LLMs · Microsoft
Phi-4 Mini (3.8B)
- Parameters
- 3.8B
- Architecture
- Dense
- Max context
- 128K tokens
- License
- mit
- Released
- 2025-02
How to run it
Quickest · Ollama
One command downloads a ready-to-run version and starts a chat.
ollama run phi4-mini:3.8bQuantized files · GGUF
Works in LM Studio, llama.cpp, Jan and Ollama. No sign-in needed.
huggingface.co/unsloth/Phi-4-mini-instruct-GGUF (opens in a new tab)
ollama run hf.co/unsloth/Phi-4-mini-instruct-GGUFOriginal weights · Hugging Face
Full-precision files from Microsoft, for fine-tuning, vLLM, Transformers.
huggingface.co/microsoft/Phi-4-mini-instruct (opens in a new tab)
Official repo · GitHub
Reference code, inference scripts and docs from Microsoft. The weights themselves download from Hugging Face.
github.com/microsoft/PhiCookBook (opens in a new tab)
git clone https://github.com/microsoft/PhiCookBookMemory needed
Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps.
| Quantization | Quality | Weights | Total at 8K context | Total at 32K context | Download |
|---|---|---|---|---|---|
Q2_K |
Low | 1.5 GB | 3.2 GB | 6.4 GB | GGUF (opens in a new tab) |
Q3_K_M |
Low | 1.9 GB | 3.6 GB | 6.8 GB | GGUF (opens in a new tab) |
Q4_K_M |
Good | 2.3 GB | 4.0 GB | 7.2 GB | GGUF (opens in a new tab) |
Q5_K_M |
High | 2.7 GB | 4.4 GB | 7.6 GB | GGUF (opens in a new tab) |
Q6_K |
High | 3.1 GB | 4.8 GB | 8.0 GB | GGUF (opens in a new tab) |
Q8_0 |
Very high | 4.1 GB | 5.7 GB | 9.0 GB | GGUF (opens in a new tab) |
F16 |
Full | 7.7 GB | 9.4 GB | 12.6 GB | GGUF (opens in a new tab) Original weights (opens in a new tab) |
Best hardware for Phi-4 Mini (3.8B)
The least expensive GPU in our catalog that runs it without offloading is the Intel Arc A380 (launch price $139) at Q4_K_M.
| Grade · points | Hardware | Quant | Memory | Speed | Fit |
|---|---|---|---|---|---|
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 15.0/15 | NVIDIA GeForce RTX 4060 | Q8_0 | 6.1 GB / 8.0 GB | 44.3 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | AMD Radeon RX 9070 XT | F16 | 10.2 GB / 16.0 GB | 41.3 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 4070 Ti SUPER | F16 | 10.2 GB / 16.0 GB | 50.4 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 4080 SUPER | F16 | 10.2 GB / 16.0 GB | 55.2 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 5070 Ti | F16 | 10.2 GB / 16.0 GB | 67.2 tok/s | Runs well |
| 74 ptsGreat (A) · 74 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 14.6/15 | NVIDIA GeForce RTX 5080 | F16 | 10.2 GB / 16.0 GB | 72.0 tok/s | Runs well |
| 72 ptsGreat (A) · 72 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.2/15 | Intel Arc B580 | Q8_0 | 6.3 GB / 12.0 GB | 56.7 tok/s | Runs well |
| 72 ptsGreat (A) · 72 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.2/15 | NVIDIA GeForce RTX 3060 12GB | Q8_0 | 6.3 GB / 12.0 GB | 58.6 tok/s | Runs well |
| 72 ptsGreat (A) · 72 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 12.2/15 | NVIDIA GeForce RTX 4070 SUPER | Q8_0 | 6.3 GB / 12.0 GB | 82.0 tok/s | Runs well |
| 69 ptsGood (B) · 69 of 100 points: Model quality 19.8/45, Memory fit 25.0/25, Speed 15.0/15, Memory use 10.1/15 | AMD Radeon RX 7900 XTX | F16 | 10.6 GB / 24.0 GB | 62.0 tok/s | Runs well |
Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.