Can the Apple M4 (16 GB) run MiniCPM-OCR-Jev?
MiniCPM-OCR-Jev·2.2B·Apple M4 (16 GB)·16 GB unified memory
Yes. The Apple M4 (16 GB) runs MiniCPM-OCR-Jev fully in unified memory at FP16 / BF16 (about 4.7 GB of 11 GB).
- best quant
- FP16 / BF16
- needs
- 4.7 GB
- usable
- 11 GB
Estimates from the model's size.Try other contexts on the MiniCPM-OCR-Jev page →
Every quantisation of MiniCPM-OCR-Jev on the Apple M4 (16 GB)
| Quant | Memory | Fit |
|---|---|---|
| Q4_K_M | 1.9 GB | fits |
| Q8_0 | 2.7 GB | fits |
| FP16 / BF16 pick | 4.7 GB | fits |
Speed is a bandwidth-based estimate for generating text (120 GB/s); prompt processing and long contexts are slower.
Run MiniCPM-OCR-Jev on the Apple M4 (16 GB)
Using FP16 / BF16, the best fit for Apple M4 (16 GB).
ollama run hf.co/openbmb/MiniCPM-OCR-Jev-gguf:F16 llama-server -hf openbmb/MiniCPM-OCR-Jev-gguf:F16brew install llama.cpp ships with Metal. Downloads the GGUF on first run.
lms get openbmb/MiniCPM-OCR-Jev-gguf@F16Then load it from the LM Studio app or with `lms load`.
OCR models that fit the Apple M4 (16 GB) comfortably
All that fit →| Model | Params | Best quant | Memory |
|---|---|---|---|
| 3.3BA1.1B | FP16 / BF16 | ≈8.0 GBFits in memory | |
| 3.3BA1.1B | FP16 / BF16 | ≈8.0 GBFits in memory | |
| 3.3BA1.1B | FP16 / BF16 | ≈8.0 GBFits in memory | |
| 1.3B | FP16 / BF16 | ≈3.5 GBFits in memory | |
| 3.4BA1.2B | FP16 / BF16 | ≈8.1 GBFits in memory | |
| 3B | FP16 / BF16 | ≈7.3 GBFits in memory |
Other GPUs that run MiniCPM-OCR-Jev
Frequently asked questions
- Can the Apple M4 (16 GB) run MiniCPM-OCR-Jev?
- Yes. The Apple M4 (16 GB) runs MiniCPM-OCR-Jev fully in unified memory at FP16 / BF16 (about 4.7 GB of 11 GB).
- Which quantisation of MiniCPM-OCR-Jev should I use on the Apple M4 (16 GB)?
- FP16 / BF16: the largest one that fits. Expected quality: lossless compared with the full-precision weights.
Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.