistral Large 3 2512Score 67
Oct 2025Voxtral Small 24B 2507Score 40
Aug 2025Mistral Medium 3.1Score 40
Aug 2025Codestral 2508Score 40
Jun 2025Mistral Small 3.2 24BScore 40
May 2025Mistral Medium 3Score 15
Mar 2025Mistral Small 3.1 24BScore 40
Feb 2025SabaScore 40
Jan 2025Mistral Small 3Score 40
Nov 2024Mistral Large 2407Score 66
Jul 2024Mistral NemoScore 40
Apr 2024Mixtral 8x22B InstructScore 63
Feb 2024Mistral LargeScore 66

价格分布

Free
0个模型
Budget
12个模型
Standard
6个模型
Premium
0个模型

功能覆盖

Vision
10/18
Function Calling
16/18
Reasoning
2/18
JSON Mode
17/18
Streaming
18/18
Web Search
0/18
Image Output
0/18

基准测试领先

Arena Elo
Mistral Medium 3.5
1427 elo
MMLU
Mistral Large
84.7%
MMLU-Pro
Mistral Large
69.4%
GPQA Diamond
Mistral Large
52.5%
HumanEval
Mistral Large
92%
MATH-500
Mistral Large
76%
IFEval
Mistral Large
86.5%

Verification & Source Coverage

把模型目录、服务状态和官方链接放在一起查看，能更快判断这个服务商页面是否可靠且最新。

目录校验时间
Aug 9, 2026

来自实时模型目录缓存。

状态校验时间
Aug 9, 2026

来自服务状态抓取。

最新发布
Apr 30, 2026

最早追踪到 Feb 26, 2024

来源覆盖
18 实时

无嵌入产品模型

APIDocsPricingPlayground

Operational Health

在定价和模型深度的同时追踪可靠性，使服务商选择建立在运营现实之上。

Full status page
当前状态
not monitored

Status monitoring via official APIs not yet available

30天正常运行
-

Rolling availability across the last 30 days.

稳定性评分
-

Composite reliability indicator from recent status behavior.

30天内事件
0

Includes degraded performance, partial outages, and major incidents.

Provider Catalog

Sort:
#ModelScore
154Mistral Medium 3.5
72
177Mistral Large 3 2512
67
184Mistral Large 2407
66
185Mistral Large
66
201Mixtral 8x22B Instruct
63
283Mistral Small 4
40
301Ministral 3 14B 2512
40
302Ministral 3 8B 2512
40
303Ministral 3 3B 2512
40
307Voxtral Small 24B 2507
40
322Mistral Medium 3.1
40
324Codestral 2508
40
331Mistral Small 3.2 24B
40
338Mistral Small 3.1 24B
40
346Saba
40
349Mistral Small 3
40
358Mistral Nemo
40
378Mistral Medium 3
15
18 models

Scoreboard

排名模型评分
#154Mistral Medium 3.572
#177Mistral Large 3 251267
#184Mistral Large 240766
#185Mistral Large66
#201Mixtral 8x22B Instruct63
#283Mistral Small 440
#301Ministral 3 14B 251240
#302Ministral 3 8B 251240
#303Ministral 3 3B 251240
#307Voxtral Small 24B 250740
#322Mistral Medium 3.140
#324Codestral 250840
#331Mistral Small 3.2 24B40
#338Mistral Small 3.1 24B40
#346Saba40
#349Mistral Small 340
#358Mistral Nemo40
#378Mistral Medium 315

类别分布

Coding
18

官方链接

PlaygroundAPIDocumentationPricing

相关

All providersCompare Mistral AI modelsFull leaderboardPricing comparisonHow benchmarks workAll benchmarksHow to choose a modelAI model familiesCost optimization guideAPI integration guidePrice changesRank changesFull leaderboard
Frequently Asked Questions

Mistral AI目前在我们平台上提供18个AI模型，平均质量评分为46/100。

Mistral AI评分最高的模型是Mistral Medium 3.5，总排名第154名，综合评分72/100。

Mistral AI模型的平均评分为46/100。他们最好的模型Mistral Medium 3.5全球排名第154名。访问我们的服务商对比页面进行详细分析。

Mistral AI Models - Rankings, Pricing & Benchmarks (2026) | LM Market Cap