# Can I run Qwen2.5 7B on Apple M4 Pro (24GB)?

> **Yes, it runs.** Runs at Q4_K_M using ~6.1 GB of ~16 GB usable. You have room for FP16 for higher quality.

Last validated: 2026-06-15.

## Memory math
- Qwen2.5 7B: 7B; Q4_K_M weighs 4.68 GB on disk.
- At Q4_K_M with a 4k context it needs ~6.1 GB (weights + KV cache + ~0.8 GB runtime overhead, ±15% with context length).
- Apple M4 Pro (24GB): 24 GB unified, ~16 GB usable for model weights.
- Headroom: ~9.9 GB; it also fits FP16 for higher quality.

## How to run
Recommended quant: Q4_K_M. Best tool on macOS: LM Studio.

```sh
ollama run qwen2.5:7b
```

## Run Qwen2.5 7B on other devices
- [Nvidia GeForce RTX 3060 (12GB)](https://localmodel.run/can-i-run/qwen2.5-7b/nvidia-rtx-3060-12gb)
- [Nvidia GeForce RTX 4070 (12GB)](https://localmodel.run/can-i-run/qwen2.5-7b/nvidia-rtx-4070-12gb)
- [Samsung Galaxy S24 Ultra](https://localmodel.run/can-i-run/qwen2.5-7b/samsung-s24-ultra)
- [Generic Android Phone (12GB RAM)](https://localmodel.run/can-i-run/qwen2.5-7b/android-generic-12gb)
- [iPhone 17 Pro](https://localmodel.run/can-i-run/qwen2.5-7b/iphone-17-pro)
- [iPhone Air](https://localmodel.run/can-i-run/qwen2.5-7b/iphone-air)

## Other models that run on Apple M4 Pro (24GB)
- [gpt-oss 20B (21B)](https://localmodel.run/can-i-run/gpt-oss-20b/apple-m4-pro-24gb)
- [DeepSeek-V2-Lite (16B)](https://localmodel.run/can-i-run/deepseek-v2-lite/apple-m4-pro-24gb)
- [Phi-4 14B (14B)](https://localmodel.run/can-i-run/phi-4-14b/apple-m4-pro-24gb)
- [Qwen2.5 14B (14B)](https://localmodel.run/can-i-run/qwen2.5-14b/apple-m4-pro-24gb)
- [Qwen3 14B (14B)](https://localmodel.run/can-i-run/qwen3-14b/apple-m4-pro-24gb)
- [DeepSeek-R1-Distill-Qwen 14B (14B)](https://localmodel.run/can-i-run/deepseek-r1-distill-qwen-14b/apple-m4-pro-24gb)

Estimate, not a guarantee. Sources: https://ollama.com/library/qwen2.5, https://huggingface.co/bartowski/Qwen2.5-7B-Instruct-GGUF, https://qwenlm.github.io/blog/qwen2.5/.
More: https://localmodel.run/can-i-run/qwen2.5-7b/apple-m4-pro-24gb