Can Ternary Bonsai 27B run on Redmi Note 14 Pro+ 5G?
What this means
Our estimate says Ternary Bonsai 27B should fit, but replies are likely to arrive slowly.
Download the Q2_0 version, which is about 7.2 GB. We expect it to use about 8.4 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 4 minutes to finish.
This result is calculated from the phone and model specifications. It has not been measured on this exact phone and setup.
Make it bearable
See how fast it feels
Estimated at 1.6 tokens/s — slower than you read. A ~300-word reply takes about 250 seconds on this phone.
Where the memory goes
| Component | Detail | GB |
|---|---|---|
| Model weights | Q2_0 GGUF (7.2 GB) + mmap overhead | 7.6 |
| KV cache | 4K context window | 0.3 |
| Runtime | llama.cpp + app overhead | 0.6 |
| Total needed | at Q2_0, 4K context | 8.4 |
| Working budget | 12 GB RAM − Android system reserve | 9 |
| Headroom | remaining inside the working budget | 0.6 |
Pick your quant
| Quant | Download | Verdict | Speed |
|---|---|---|---|
| Q2_0 BEST HERE | 7.2 GB | ! Runs, barely | ~1.6 tokens/s |
| Q2_g64 | 7.6 GB | ! Runs, barely | ~1.5 tokens/s |
Use the supported phone path
The current App Store and Play Store versions of PocketPal cannot load this tracked ternary file. Memory fit does not make this exact download beginner-ready.
For the same kind of chat in PocketPal, use this supported build instead:
Check Bonsai 27B (1-bit)