README.md

---
license: mit
language:
- en
base_model: juanquivilla/sotto-cleanup-lfm25-350m
tags:
- speech-to-text
- transcript-cleanup
- text-correction
- asr-post-processing
- LFM
- LiquidAI
- mlx
- mlx-5bit
pipeline_tag: text-generation
---

# SottoASR Transcript Cleanup — LFM2.5-350M MLX 5-bit (soup_30)

[sottoasr.app](https://sottoasr.app) · [Full precision (bf16)](https://huggingface.co/juanquivilla/sotto-cleanup-lfm25-350m) · [MLX 4-bit (smaller)](https://huggingface.co/juanquivilla/sotto-cleanup-lfm25-350m-mlx-4bit)

## Overview

MLX 5-bit affine quantization of [juanquivilla/sotto-cleanup-lfm25-350m](https://huggingface.co/juanquivilla/sotto-cleanup-lfm25-350m). Recommended for Apple Silicon — best size/quality trade-off.

## What's new in soup_30

soup_30 extends v45 with targeted training data for five failure modes (multi-number sentences,
year-context drift, disconnected number lists, within-input duplicates, long-form preservation),
each generated programmatically and audited with a Qwen3.6-27B judge.

| Metric | v45 | **soup_30** |
|---|---:|---:|
| Number accuracy | 95.9% | **96.5%** |
| Adversarial benchmark (greedy) | 76% | **86%** |

See the [bf16 model card](https://huggingface.co/juanquivilla/sotto-cleanup-lfm25-350m) for the full pipeline and benchmark numbers.

## Quantization Recipe

```bash
mlx_lm.convert \
  --hf-path juanquivilla/sotto-cleanup-lfm25-350m \
  --mlx-path sotto-cleanup-lfm25-350m-mlx-5bit \
  -q --q-bits 5 --q-group-size 64 \
  --trust-remote-code
```

## Usage

```python
from mlx_lm import load, generate
from mlx_lm.sample_utils import make_sampler

model, tokenizer = load("juanquivilla/sotto-cleanup-lfm25-350m-mlx-5bit")
sampler = make_sampler(temp=0.0)

text = "talk about server three sixty"
prompt = f"### Input:\n{text}\n\n### Output:\n"
output = generate(model, tokenizer, prompt=prompt, max_tokens=512, sampler=sampler)
if "###" in output:
    output = output[:output.index("###")].strip()
print(output)
```

## License

MIT