Can I run Phi 4 Mini Instruct locally?
Phi 4 Mini Instruct by microsoft has 3.8B parameters. At 4-bit with an 8K context it needs about 4.1 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track. See all models that fit in 8 GB of VRAM.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from bartowski/microsoft_Phi-4-mini-instruct-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | rough | 2.8 GB | 3.3 GB | 6.3 GB | 18.3 GB |
| 3-bit (Q3_K_M) • | usable | 3.3 GB | 3.8 GB | 6.8 GB | 18.8 GB |
| 4-bit (Q4_K_M) • | good, most popular | 3.6 GB | 4.1 GB | 7.1 GB | 19.1 GB |
| 5-bit (Q5_K_M) • | very good | 3.9 GB | 4.4 GB | 7.4 GB | 19.4 GB |
| 6-bit (Q6_K) • | near lossless | 4.2 GB | 4.7 GB | 7.7 GB | 19.7 GB |
| 8-bit (Q8_0) • | effectively lossless | 5.1 GB | 5.6 GB | 8.6 GB | 20.6 GB |
| 16-bit (F16/BF16) | original weights | 8.4 GB | 8.9 GB | 11.9 GB | 23.9 GB |
Phi 4 Mini Instruct on Ollama: download size and VRAM
ollama run phi4-mini:3.8b | Source | File | Download size | GPU memory at 8K ctx (est.) |
|---|---|---|---|
| Ollama download size | phi4-mini:3.8b (Q4_K_M) | 2.5 GB | 4.1 GB |
| Our GGUF (Hugging Face) | bartowski/microsoft_Phi-4-mini-instruct-GGUF (Q4_K_M) | 2.5 GB | 4.1 GB |
Download sizes are decimal GB, as ollama.com lists them; memory figures on this site use binary GB (GiB). GPU memory for the Ollama tag is its file size plus this model's KV cache at 8K context and runtime overhead; Ollama's default context window may be smaller (set num_ctx to change it). Source: ollama.com/library/phi4-mini:3.8b, checked Nov 17, 2026. Refreshed weekly; shown only when the tag's architecture and parameter count match this model. All Ollama models with sizes and VRAM.
Which GPUs and Macs run Phi 4 Mini Instruct?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
| Hardware | Memory | Fits? | Best quant | Needs | Max tok/s (est.) | Details |
|---|---|---|---|---|---|---|
| RTX 3060 12GB | 12 GB | Fits | F16 | 8.9 GB | 47 | details |
| RTX 4060 Ti 16GB | 16 GB | Fits | F16 | 8.9 GB | 38 | details |
| RTX 3090 | 24 GB | Fits | F16 | 8.9 GB | 122 | details |
| RTX 4090 | 24 GB | Fits | F16 | 8.9 GB | 131 | details |
| RTX 5090 | 32 GB | Fits | F16 | 8.9 GB | 234 | details |
| RTX 5070 Ti | 16 GB | Fits | F16 | 8.9 GB | 117 | details |
| RX 7900 XTX | 24 GB | Fits | F16 | 8.9 GB | 125 | details |
| Mac M4 Pro 48GB | 48 GB | Fits | F16 | 8.9 GB | 36 | details |
| Mac M4 Max 128GB | 128 GB | Fits | F16 | 8.9 GB | 71 | details |
| Mac M3 Ultra 512GB | 512 GB | Fits | F16 | 8.9 GB | 107 | details |
| DGX Spark | 128 GB | Fits | F16 | 8.9 GB | 36 | details |
| RTX 4070 | 12 GB | Fits | F16 | 8.9 GB | 66 | details |
| RTX 5060 Ti 16GB | 16 GB | Fits | F16 | 8.9 GB | 58 | details |
