ternary-models: VLMs, Multimodal & Audio
Collection
Ternary-quantized models for architectures GGUF can't handle. tritplane3 scheme. โข 16 items โข Updated โข 2
How to use AsadIsmail/whisper-small-ternary with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("automatic-speech-recognition", model="AsadIsmail/whisper-small-ternary") # Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("AsadIsmail/whisper-small-ternary", dtype="auto")Ternary-quantized version of openai/whisper-small.
| Property | Value |
|---|---|
| Parameters | 244M |
| Quantization | tritplane3 (120 decoder layers) |
| Stored size | 127 MB |
| Compression | 1.26ร |
Part of ternary-models.
Base model
openai/whisper-small