Modeldex
Microsoft

Phi-4

CTX128K
42Quality
Text
TextInput$0.13/M
TextOutput$0.50/M

Open-weight instruction model for adaptable chat and self-hosted production workloads

Open weightsMIT453/1044
Drag to turnCard front showing

MicrosoftTextCommon

Phi-4

Open-weight instruction model for adaptable chat and self-hosted production workloads

phi-4
Type
Text
Quality
42 · Top 58% on the Epoch Capabilities Index
Context window
128,000 tokens
Max output
4,096 tokens
Input price
$0.13 / 1M tokens
Output price
$0.50 / 1M tokens
Cache read
—
Cache write
—
Parameters
15B (14,659,507,200)
Downloads
370K in the last 30 days
Likes
2.3K
Access to weights
Open download
Released
Dec 11, 2024
Knowledge cutoff
Oct 2023
Takes in
Text
Puts out
Text
Reasoning
No
Tool calling
No
Structured output
No
File attachments
No
Access
Open weights
License
MIT
Rarity
Common
Set number
453/1044

License read from the model’s Hugging Face card. Specs from models.dev, downloads from Hugging Face, last changed Oct 19, 2026. Official source: learn.microsoft.com.

How good is it?

Public leaderboard scores for Phi-4, refreshed daily from Epoch AI and LMArena.

Top 58% of the 277 models on the Epoch Capabilities Index: #160.

Quality is the share of a leaderboard’s models this one beats, from 0 to 100. Chat models are rated on Epoch AI’s Capabilities Index when it lists them, otherwise on LMArena; image and video models on LMArena’s image and video arenas.

Leaderboards

Benchmarks Epoch AI runs

Scores from Epoch AI and LMArena, licensed CC BY 4.0. Boards list models under their own names, so a close match is shown with the name it was scored under. Credits

Where to run it

4 providers sell Phi-4 through an API, and 2 Hugging Face hosts serve it. Prices are per million tokens, cheapest first.

API providers

ProviderInputOutputContext
Kilo GatewayCheapest paidmicrosoft/phi-4$0.056$0.1116K
OpenRoutermicrosoft/phi-4$0.07$0.1416K
AzureOfficialphi-4$0.13$0.50128K
Azure Cognitive Servicesphi-4$0.13$0.50128K

Hugging Face Inference Providers

HostInputOutputContextSpeedFirst token
DeepInfraFastest$0.07$0.1416K78 tok/s0.3s
Featherless—————

From models.dev and Hugging Face, checked daily. Resellers’ listings can lag behind their price pages, so confirm with the provider before you commit.

Face off

Current text models from other labs at a similar output price.

More from Microsoft

Microsoft

harrier-oss-v1-270m

SIZE268M
Embedding
TextDownloads130K/mo
–Likes210

Embedding model from Microsoft, 268M parameters.

Open weightsMIT843/1044
Microsoft

harrier-oss-v1-0.6b

SIZE596M
Embedding
TextDownloads280K/mo
–Likes320

Embedding model from Microsoft, 596M parameters.

Open weightsMIT842/1044
Microsoft

VibeVoice-ASR

SIZE8.7B
Speech to text
AudioDownloads690K/mo
TextLikes1.3K

Speech recognition model from Microsoft, 8.7B parameters.

Open weightsMIT757/1044
Microsoft

VibeVoice-Realtime-0.5B

SIZE1B
Text to speech
TextDownloads290K/mo
AudioLikes1.3K

Text-to-speech model from Microsoft, 1B parameters.

Open weightsMIT719/1044
Microsoft

TRELLIS.2-4B

3D generation
ImageDownloads2.3M/mo
–Likes1.4K

Image-to-3D model from Microsoft.

Open weightsMIT712/1044
Microsoft

VibeVoice-1.5B

SIZE2.7B
Text to speech
TextDownloads540K/mo
AudioLikes2.5K

Text-to-speech model from Microsoft, 2.7B parameters.

Open weightsMIT644/1044
Microsoft

Phi-4-reasoning-plus

CTX32K
Reasoning
TextInput$0.13/M
TextOutput$0.50/M

Reasoning model for deliberate analysis, multi-step problem solving, and tool use

Open weightsMIT458/1044
Microsoft

Phi-4-reasoning

CTX32K
Reasoning
TextInput$0.13/M
TextOutput$0.50/M

Reasoning model for deliberate analysis, multi-step problem solving, and tool use

Open weightsMIT457/1044
Deck0/4
  1. Empty slot
  2. Empty slot
  3. Empty slot
  4. Empty slot