| RTX 4060 | 8 GB | Fits | Q8_0 | 5.6 GB | 67 | details |
| Arc B580 | 12 GB | Fits | F16 | 8.9 GB | 59 | details |
| RTX 3060 Ti | 8 GB | Fits | Q8_0 | 5.6 GB | 110 | details |
| RTX 3070 | 8 GB | Fits | Q8_0 | 5.6 GB | 110 | details |
| RTX 3080 10GB | 10 GB | Fits | Q8_0 | 5.6 GB | 186 | details |
| RTX 3080 12GB | 12 GB | Fits | F16 | 8.9 GB | 119 | details |
| RTX 3090 Ti | 24 GB | Fits | F16 | 8.9 GB | 131 | details |
| RTX 4060 Ti 8GB | 8 GB | Fits | Q8_0 | 5.6 GB | 71 | details |
| RTX 4070 Super | 12 GB | Fits | F16 | 8.9 GB | 66 | details |
| RTX 4070 Ti | 12 GB | Fits | F16 | 8.9 GB | 66 | details |
| RTX 4070 Ti Super | 16 GB | Fits | F16 | 8.9 GB | 88 | details |
| RTX 4080 | 16 GB | Fits | F16 | 8.9 GB | 93 | details |
| RTX 4080 Super | 16 GB | Fits | F16 | 8.9 GB | 96 | details |
| RTX 5060 | 8 GB | Fits | Q8_0 | 5.6 GB | 110 | details |
| RTX 5070 | 12 GB | Fits | F16 | 8.9 GB | 88 | details |
| RTX 5080 | 16 GB | Fits | F16 | 8.9 GB | 125 | details |
| 2x RTX 3090 | 2×24 GB | Fits | F16 | 8.9 GB | 122 | details |
| 2x RTX 4090 | 2×24 GB | Fits | F16 | 8.9 GB | 131 | details |
| 2x RTX 5090 | 2×32 GB | Fits | F16 | 8.9 GB | 234 | details |
| RTX A6000 | 48 GB | Fits | F16 | 8.9 GB | 100 | details |
| RTX 6000 Ada | 48 GB | Fits | F16 | 8.9 GB | 125 | details |
| RTX PRO 6000 Blackwell | 96 GB | Fits | F16 | 8.9 GB | 234 | details |
| L40S | 48 GB | Fits | F16 | 8.9 GB | 113 | details |
| A100 40GB | 40 GB | Fits | F16 | 8.9 GB | 203 | details |
| A100 80GB | 80 GB | Fits | F16 | 8.9 GB | 266 | details |
| H100 PCIe | 80 GB | Fits | F16 | 8.9 GB | 261 | details |
| H100 SXM | 80 GB | Fits | F16 | 8.9 GB | 437 | details |
| H200 | 141 GB | Fits | F16 | 8.9 GB | 626 | details |
| RX 7600 | 8 GB | Fits | Q8_0 | 5.6 GB | 71 | details |
| RX 7700 XT | 12 GB | Fits | F16 | 8.9 GB | 56 | details |
| RX 7800 XT | 16 GB | Fits | F16 | 8.9 GB | 81 | details |
| RX 7900 GRE | 16 GB | Fits | F16 | 8.9 GB | 75 | details |
| RX 7900 XT | 20 GB | Fits | F16 | 8.9 GB | 104 | details |
| RX 9070 | 16 GB | Fits | F16 | 8.9 GB | 83 | details |
| RX 9070 XT | 16 GB | Fits | F16 | 8.9 GB | 83 | details |
| Ryzen AI Max+ 395 128GB | 128 GB | Fits | F16 | 8.9 GB | 33 | details |
| Arc A770 16GB | 16 GB | Fits | F16 | 8.9 GB | 73 | details |
| Mac M1 16GB | 16 GB | Fits | F16 | 8.9 GB | 8.9 | details |
| Mac M2 24GB | 24 GB | Fits | F16 | 8.9 GB | 13 | details |
| Mac M3 24GB | 24 GB | Fits | F16 | 8.9 GB | 13 | details |
| Mac M4 16GB | 16 GB | Fits | F16 | 8.9 GB | 16 | details |
| Mac M4 32GB | 32 GB | Fits | F16 | 8.9 GB | 16 | details |
| Mac M1 Pro 32GB | 32 GB | Fits | F16 | 8.9 GB | 26 | details |
| Mac M3 Pro 36GB | 36 GB | Fits | F16 | 8.9 GB | 20 | details |
| Mac M4 Pro 24GB | 24 GB | Fits | F16 | 8.9 GB | 36 | details |
