---
license: apache-2.0
base_model: openbmb/MiniCPM5-2B
pipeline_tag: text-generation
tags:
- executorch
library_name: executorch
---

# minicpm-5

This repository hosts the **MiniCPM5-2B** model exported for the
[React Native ExecuTorch](https://www.npmjs.com/package/react-native-executorch)
library as ExecuTorch `.pte` programs, ready to run on device.

Source: [openbmb/MiniCPM5-2B](https://huggingface.co/openbmb/MiniCPM5-2B),
revision `f97400052a43d642bbc6e9975e2397e3ae6a6b52` (final post-trained checkpoint).

## Variants

| Path | Backend | Precision | Context length |
| --- | --- | --- | --- |
| `mlx/minicpm_5_2b_mlx_4w.pte` | mlx | int4 | 8,192 |
| `xnnpack/minicpm_5_2b_xnnpack_8da4w.pte` | xnnpack | 8da4w | 8,192 |

## Repository structure

```
mlx/config.json                           1.3 KiB
mlx/minicpm_5_2b_mlx_4w.pte               1.3 GiB
tokenizer.json                            5.0 MiB
tokenizer_config.json                    97.5 KiB
xnnpack/config.json                       1.3 KiB
xnnpack/minicpm_5_2b_xnnpack_8da4w.pte    1.3 GiB
```

## Export notes

128-token prefill and float32 logits. The chat template supports `enable_thinking`: `true` opens reasoning and `false`
prefills an empty reasoning block. History omits reasoning.
MLX requires Apple silicon.

## Compatibility

These files are published for the **ExecuTorch v1.4.1** runtime. ExecuTorch
gives no forward compatibility guarantee, so an older runtime may fail to load
them.

To use them in React Native ExecuTorch, pass the model constant shipped in the
library's model registry to the corresponding task pipeline. See the
[documentation](https://docs.swmansion.com/react-native-executorch/docs/fundamentals/downloading-models).

To load these files in your own ExecuTorch runtime, read the
[compatibility note](https://github.com/pytorch/executorch/blob/main/runtime/COMPATIBILITY.md)
first.