text model · Qwen2.5 · Android
Can I run Qwen2.5 72B on Generic Android Phone (12GB RAM)?
No. Qwen2.5 72B needs ~50.2 GB even at Q4_K_M, but Generic Android Phone (12GB RAM) only has ~8.5 GB usable.
Needs ~50.2 GB even at Q4_K_M, but only ~8.5 GB is usable.
That figure is at a 4k context and moves about ±15% as context length changes.
The gap is about 41.7 GB: Qwen2.5 72B needs roughly 50.2 GB at Q4_K_M and Generic Android Phone (12GB RAM) leaves only about 8.5 GB usable for a model. The lightest tracked hardware that runs Qwen2.5 72B is the Apple M4 Max (128GB) at 128 GB. See Qwen2.5 72B on Apple M4 Max (128GB).
- Q4_K_M needed
- ~50.2 GB
- Usable on device
- ~8.5 GB
- Device memory
- 12 GB
Which quant fits
How to run it
On Android use PocketPal AI (Polished app, download GGUF and run offline.).
- Parameters
- 72B
- Q4_K_M size
- 47.42 GB
- Q8_0 size
- 77.26 GB
- Context
- 128k
- Ollama tag
- qwen2.5:72b
- Memory
- 12 GB ram
- Usable for weights
- ~8.5 GB
- Best runtime
- llama.cpp (PocketPal) or MLC-LLM
What you can run instead
Run Qwen2.5 72B on other hardware
FAQ
Can Generic Android Phone (12GB RAM) run Qwen2.5 72B?
No. Qwen2.5 72B needs ~50.2 GB even at Q4_K_M, but Generic Android Phone (12GB RAM) only has ~8.5 GB usable.
How much memory does Qwen2.5 72B need?
Generic Android Phone (12GB RAM) does not have enough memory. At Q4_K_M the weights are ~47.42 GB; with KV cache and runtime overhead, budget ~50.2 GB at a 4k context.
What is the best tool to run Qwen2.5 72B on Android?
On Android, PocketPal AI (Polished app, download GGUF and run offline.) is the go-to option. NPU acceleration is limited and chip-specific; most apps run on CPU. Expect 1B-4B class.
Embed this
A live badge for your model card or README, updated as the data is. [](https://localmodel.run/can-i-run/qwen2.5-72b/android-generic-12gb) Sources
- github.com/a-ghorbani
- github.com/google-ai-edge
- github.com/mlc-ai
- github.com/shubham0204
- github.com/Vali-98
- huggingface.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF
- huggingface.co/bartowski/Qwen2.5-72B-Instruct-GGUF
- layla-network.ai
- lmarena.ai
- ollama.com/library/llama3.1:8b
- ollama.com/library/qwen2.5
- qwenlm.github.io
Weights are measured from GGUF files; KV cache and overhead are computed, so totals can vary ~15% with context and runtime. Any tok/s is a bandwidth estimate. See methodology.