Downloads · 30 days
6
10% of all-time downloads
whisper-large-v3-turbo
basecompute/whisper-large-v3-turbo
whisper-large-v3-turbo is a automatic speech recognition model from basecompute. Use it when you need speech turned into text. It is set up for basert. The card lists the license as apache-2.0.
BaseRT .base build of openai/whisper-large-v3-turbo, OpenAI's 809M-parameter Whisper speech-recognition model (turbo: 4-layer decoder distillation of large-v3, 128 mel bins), for fast local transcription on Apple Sili…
All-time downloads
58
Public
Repo size
6.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.base3.1 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- Library
- basert
- License
- apache-2.0
- Access
- Public
- Created
- Aug 5, 2026
- Updated
- Aug 6, 2026
- SHA
- 6572da45
Base models
Model card
From the Hugging Face model README
whisper-large-v3-turbo
BaseRT .base build of
openai/whisper-large-v3-turbo,
OpenAI's 809M-parameter Whisper speech-recognition model (turbo: 4-layer decoder distillation of large-v3, 128 mel bins),
for fast local transcription on Apple Silicon.
Files
| File | Precision | Size |
|---|---|---|
whisper-large-v3-turbo-F16.base | float16 | 1.62 GB |
whisper-large-v3-turbo-Q8.base | 8-bit linears, f16 embeddings/conv/norms | 913 MB |
whisper-large-v3-turbo-Q4.base | 4-bit linears, f16 embeddings/conv/norms | 569 MB |
F16 and Q8 are transcription-quality equivalent (Q8 passes the same word-error parity gates against reference openai-whisper). Q4 is the smallest and remains accurate; on some smaller variants it can occasionally repeat a word in timestamped beam decoding.
Usage
curl -LsSf https://basecompute.co/install.sh | sh
basert serve --model whisper-large-v3-turbo-F16.base
POST /v1/audio/transcriptions (multipart or JSON) returns json, text,
srt, vtt, or verbose_json (with per-segment avg_logprob /
no_speech_prob / compression_ratio / temperature and the detected
language), with optional SSE streaming and POST /v1/audio/translations. Supported
request fields: language (or "auto" to detect), prompt
(initial prompt / vocabulary bias), task (transcribe/translate). Or transcribe directly
from the CLI:
basert-transcribe whisper-large-v3-turbo-F16.base audio.wav --lang auto
Multilingual: pass language (e.g. --lang de), or auto to detect, across the 100 supported languages. Note: turbo was distilled without translation training, so task=translate is rejected.
Released under the apache-2.0 license, inherited from the base model.
- Task
- Automatic Speech Recognition
- Library
- basert
- License
- apache-2.0
- Created
- Aug 5, 2026
- Updated
- Aug 6, 2026