测试版
排行榜/Phi 4
Microsoft logo

Phi 4

开源

来自 Microsoft · 发布于 2025-01-10

43.2
平均分
$0.07/1M
输入价格
$0.14/1M
输出价格
16K tokens (~8 books)
上下文窗口
text
类型

Tested on 16 benchmarks with 43.2% average. Top scores: Chatbot Arena Elo — Overall (1255.4%), MMLU (79.7%), IFEval (68.8%).

基准测试类别分数Bar
Chatbot Arena Elo — Overallarena1255.4
MMLUknowledge79.7
IFEvallanguage68.8
MATH level 5math64.9
Lech Mazur Writingknowledge62.6
BBH (HuggingFace)general55.3
MATH Level 5math50.0
MMLU-PROknowledge48.6
GPQA diamondknowledge41.4
OTIS Mock AIME 2024-2025math13.7
Balrogknowledge11.6
GPQAknowledge11.5
Artificial Analysis — Coding Indexspeed11.2
Artificial Analysis — Quality Indexspeed10.4
MUSRreasoning10.1
Artificial Analysis — Agentic Indexspeed0.0