text model · Granite · Android
Can I run Granite 4.0 H Small on Samsung Galaxy S25 Ultra (16GB, 1TB config only)?
No. Granite 4.0 H Small needs ~20.3 GB even at Q4_K_M, but Samsung Galaxy S25 Ultra (16GB, 1TB config only) only has ~12 GB usable.
Needs ~20.3 GB even at Q4_K_M, but only ~12 GB is usable.
That figure is at a 4k context and moves about ±15% as context length changes.
The gap is about 8.3 GB: Granite 4.0 H Small needs roughly 20.3 GB at Q4_K_M and Samsung Galaxy S25 Ultra (16GB, 1TB config only) leaves only about 12 GB usable for a model. The lightest tracked hardware that runs Granite 4.0 H Small is the Nvidia GeForce RTX 4090 (24GB) at 24 GB. See Granite 4.0 H Small on Nvidia GeForce RTX 4090 (24GB).
- Q4_K_M needed
- ~20.3 GB
- Usable on device
- ~12 GB
- Device memory
- 16 GB
Which quant fits
How to run it
On Android use PocketPal AI (Polished app, download GGUF and run offline.).
- Parameters
- 32B (MoE, 9B active)
- Q4_K_M size
- 18.14 GB
- Q8_0 size
- 31.91 GB
- Context
- 128k
- Ollama tag
- granite4:32b-a9b-h
- Memory
- 16 GB ram
- Usable for weights
- ~12 GB
- Power draw
- ~8 W
- Best runtime
- llama.cpp (PocketPal) or MLC-LLM (Adreno GPU path)
What you can run instead
Run Granite 4.0 H Small on other hardware
FAQ
Can Samsung Galaxy S25 Ultra (16GB, 1TB config only) run Granite 4.0 H Small?
No. Granite 4.0 H Small needs ~20.3 GB even at Q4_K_M, but Samsung Galaxy S25 Ultra (16GB, 1TB config only) only has ~12 GB usable.
How much memory does Granite 4.0 H Small need?
Samsung Galaxy S25 Ultra (16GB, 1TB config only) does not have enough memory. At Q4_K_M the weights are ~18.14 GB; with KV cache and runtime overhead, budget ~20.3 GB at a 4k context. It is a Mixture-of-Experts model (32B total / 9B active), so all experts must stay in memory; memory tracks total params, not active params.
What is the best tool to run Granite 4.0 H Small 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/granite-4-h-small/samsung-s25-ultra-16gb) Sources
- gadgetversus.com
- github.com/a-ghorbani
- github.com/google-ai-edge
- github.com/mlc-ai
- github.com/shubham0204
- github.com/Vali-98
- gsmarena.com/samsung_galaxy_s25_ultra-12750.php
- gsmarena.com/samsung_galaxy_s25_ultra-13322.php
- huggingface.co/ibm-granite/granite-4.0-h-small
- huggingface.co/ibm-granite/granite-4.0-h-small-GGUF
- layla-network.ai
- ollama.com
- sammyfans.com
- sammyguru.com
- samsung.com
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.