| Mac M4 Pro 64GB | 64 GB | Fits | F16 | 8.9 GB | 36 | details |
| Mac M1 Max 64GB | 64 GB | Fits | F16 | 8.9 GB | 52 | details |
| Mac M2 Max 96GB | 96 GB | Fits | F16 | 8.9 GB | 52 | details |
| Mac M3 Max 128GB | 128 GB | Fits | F16 | 8.9 GB | 52 | details |
| Mac M4 Max 64GB | 64 GB | Fits | F16 | 8.9 GB | 71 | details |
| Mac M2 Ultra 192GB | 192 GB | Fits | F16 | 8.9 GB | 104 | details |
| Mac M3 Ultra 256GB | 256 GB | Fits | F16 | 8.9 GB | 107 | details |
Questions
How much VRAM does Phi 4 Mini Instruct need?
About 4.1 GB at 4-bit (Q4_K_M) with an 8K context: 2.3 GB of weights (real GGUF file size), 1.0 GB of KV cache and 0.8 GB of runtime overhead. At 8-bit it needs about 5.6 GB.
What is the smallest GPU that runs Phi 4 Mini Instruct?
At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 4060 (8 GB).
How big is the Ollama download for Phi 4 Mini Instruct?
2.5GB for the phi4-mini:3.8b tag (Q4_K_M), as listed on ollama.com. Run it with "ollama run phi4-mini:3.8b". With an 8K context, plan on about 4.1 GB of GPU memory.
Can I run Phi 4 Mini Instruct on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M1 16GB or larger, using the default macOS GPU memory limit.
Similar models
Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.
Phi 3 Mini 4K Instruct
3.8B · 5.7 GB at 4-bitPhi 3.5 Mini Instruct
3.8B · 5.7 GB at 4-bitPhi 3 Mini 128K Instruct
3.8B · 5.7 GB at 4-bitQwen2.5 VL 3B Instruct
3.8B · 2.8 GB at 4-bitGranite Vision 4.1 4B
4.0B · 3.2 GB at 4-bitGranite 4.2 3B
3.7B · 3.3 GB at 4-bitEmbed this
Free badge and "Will it run on my GPU?" widget for model cards, READMEs and blogs. No cookies, no tracking. All options.
Copy a snippet
Markdown badge
[](https://runsonmy.com/model/phi-4-mini-instruct/) Widget (script)
<div class="runsonmy-widget" data-model="phi-4-mini-instruct"><a href="https://runsonmy.com/model/phi-4-mini-instruct/">Will Phi 4 Mini Instruct run on my GPU? Check on RunsOnMy</a></div>
<script async src="https://runsonmy.com/embed.js"></script> Widget (iframe)
<iframe src="https://runsonmy.com/embed/m/phi-4-mini-instruct/" width="100%" height="220" style="max-width:440px;border:0" loading="lazy" title="Will Phi 4 Mini Instruct run on my GPU? (RunsOnMy)"></iframe> Model facts
| Hugging Face repo | microsoft/Phi-4-mini-instruct |
| Parameters | 3,836,021,760 (3.8B) |
| Layers / hidden size | 32 / 3072 |
| Attention | GQA; 24 heads, 8 KV heads, head dim 128 |
| KV cache per token (FP16) | 128.0 KiB |
| Max context | 131,072 tokens |
| License | mit |
| Released on HF | Feb 19, 2025 |
| Downloads (30d) / likes | 301,567 / 891 |
Sources: Hugging Face model API · config.json (microsoft/Phi-4-mini-instruct) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.