Can OneJev 9B run on Pixel 8a?
Estimated weights fit · custom workflow required
Weight fit is not phone app support
The tracked Q2_K weights are 3.8GB; the generic working-memory estimate is 5.2GB against 6GB usable. Image inputs need additional projector memory.
OneJev scores typed decisions in a single forward pass, not chat replies. The decode estimates below are not measured decision latency. The publisher's qev server and System One client are required; AICanRun has not verified a phone app route.
Not measured
decision latency
5.2GB
needed at Q2_K
Q2_K
quant checked
needs 5.2 GBusable 6 GB
Google Tensor G38 GB RAMCPU / GPU
Where the memory goes
| Component | Detail | GB |
|---|---|---|
| Model weights | Q2_K GGUF (3.8 GB) + mmap overhead | 4 |
| KV cache | 4K context window | 0.6 |
| Runtime | llama.cpp + app overhead | 0.6 |
| Total needed | at Q2_K, 4K context | 5.2 |
| Working budget | 8 GB RAM − Android system reserve | 6 |
| Headroom | remaining inside the working budget | 0.8 |
Pick your quant
| Quant | Download | Verdict | Speed |
|---|---|---|---|
| Q2_K BEST HERE | 3.8 GB | ✓ Runs great | ~8.1 tokens/s |
| Q3_K_M | 4.6 GB | ✕ Won't fit | won't fit |
| IQ4_XS | 5.2 GB | ✕ Won't fit | won't fit |
| Q4_K_S | 5.4 GB | ✕ Won't fit | won't fit |
| Q4_K_M | 5.6 GB | ✕ Won't fit | won't fit |
| Q5_K_M | 6.5 GB | ✕ Won't fit | won't fit |
| Q6_K | 7.4 GB | ✕ Won't fit | won't fit |
| Q8_0 | 9.5 GB | ✕ Won't fit | won't fit |
Use the right workflow for this model
The model fits in memory, but it is not an ordinary chat model.
This model is designed to return calibrated probabilities for typed choices from text or images. PocketPal does not provide the required the publisher's qev server and System One client; image inputs also need a vision projector, so the tracked file is not a beginner phone-chat path.
Read the publisher workflow ↗If you only want a normal private chat, use this simpler model instead:
Check Llama 3.2 3BRelated checks
More on Pixel 8a
OneJev 9B on other phones