Phi-4
Small language model by Microsoft with strong reasoning for its size.
About
Phi-4 by Microsoft Research is a 14B parameter language model that reaches reasoning, math, and coding performance competitive with larger models. Its training extends the Phi-3 data recipe with synthetic textbook-style content for teaching reasoning, plus filtered web data and curated chat-format supervision, with a small share of multilingual data. It targets memory-constrained reasoning workloads. Released under the MIT license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- Large Language Models (LLMs)
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Easy (2/5)
- License
- MIT
- Minimum VRAM
- 8 GB
- Added
- Apr 3, 2026
Related Tools
Open-source code LLM family by IBM for enterprise code generation.
High-performance open-weight MoE LLM with 671B total parameters.
Hybrid SSM-Transformer model by AI21 Labs combining Mamba with attention layers.
State-space model for language with linear-time sequence modeling.
Family of small language models by Hugging Face for on-device use.
Open-weight LLM by Meta in 8B and 70B sizes with strong general capabilities.