Downloads · 30 days
18
7% of all-time downloads
VoiceScribe/qwen3-5-0.8b-dictation-corrector-mlx-mxfp4-g32
qwen3-5-0.8b-dictation-corrector-mlx-mxfp4-g32 is a text generation model from VoiceScribe. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as apache-2.0.
RAM-constrained V15 R-3 corrector for low-memory Apple Silicon (M1 8 GB and M2 16 GB-1 hosts). PTQ on V15's bf16-fused weights via mlx-lm --q-mode mxfp4 --q-group-size 32. No adapter retraining; sidesteps the noisy-Q4…
Downloads · 30 days
18
7% of all-time downloads
All-time downloads
257
Public
Parameters
752M
420 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors401 MB · 95%
How the weights are stored.
U32752M · 100%
From the Hugging Face model README
RAM-constrained V15 R-3 corrector for low-memory Apple Silicon (M1 8 GB and M2 16 GB-1 hosts). PTQ on V15's bf16-fused weights via mlx-lm --q-mode mxfp4 --q-group-size 32. No adapter retraining; sidesteps the noisy-Q4-gradient failure mode that doomed the train-on-Q4 V14/V16 lineage.
| Eval set | V15 prod (Q8 g=64) | This mirror | V14 4-bit baseline |
|---|---|---|---|
| seed_v5 wild (50 rows, hard-neg discourse) | 100% | 100% | 74.1% |
| ECHO15 long-form WER vs Gemini-3.1 ref | 0.210 | 0.164 | not tested |
| Dev-corpus long-form WER vs Gemini-3.5-Flash ref | 0.141 | 0.140 | not tested |
| Disk size | 782 MB | 400 MB | 424 MB |
| Warm RSS (process) | 1573 MB | 1244 MB | ~1100 MB |
| Bits/weight (avg) | ~8.5 | {4.258} | ~4.5 |
This mirror is strictly Pareto-optimal vs V15 Q8 on every axis:
smaller, lower RSS, equal-or-better quality. See companion mirror
qwen3-5-0.8b-dictation-corrector-mlx-mixed46-g32
for the M2 16 GB+ (16 GiB+ unified) tier.
Best for: M1 8 GB (8-15 GiB unified).
The voice-scribe-macos installer (WP#1074) ships both this mirror and
its sibling and picks one at install time via sysctl hw.memsize:
mxfp4 g=32 (this mirror, 400 MB / 1244 MB RSS)mixed46 g=32 (482 MB / 1326 MB RSS)from mlx_lm import load, generate
from mlx_lm.sample_utils import make_sampler
model, tokenizer = load("VoiceScribe/qwen3-5-0.8b-dictation-corrector-mlx-mxfp4-g32")
messages = [
{"role": "system", "content": "Корректор русской диктовки. Убери слова-паразиты. Нормализуй IT-термины. Не меняй смысл."},
{"role": "user", "content": "Эм, докер мониторит, ну, бэкенд через гитхаб экшнс"},
]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True, enable_thinking=False
)
sampler = make_sampler(temp=0.0, top_p=1.0)
out = generate(model, tokenizer, prompt=prompt, max_tokens=200, sampler=sampler)
print(out) # → "Docker мониторит бэкенд через GitHub Actions"
Source: V15 R-3 LoRA-fused bf16 weights (the predecessor of the V15 Q8 production mirror). PTQ via mlx-lm 0.31.x:
python -m mlx_lm convert \
--hf-path /path/to/v15-r3-fused-bf16 \
--mlx-path ./qwen3-5-0.8b-dictation-corrector-mlx-mxfp4-g32 \
{ --quantize --q-mode mxfp4 --q-group-size 32 \
} --dtype bfloat16
The bf16-fused weights are the canonical V15 source; the V15 Q8 production mirror was produced from the same source via uniform Q8 quantisation. This mirror simply uses a smarter quantiser on the same proven training signal.
Predecessor approaches (V14, V16) attempted to train a LoRA adapter directly on a Q4-quantised base. Both produced regressions because Q4 forward gradients are too noisy — the adapter learns a sloppier signal than the bf16-base V15 path.
This mirror reverses the order: train on bf16 (proven V15 path), then quantise smarter. Result: the adapter signal is preserved faithfully; the only loss is from re-quantising weights with a more aggressive bit budget on layers where it doesn't hurt.
The {mxfp4 (E2M1 floating-point) numerics} keeps the high-magnitude channels (attention
output projections — where the LoRA-trained signal concentrates) at
high precision while compressing the MLP feedforward layers (which
tolerate aggressive quant). Result: 38-49% smaller disk + 16-21% lower
RSS at zero (or better) quality vs the uniform Q8 mirror.
докер→Docker,
гитхаб→GitHub), preserves meaning verbatim.@software{{voicescribe-v15-mxfp4-2026,
title = {{Voice Scribe Russian Dictation Corrector V15 R-3 (mxfp4 {Apple Silicon M1 8 GB tier})}},
author = {{Sabynin, Andrey}},
year = {{2026}},
note = {{WP#1067 R&D + WP#1074 productisation}},
url = {{https://huggingface.co/VoiceScribe/qwen3-5-0.8b-dictation-corrector-mlx-mxfp4-g32}}
}}
qwen3-5-0.8b-dictation-corrector-mlx-mixed46-g32 — sibling tierdocs/wp1067-q4-rd/CYCLES_RESULTS.md (PR #148)