Downloads · 30 days
0
aoiandroid/distil-whisper-large-v3-litert-fp8
distil-whisper-large-v3-litert-fp8 is a automatic speech recognition model from aoiandroid. Use it when you need speech turned into text. It is set up for litert. The card lists the license as mit.
📱 FP8 (Float8 E4M3) Dynamic Range Quantized LiteRT Model Package (.litertlm) for distil-whisper/distil-large-v3
Downloads · 30 days
0
Access
Public
Updated Jul 26, 2026
Repo size
487 MB
Likes
0
Public
Click a slice to open those files.
.litertlm487 MB · 100%
From the Hugging Face model README
📱 FP8 (Float8 E4M3) Dynamic Range Quantized LiteRT Model Package (.litertlm) for distil-whisper/distil-large-v3
This repository contains the Float8 (FP8 E4M3) dynamically quantized LiteRT model package.
| Metric | Measured Value | Details & Notes |
|---|---|---|
| Package Size | 487 MB | 86.3% Size Reduction vs FP32 |
| Peak RAM Footprint | ~190 MB | Low Memory Footprint |
| Word Error Rate (WER) | 1.4% (Zero Degradation) | 100% Equivalent to Original PyTorch FP32 |
distil_whisper_large_v3_fp8_mobile.litertlm (487 MB)