Skip to main content

Phi-4

MicrosoftPhiOpen WeightMIT · Commercial OK

Description

phi-4 is a state-of-the-art open model built to excel at advanced reasoning, coding, and knowledge tasks. It leverages a blend of synthetic data, filtered web data, academic texts, and supervised fine-tuning for precision, alignment, and safety.

Release Date
2024-12-12
Parameters
14.7B
Context Length
16K
Modalities
text

Capability Radar

25
general
24
coding
30
reasoning
36
science
28
agents
0
multimodal

Rankings

Domain#RankScoreSource
Code Ranking508
10.0
AA
General Ranking502
21.0
AA
Science383
35.0
AA

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Biology

GPQANYU + Cohere + Anthropic (2023)65.8%SR

Code

HumanEvalOpenAI (2021)82.6%SR
LiveCodeBench53.8%SR

Creativity

Arena Hard73.3%SR

Factuality

SimpleQA3.0%SR

Finance

MMLU84.8%SR
MMLU-Pro74.3%SR

General

IFEvalGoogle Research (2023)83.4%SR
PhiBench70.6%SR
LiveBench47.6%SR

Long Context

FlenQA97.7%SR

Math

MGSM80.6%SR
MATH80.4%SR
OmniMath76.6%SR
DROP75.5%SR
AIME 202475.3%SR
AIME 202562.9%SR

Reasoning

HumanEval+92.9%SR

AA Evaluation Indices

(Artificial Analysis)
Math Index(Artificial Analysis)
18.0
Intelligence Index(Artificial Analysis)
4.6
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))
0.8
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.7
Gpqa(NYU + Cohere + Anthropic (2023))
0.6
Scicode(UIUC + Argonne National Lab (2024))
0.3
Ifbench(Google Research (2023))
0.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.2
Aime 25(MAA (Mathematical Association of America))
0.2
Aime(MAA (Mathematical Association of America))
0.1
Hle(Center for AI Safety + Scale AI (2025))
0.0
Terminalbench Hard(Stanford × Laude Institute (2026))
0.0
Lcr(Artificial Analysis)
0.0
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.0

LLM Stats Category Scores

(LLM Stats (zeroeval))
Legal
80
Language
80
Finance
80
Healthcare
80
Code
80
Creativity
80
Writing
80
Math
70
Reasoning
70
General
70
Physics
60
Structured Output
60
Instruction Following
60
Biology
60
Chemistry
60
Factuality
0

Pricing

Input Price$0.125 / 1M tokens
Output Price$0.5 / 1M tokens
Blended Price (3:1)$0.219 / 1M tokens

Speed

Tokens/sec40.9
Time to First Token0.57s
Time to Answer0.57s

Provider Price Ranking

Provider Price Ranking

5 providers

Cheapest: OpenRouterMost Expensive: Azure
ProviderInputOutput
1OpenRouterCheapest
$0.07
$0.14
2Kilo Gateway
$0.07
$0.14
3MicrosoftPRIMARY
$0.125
$0.5
4Azure Cognitive Services
$0.125
$0.5
5Azure
$0.125
$0.5

Compare pricing across different API providers for this model.

External Sources