| --- |
| license: mit |
| base_model: openai/whisper-small |
| base_model_relation: quantized |
| language: |
| - en |
| - multilingual |
| pipeline_tag: automatic-speech-recognition |
| tags: |
| - automatic-speech-recognition |
| - whisper |
| - whisper.cpp |
| - ggml |
| - quantized |
| - on-device |
| --- |
| |
| # Whisper ggml models (availability mirror) |
|
|
| **This repository redistributes existing model weights unchanged.** It exists so that a |
| published mobile application has a distribution endpoint whose availability we control. |
| No model was trained, fine-tuned, or converted here. |
|
|
| ## Provenance |
|
|
| | | | |
| |---|---| |
| | Original models | [OpenAI Whisper](https://github.com/openai/whisper) — **Copyright (c) 2022 OpenAI**, MIT License | |
| | ggml conversion & quantization | [ggerganov/whisper.cpp](https://huggingface.co/ggerganov/whisper.cpp) — **Copyright (c) 2023-2026 The ggml authors**, MIT License | |
| | This repository | Byte-identical copy of three files from the above. **No files were altered.** | |
|
|
| ## Contents |
|
|
| | File | Source model | Quantization | Size | |
| |---|---|---|---| |
| | `ggml-base.en-q5_1.bin` | Whisper base.en | Q5_1 | ~57 MB | |
| | `ggml-small.en-q5_1.bin` | Whisper small.en | Q5_1 | ~190 MB | |
| | `ggml-small-q5_1.bin` | Whisper small (multilingual) | Q5_1 | ~190 MB | |
| |
| Each file has been verified by SHA-256 against the upstream repository at the time of upload. |
| |
| ## License |
| |
| **MIT**, inherited from both upstream projects. The full text is included in this |
| repository: [LICENSE](./LICENSE). |
| |
| - OpenAI Whisper — Copyright (c) 2022 OpenAI |
| - whisper.cpp / ggml — Copyright (c) 2023-2026 The ggml authors |
| |
| The MIT License permits redistribution provided the copyright notice and permission |
| notice are retained, which this repository does. |
| |
