---
license: mit
language:
- en
- multilingual
base_model:
- openai/whisper-medium
library_name: transformers
pipeline_tag: automatic-speech-recognition
tags:
- whisper
- speech-to-text
- mlx
- quantized
- q8
- apple-silicon
inference: false
---

# whisper-medium-mlx-q8

`whisper-medium-mlx-q8` is an MLX-ready Whisper speech-to-text checkpoint derived from `openai/whisper-medium` for local transcription on Apple Silicon.

## Intended use

- Local speech-to-text transcription on Apple Silicon
- Batch or interactive audio transcription experiments
- Multilingual ASR workflows when supported by the upstream Whisper checkpoint

## Out of scope

- Safety-critical decisions without domain expert review
- Claims of benchmark superiority not backed by published evaluation data
- Non-MLX runtime guarantees; this card documents the shipped HF checkpoint, not every possible serving stack
- Speaker diarization, clinical interpretation, or audio enhancement

## Training and conversion metadata

| Parameter | Value |
|---|---|
| Repository | `LibraxisAI/whisper-medium-mlx-q8` |
| Base model | `openai/whisper-medium` |
| Task | `automatic-speech-recognition` |
| Library | `transformers` |
| Format | MLX / Apple Silicon checkpoint |
| Quantization | Q8 |
| Architecture | Not declared in config |
| Model files | 1 |
| Config model_type | `whisper` |

This card only reports metadata present in the Hugging Face repository, existing card frontmatter, or public config files. Missing benchmark, dataset, or training-run details are left explicit rather than reconstructed.

## Tested inference path

> **Inference for this checkpoint has been tested with [`LibraxisAI/mlx-batch-server`](https://github.com/LibraxisAI/mlx-batch-server).**\
> This is the recommended tested path for operator-controlled local inference on Apple Silicon.

| Aspect | Status |
|---|---|
| Tested runtime | `LibraxisAI/mlx-batch-server` |
| Target hardware | Apple Silicon |
| Inference mode | Local / self-hosted |
| Hugging Face Hosted Inference | Disabled for this repository (`inference: false`) |

This does not claim compatibility with every possible serving stack. It documents the path that has been exercised for this published checkpoint.

## Usage

### Python

```python
import mlx_whisper

result = mlx_whisper.transcribe(
    "audio.wav",
    path_or_hf_repo="LibraxisAI/whisper-medium-mlx-q8",
)
print(result["text"])
```

### Notes

- Use local audio files supported by `mlx_whisper`.
- For long recordings, split audio into manageable chunks before transcription.

## Example output

No public sample output is currently declared for this checkpoint.

## Quantization notes

| Aspect | Original/base checkpoint | This checkpoint |
|---|---|---|
| Lineage | `openai/whisper-medium` | `LibraxisAI/whisper-medium-mlx-q8` |
| Runtime target | Upstream runtime format | MLX on Apple Silicon |
| Quantization | Base precision or upstream-declared format | Q8 |
| Published quality delta | Not declared in public metadata | Not declared in public metadata |

## Limitations

- No public benchmarks for this checkpoint are declared in the model metadata.
- No public benchmark claims are made by this card unless listed in the frontmatter.
- Validate outputs on your own domain data before relying on this checkpoint.
- Memory use and speed depend heavily on Apple Silicon generation, unified-memory size, audio duration, and language complexity.

## License

`mit`. Check the upstream/base model license as well when a base model is declared.

## Citation

```bibtex
@misc{libraxisai-whisper-medium-mlx-q8,
  title = {whisper-medium-mlx-q8},
  author = {LibraxisAI},
  year = {2026},
  howpublished = {\url{https://huggingface.co/LibraxisAI/whisper-medium-mlx-q8}},
  note = {MLX checkpoint published by LibraxisAI}
}
```
---

𝚅𝚒𝚋𝚎𝚌𝚛𝚊𝚏𝚝𝚎𝚍. with AI Agents by VetCoders (c)2024-2026 LibraxisAI