Can Grug 12B run on Xiaomi 15T Pro?
YES — Runs, barely
Grade C · 54/100 · formula estimate
What this means
Our estimate says Grug 12B should fit, but there is very little memory left for the operating system and other apps.
Download the Q4_K_S version, which is about 7.2 GB. We expect it to use about 9 GB of the roughly 9 GB available to a model on this phone.
At the estimated speed, a roughly 300-word answer may take about 75 seconds to finish.
This result is calculated from the phone and model specifications. It has not been measured on this exact phone and setup.
5.3tokens/s
estimated · reading pace
9GB
needed at Q4_K_S
Q4_K_S
quant checked
C
grade · 54/100
needs 9 GBusable 9 GB
Dimensity 9400+12 GB RAMCPU / GPU
Make it bearable
1
Drop context to 2K — saves ~0.4 GB of KV cache and a little speed.
2
Close every other app before loading — the 0 GB headroom is real; Android may kill the app otherwise.
3
Expect throttling after ~10 min of sustained generation on a phone chassis.
See how fast it feels
Estimated at 5.3 tokens/s — reading pace. A ~300-word reply takes about 75 seconds on this phone.
Where the memory goes
| Component | Detail | GB |
|---|---|---|
| Model weights | Q4_K_S GGUF (7.2 GB) + mmap overhead | 7.6 |
| KV cache | 4K context window | 0.8 |
| Runtime | llama.cpp + app overhead | 0.6 |
| Total needed | at Q4_K_S, 4K context | 9 |
| Usable on Xiaomi 15T Pro | 12 GB RAM − Android system reserve | 9 |
| Headroom | left for the OS and your other apps | 0 |
Pick your quant
| Quant | Download | Verdict | Speed |
|---|---|---|---|
| Q2_K | 5.1 GB | ! Runs, barely | ~7.5 tokens/s |
| Q3_K_M | 6.3 GB | ! Runs, barely | ~6.1 tokens/s |
| IQ4_XS | 6.8 GB | ! Runs, barely | ~5.6 tokens/s |
| Q4_0 | 7.1 GB | ! Runs, barely | ~5.4 tokens/s |
| Q4_K_S BEST HERE | 7.2 GB | ! Runs, barely | ~5.3 tokens/s |
| Q4_K_M | 7.7 GB | ✕ Won't fit | won't fit |
| Q5_K_M | 8.8 GB | ✕ Won't fit | won't fit |
| Q6_K | 10.2 GB | ✕ Won't fit | won't fit |
| Q8_0 | 12.7 GB | ✕ Won't fit | won't fit |
Get it running in 3 steps
1
Install PocketPal from Google Play — free, no account needed.
2
Download "Grug 12B Q4_K_S" in the in-app model hub — 7.2 GB, use Wi-Fi. Search Hugging Face for
bartowski/kai-os_Grug-12B-GGUF and pick the Q4_K_S file.3
Set context to 4K — expect ~5.3 tokens/s on this phone.
Related checks
More on Xiaomi 15T Pro
Grug 12B on other phones