text model · Qwen2.5-VL · Android
Can I run Qwen2.5-VL 3B on Samsung Galaxy S26 Ultra (16GB, 1TB config)?
Yes. Qwen2.5-VL 3B runs on Samsung Galaxy S26 Ultra (16GB, 1TB config) at Q4_K_M (~4.4 GB of ~12 GB usable).
Runs at Q4_K_M using ~4.4 GB of ~12 GB usable. You have room for FP16 for higher quality.
That figure is at a 4k context and moves about ±15% as context length changes. Samsung Galaxy S26 Ultra (16GB, 1TB config) leaves ~7.6 GB of headroom, room to step up to FP16 for higher quality.
- Q4_K_M needed
- ~4.4 GB
- Usable on device
- ~12 GB
- Device memory
- 16 GB
- Best quant
- Q4_K_M
Which quant fits
How to run it
On Android use PocketPal AI (Polished app, download GGUF and run offline.).
- Parameters
- 3.75B
- Q4_K_M size
- 3.05 GB
- Q8_0 size
- 5.1 GB
- Context
- 32k
- Memory
- 16 GB ram
- Usable for weights
- ~12 GB
- Best runtime
- llama.cpp (PocketPal) or MLC-LLM (Adreno GPU path)
You could also run
Run Qwen2.5-VL 3B on other hardware
FAQ
Can Samsung Galaxy S26 Ultra (16GB, 1TB config) run Qwen2.5-VL 3B?
Yes. Qwen2.5-VL 3B runs on Samsung Galaxy S26 Ultra (16GB, 1TB config) at Q4_K_M (~4.4 GB of ~12 GB usable).
How much memory does Qwen2.5-VL 3B need?
Samsung Galaxy S26 Ultra (16GB, 1TB config) has room to spare. At Q4_K_M the weights are ~3.05 GB; with KV cache and runtime overhead, budget ~4.4 GB at a 4k context.
What is the best tool to run Qwen2.5-VL 3B 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-vl-3b/samsung-s26-ultra) Sources
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.