Back

Microsoft

Phi-4

Open weightsHugging Face
Intelligence
5.9
Rank 640 of 744 measured models
Output speed
41.4 t/s
Rank 358 of 403 measured models
Price/1M
$0.219
Rank 177 of 659 measured models
Context window
16K
Rank 1100 of 1199 measured models

Phi-4 scores 5.9 on the Artificial Analysis Intelligence Index, ranking 640th of 744 measured models. The closest models on this index are Claude 3 Sonnet (5.9), Gemini 1.5 Flash (May '24) (5.9), and Granite 4.1 3B (5.9).

Artificial Analysis Indices
Intelligence5.9
Rank 640 of 744 measured models
Math18.0
Rank 215 of 270 measured models
Standard Benchmarks
MMLU Pro71.4%
Rank 212 of 349 measured models
GPQA Diamond57.5%
Rank 424 of 619 measured models
HLE3.8%
Rank 593 of 664 measured models
LiveCodeBench23.1%
Rank 263 of 343 measured models
MATH-50081.0%
Rank 106 of 202 measured models
AIME 202414.3%
Rank 114 of 194 measured models
AIME 202518.0%
Rank 215 of 270 measured models
IFBench23.5%
Rank 424 of 450 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy