Phi 4 Mini Reasoning — Hardware Requirements & GPU Compatibility
ChatMathCodeReasoningFunctions
Phi-4-mini-reasoning is Microsoft's compact 3.8-billion-parameter model in the Phi-4 family, fine-tuned specifically for multi-step, logic-intensive mathematical reasoning using synthetic math data distilled from a larger teacher model. It targets tasks like formal proof generation, symbolic computation, and advanced word problems, and is designed to run in memory- and latency-constrained environments such as educational tools or embedded tutoring systems, rather than for general knowledge or open-domain chat. It supports a 128K token context window and is released under the MIT license. At 3.8 billion parameters, it runs comfortably on almost any modern laptop or consumer GPU, even at higher precision, making it one of the more accessible reasoning-focused models to run locally.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 32K)Best on 24 GB (Q4_K_M · ~228 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~25 tok/s)
MIT
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Phi 4 Mini Reasoning
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
ollama run hf.co/lmstudio-community/Phi-4-mini-reasoning-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 19.2 GB free · 16K context
Download: Phi-4-mini-reasoning-Q4_K_M.gguf2.49 GBlmstudio-community
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (7 GB headroom) · Q4_K_M
- Generation speed
- ~82 tok/s
- generation speed
- Cost per 1M output tokens
- $0.28
- per 1M output tokens
How Much VRAM Does Phi 4 Mini Reasoning Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 2.2 GB | 6.2 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~298 tok/s | |
| 2.3 GB | 6.3 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~291 tok/s | |
| 2.4 GB | 6.5 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~269 tok/s | |
| 2.5 GB | 6.5 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 128K context · ~263 tok/s | |
| 2.9 GB | 6.9 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~228 tok/s | |
| 3.3 GB | 7.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~199 tok/s | |
| 3.7 GB | 7.8 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~176 tok/s | |
| 4.4 GB | 8.4 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~149 tok/s |
Selected Q4_K_M
4-bit medium quantization4.8GBat 16K context
- Weights
- 2.3 GB
- KV cache
- 2.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 19.3 GB free · ~228 tok/s
≥50 tok/s needs ≥221 GB/s · RTX 5090 ~406, RTX 3090 Ti ~228
Which Hardware Can Run Phi 4 Mini Reasoning?
Q4_K_M · 2.9 GBPhi 4 Mini Reasoning (Q4_K_M) requires 2.9 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 4+ GB is recommended. Using the full 131K context window can add up to 16.9 GB, bringing total usage to 19.8 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Phi 4 Mini Reasoning, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 3.3 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~172Simulate output on RTX 3080RTX 2080 Ti~140Simulate output on RTX 2080 TiRTX 3070 Ti~138Simulate output on RTX 3070 TiM3*8 GB~25Simulate output on iMac 24" M3 (8 GB)Orin Nano~23Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~17Simulate output on iPhone 1720 GB
fits · 15.3 GB free · up to 128K ctx
1 GPU
RX 7900 XT~167Simulate output on RX 7900 XT32–40 GB
fits · 27.3 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~406Simulate output on RTX 5090M5 Max36 GB~112Simulate output on Mac Studio M5 Max (36 GB)M532 GB~38Simulate output on MacBook Air 13" M5 (32 GB)Orin~46Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~31Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Phi 4 Mini Reasoning
Phi 4 family · 7 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite Vision 4.1 4BIBM | — | 4 GB | ~228 | same size |
| Qwen3 4B Instruct 2507Alibaba | — | 4.2 GB | ~226 | same size |
| MiniCPM V 4OpenBMB | — | 3.1 GB | ~236 | same size |
| Qwen2.5 VL 3B InstructAlibaba | — | 3.2 GB | ~249 | same size |
| CAJAL 4BAgnuxo | — | 4.8 GB | ~248 | same size |
| Gemma 4 E4B IT AssistantGoogle | — | 2.7 GB | ~243 | same size |
Fine-tunes & variants(2)
Specifications
- Publisher
- Microsoft
- Family
- Phi 4
- Parameters
- 3.8B
- Architecture
- Phi3ForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 200,064
- Release Date
- 2025-04-29
- License
- MIT
Frequently Asked Questions
How much VRAM does Phi 4 Mini Reasoning need?
Phi 4 Mini Reasoning requires 2.9 GB of VRAM at Q4_K_M, or 8.2 GB at BF16. Full 131K context adds up to 16.9 GB (19.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3.8B × 4.8 bits ÷ 8 = 2.3 GB
KV Cache + Overhead ≈ 0.6 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 17.5 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for Phi 4 Mini Reasoning?
For Phi 4 Mini Reasoning, Q4_K_M (2.9 GB) offers the best balance of quality and VRAM usage. Q4_K_L (2.9 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 1.6 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Phi 4 Mini Reasoning on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Phi 4 Mini Reasoning. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.9 GB of usable unified memory (RAM minus macOS overhead).
Can I run Phi 4 Mini Reasoning locally?
Yes — Phi 4 Mini Reasoning can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Phi 4 Mini Reasoning?
At Q4_K_M, Phi 4 Mini Reasoning can reach ~1673 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~228 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 ÷ 2.87 × 0.65 = ~1812 tok/s
Estimated speed at Q4_K_M (2.9 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Phi 4 Mini Reasoning?
At Q4_K_M, the download is about 2.30 GB. The full-precision BF16 version is 7.67 GB. The smallest option (IQ2_XXS) is 1.05 GB.
Which GPUs can run Phi 4 Mini Reasoning?
52 consumer GPUs can run Phi 4 Mini Reasoning at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Phi 4 Mini Reasoning?
163 devices with unified memory can run Phi 4 Mini Reasoning at Q4_K_M (2.9 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.