Back to Models
microsoft logo
microsoft/phi-4
Not Available

Phi 4

[Microsoft Research](/microsoft) Phi-4 is designed to perform well in complex reasoning tasks and can operate efficiently in situations with limited memory or where quick responses are needed. At 14 billion parameters, it was trained on a mix of high-quality synthetic datasets, data from curated websites, and academic materials. It has undergone careful improvement to follow instructions accurately and maintain strong safety standards. It works best with English language inputs. For more information, please see [Phi-4 Technical Report](https://arxiv.org/pdf/2412.08905)

1/10/2025
16,384 tokens
#197 Text (Japanese)
Specifications

Modalities

Input
text
Output
text

Supported Parameters

frequency_penalty
max_tokens
min_p
presence_penalty
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
top_k
top_p

Max Output Tokens

16,384
Leaderboard
Text
๐Ÿ†OverallELO: 1,256
#282
๐Ÿ‡ฏ๐Ÿ‡ตJapaneseELO: 1,195
#197
๐Ÿ‡จ๐Ÿ‡ณChineseELO: 1,267
#267
๐Ÿ‡ฐ๐Ÿ‡ทKoreanELO: 1,196
#213
๐Ÿ‡ฌ๐Ÿ‡งEnglishELO: 1,272
#280
frenchELO: 1,265
#220
germanELO: 1,256
#217
spanishELO: 1,265
#218
russianELO: 1,245
#272
๐Ÿ’ปCodingELO: 1,306
#268
๐ŸงฎMathELO: 1,265
#252
โœ๏ธCreative WritingELO: 1,210
#283
๐Ÿ“Instruction FollowingELO: 1,245
#276
๐ŸŒถ๏ธHard PromptsELO: 1,278
#270
๐Ÿ’ฌMulti-TurnELO: 1,241
#276