Phi-4 14B
Phi · MIT
Microsoft's efficient reasoning and coding model with high performance per parameter.
- Parameters
- 14.7B
- Architecture
- Dense
- Context
- 32,768 tokens
- Released
- 2025-02-26
- Engines
- llama.cpp, ollama, vLLM, TGI
- Builder Tools
- Cursor, Continue, Aider, Open WebUI, LM Studio
Parameters
14.7B
VRAM
12.6 GB
Context
32K
Formats
5
GPUs
26
Phi-4 14B (14.7B) requires 12.6 GB VRAM at recommended quality (Q6_K). At efficient quality (Q4_K_M), it fits in 8.4 GB VRAM, making it compatible with the NVIDIA RTX 4060 Laptop (40-60W). On NVIDIA Grace Blackwell Ultra GB300, expect approximately 220 tok/s at Q8_0. For the best experience, Budget Home AI Server ($1,162) is recommended.
Source: OwnRig methodology
12.6 GB
Q6_K
11 GB
32K tokens
Reasoning
VRAM Requirements
| Quality | Quantization | VRAM | File Size |
|---|---|---|---|
| full | Q8_0 | 16.2 GB | 14.5 GB |
| recommended | Q6_K | 12.6 GB | 11 GB |
| recommended | Q5_K_M | 10.5 GB | 9 GB |
| efficient | Q4_K_M | 8.4 GB | 7.2 GB |
| compressed | Q3_K_M | 6.8 GB | 5.8 GB |
Context Length Impact
KV cache VRAM at Q6_K quality. Longer context = more memory.
| Context | KV Cache | Total VRAM |
|---|---|---|
| 2K | 205 MB | 12.8 GB |
| 4K | 512 MB | 13.1 GB |
| 8K | 1 GB | 13.6 GB |
| 16K | 1.9 GB | 14.5 GB |
| 32K | 3.8 GB | 16.4 GB |
Compatible GPUs
26 devicesShowing 26 of 26 entries
Builder Context
Phi-4 14B is commonly used with Cursor, Continue, Aider, Open WebUI, LM Studio. For an AI coding workflow, pair it with an embedding model like nomic-embed-text for local RAG.
Recommended Builds
Complete PC builds that can run Phi-4 14B.
Frequently Asked Questions
- How much VRAM does Phi-4 14B need?
- Phi-4 14B requires 12.6 GB VRAM at recommended quality (Q6_K). At lower quality settings, it can fit in as little as 6.8 GB.
- What is the best GPU for Phi-4 14B?
- The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Phi-4 14B, achieving 220 tok/s at Q8_0 with an excellent rating.
- Can I run Phi-4 14B on an RTX 4060 Ti?
- Yes. On the NVIDIA GeForce RTX 4060 Ti 16GB, Phi-4 14B runs at 28 tok/s (Q4_K_M, good).
- What quantization should I use for Phi-4 14B?
- For the best quality, use Q6_K (12.6 GB VRAM). If your GPU has limited VRAM, Q3_K_M (6.8 GB) is the most efficient option with acceptable quality.
- Is Phi-4 14B good for coding?
- Yes. Phi-4 14B is used with Cursor, Continue, Aider, Open WebUI, LM Studio for local AI coding. For the best coding experience, pair it with an embedding model for local RAG.
Related Guides
Tutorial
The Complete Guide to Running LLMs Locally
Run large language models locally: hardware needs, Ollama and llama.cpp, model picks by use case, and quantization.
Explainer
VRAM: The Only Spec That Matters for AI
VRAM for local AI: what it is, why models need it, how quantization cuts requirements, and a VRAM table for major models.
Data confidence: estimated. Source
VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Phi is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.