I don't know why the model large-v3 in https://huggingface.co/ggerganov/whisper.cpp does have errors or not work for me, that's why I have converted pt --> fp32 --> quants manually
| Filename | Size | Quantization |
|---|---|---|
| ggml-large-v3-fp32.bin | 3.1 GB | FP32 |
| ggml-large-v3-q8_0.bin | 1.66 GB | Q8_0 |
| ggml-large-v3-q6_k.bin | 1.28 GB | Q6_K |
| ggml-large-v3-q5_k.bin | 1.08 GB | Q5_K |
| ggml-large-v3-q4_k.bin | 889 MB | Q4_K |
| ggml-large-v3-q3_k.bin | 686 MB | Q3_K |
| ggml-large-v3-q2_k.bin | 530 MB | Q2_K |
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
Model tree for adriabama06/whisper-large-v3-ggml
Base model
openai/whisper-large-v3