Phi 4 — Hardware Requirements & GPU Compatibility
ChatMathCode
Microsoft Phi 4 is a 14-billion parameter language model from Microsoft Research's Phi series, designed to deliver strong reasoning, mathematical, and coding performance at an efficient size. Phi 4 continues the Phi family's focus on maximizing capability per parameter through high-quality training data curation, achieving benchmark scores that rival much larger models on reasoning and STEM tasks. The model runs well on consumer GPUs with 12-16GB of VRAM in quantized formats. It excels at mathematical problem solving, code generation, and structured reasoning. Released under the MIT license.
814.3K downloads
2.3K likesSpecifications
- Publisher
- Microsoft
- Family
- Phi 4
- Parameters
- 14.7B
- Architecture
- Phi3ForCausalLM
- Context Length
- 16,384 tokens
- Vocabulary Size
- 100,352
- Release Date
- 2024-12-11
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does Phi 4 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 7.0 GB | 9.9 GB | 6.23 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 7.1 GB | 10.1 GB | 6.41 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 7.9 GB | 10.8 GB | 7.15 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 8.1 GB | 11.0 GB | 7.33 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 9.5 GB | 12.4 GB | 8.80 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 11.2 GB | 14.1 GB | 10.44 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 12.8 GB | 15.8 GB | 12.09 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 15.4 GB | 18.3 GB | 14.66 GB | 8-bit quantization, near-lossless |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Phi 4?
Q4_K_M · 9.5 GBPhi 4 (Q4_K_M) requires 9.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 13+ GB is recommended. Using the full 16K context window can add up to 2.9 GB, bringing total usage to 12.4 GB. 39 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run Phi 4?
Q4_K_M · 9.5 GB49 devices with unified memory can run Phi 4, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, iPad Pro M5 13" (16 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightWhere to Download Phi 4
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Benchmarks
Benchmark details →Related Models
Frequently Asked Questions
- How much VRAM does Phi 4 need?
Phi 4 requires 9.5 GB of VRAM at Q4_K_M, or 30.0 GB at BF16. Full 16K context adds up to 2.9 GB (12.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 14.7B × 4.8 bits ÷ 8 = 8.8 GB
KV Cache + Overhead ≈ 0.7 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 3.6 GB (at full 16K context)
VRAM usage by quantization
Q4_K_M9.5 GBQ4_K_M + full context12.4 GB- Can NVIDIA GeForce RTX 4090 run Phi 4?
Yes, at Q8_0 (15.4 GB) or lower. Higher quantizations like BF16 (30.0 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for Phi 4?
For Phi 4, Q4_K_M (9.5 GB) offers the best balance of quality and VRAM usage. Q4_K_L (9.7 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XS at 5.1 GB.
VRAM requirement by quantization
IQ2_XS5.1 GBQ3_K_S7.1 GBQ4_K_S9.0 GBQ4_K_M ★9.5 GBQ5_K_S10.8 GBBF1630.0 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Phi 4 on a Mac?
Phi 4 requires at least 5.1 GB at IQ2_XS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Phi 4 locally?
Yes — Phi 4 can run locally on consumer hardware. At Q4_K_M quantization it needs 9.5 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Phi 4?
At Q4_K_M, Phi 4 can reach ~462 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~69 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 9.5 × 0.65 = ~546 tok/s
Estimated speed at Q4_K_M (9.5 GB)
~546 tok/s~69 tok/s~546 tok/s~462 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Phi 4?
At Q4_K_M, the download is about 8.80 GB. The full-precision BF16 version is 29.32 GB. The smallest option (IQ2_XS) is 4.40 GB.
- Which GPUs can run Phi 4?
39 consumer GPUs can run Phi 4 at Q4_K_M (9.5 GB). Top options include AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 6900 XT, AMD Radeon RX 6700 XT. 26 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Phi 4?
52 devices with unified memory can run Phi 4 at Q4_K_M (9.5 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.