Phi-2
Parameters
2.7B
Context Length
2K
Modality
Text
Architecture
Dense
License
MIT License
Release Date
12 Oct 2023
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
1x RTX 4090
24GB VRAM
Datacenter
1x NVIDIA A100
80GB VRAM
Apple Silicon
1x Apple M3 Max
128GB VRAM
2,048 tokens
Consumer
1x RTX 4090
24GB VRAM
Datacenter
1x NVIDIA A100
80GB VRAM
Apple Silicon
1x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
No evaluation benchmarks for Phi-2 available.
Rankings
Overall Rank
-
Coding Rank
-
About Phi-2
Phi-2 is Microsoft's 2.7B parameter small language model engineered for advanced reasoning, mathematical logic, and code generation. Trained on 1.4T textbook-quality tokens, it delivers benchmark performance rivaling significantly larger models.
Technical Specifications
Attention
Attention Structure
Multi-Head Attention
Attention Heads
32
Key-Value Heads
32
Attention Head Dimension
-
Position Embedding
ROPE
RoPE Theta
10,000
Sliding Window Attention
No
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
-
Linear Attention Ratio
-
Normalization
Layer Normalization
Activation Function
GELU
Dimensions
Hidden Dimension Size
2,048
Number of Layers
32
FFN Intermediate Size (Dense)
10,240
Multi-Token Prediction Heads
-
Tokenizer
Vocabulary Size
51,200
About Phi-2
Microsoft's Phi-2 is a 2.7 billion parameter Transformer-based model, developed for efficient language understanding and reasoning. Its technical innovations include training on "textbook-quality" synthetic and filtered web data, alongside scaled knowledge transfer from its predecessor, Phi-1.5, facilitating emergent capabilities within a compact architecture.
Other Phi-2 Models
- No related models available