Can GRM 3.2 Cliff 9B run on Pixel 10 Pro?
YES — Runs, barely
Formula estimate
What this means
Our estimate says GRM 3.2 Cliff 9B should fit, but replies are likely to arrive slowly.
Download the Q4_K_M version, which is about 5.9 GB. We expect it to use about 7.5 GB of the roughly 12 GB available to a model on this phone.
At the estimated speed, a roughly 300-word answer may take about 77 seconds to finish.
This result is calculated from the phone and model specifications. It has not been measured on this exact phone and setup.
5.2tokens/s
estimated · reading pace
7.5GB
needed at Q4_K_M
Q4_K_M
quant checked
needs 7.5 GBusable 12 GB
Google Tensor G516 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 4.5 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 5.2 tokens/s — reading pace. A ~300-word reply takes about 77 seconds on this phone.
Where the memory goes
| Component | Detail | GB |
|---|---|---|
| Model weights | Q4_K_M GGUF (5.9 GB) + mmap overhead | 6.2 |
| KV cache | 4K context window | 0.7 |
| Runtime | llama.cpp + app overhead | 0.6 |
| Total needed | at Q4_K_M, 4K context | 7.5 |
| Working budget | 16 GB RAM − Android system reserve | 12 |
| Headroom | remaining inside the working budget | 4.5 |
Pick your quant
| Quant | Download | Verdict | Speed |
|---|---|---|---|
| Q2_K | 4.1 GB | ! Runs, barely | ~7.5 tokens/s |
| Q3_K_M | 4.9 GB | ! Runs, barely | ~6.3 tokens/s |
| IQ4_XS | 5.2 GB | ! Runs, barely | ~5.9 tokens/s |
| Q4_0 | 5.5 GB | ! Runs, barely | ~5.6 tokens/s |
| Q4_K_S | 5.6 GB | ! Runs, barely | ~5.5 tokens/s |
| Q4_K_M BEST HERE | 5.9 GB | ! Runs, barely | ~5.2 tokens/s |
| Q5_K_M | 6.9 GB | ! Runs, barely | ~4.4 tokens/s |
| Q6_K | 7.7 GB | ! Runs, barely | ~4 tokens/s |
| Q8_0 | 9.5 GB | ! Runs, barely | ~3.2 tokens/s |
Get your first offline chat working
Recommended app: PocketPal. Follow the point-and-click steps below. The speed above is an estimate, not a measurement from this exact app and phone.
1
Install or update PocketPal from Google Play. It is free and does not require an account. Use a current version so its loader supports newer model architectures.
2
Open the exact model. In PocketPal, go to Models → + → Add from Hugging Face, then paste
bartowski/OrionLLM_GRM-3.2-Cliff-GGUF.3
Choose the Q4_K_M GGUF file. The download is about 5.9 GB, so use Wi-Fi and keep the app open. Choose the main GGUF weights, not a vision projector, mmproj, or other helper file.
4
Tap Download, then Load. Start with a 4K (4096-token) context. Keep the app’s default Android backend for the first run.
5
Send a simple first prompt. Try “Explain why the sky is blue in three sentences.” This page estimates about 5.2 tokens/s, but that number is not a PocketPal measurement unless it carries a ✓ Verified label.
6
Confirm it is really offline. After the first reply, turn on airplane mode and ask a second question. If it still answers, the model is running on your phone.
If it does not work
- Model not listed: update PocketPal and paste the exact repository
bartowski/OrionLLM_GRM-3.2-Cliff-GGUF. - App closes while loading: close other apps, restart the phone, and try 2K context. If it still closes, choose a smaller model.
- No offline reply: confirm that the Q4_K_M GGUF file is loaded in the chat rather than a remote model.
Related checks
More on Pixel 10 Pro
GRM 3.2 Cliff 9B on other phones