---
title: CrisperWhisper2.0-GGML
canonical_url: "https://www.modelscope.ai/models/drbaph-org/CrisperWhisper2.0-GGML"
md_url: "https://www.modelscope.ai/models/drbaph-org/CrisperWhisper2.0-GGML.md"
repository: drbaph-org/CrisperWhisper2.0-GGML
last_updated: 2026-09-21
license: other
base_model:
  - nyralabs/CrisperWhisper2.0_small
  - nyralabs/CrisperWhisper2.0_medium
  - nyralabs/CrisperWhisper2.0_turbo
  - nyralabs/CrisperWhisper2.0_large
base_model_relation: finetune
library_name:
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - de
downloads: 50
stars: 0
tags:
  - speech-recognition
  - verbatim
  - disfluency
  - whisper
  - word-timestamps
  - cpp
  - ggml
  - cuda
  - cpu
  - "arxiv:2607.18934"
---

# CrisperWhisper2.0-GGML

> CrisperWhisper2.0-GGML - An open-source model by drbaph-org on ModelScope. CrisperWhisper 2.0 GGML

drbaph-org/CrisperWhisper2.0-GGML is a machine learning model on ModelScope. licensed under other. derived from nyralabs/CrisperWhisper2.0_small、nyralabs/CrisperWhisper2.0_medium、nyralabs/CrisperWhisper2.0_turbo.

- **Repository**: drbaph-org/CrisperWhisper2.0-GGML
- **License**: other
- **Base model**: nyralabs/CrisperWhisper2.0_small, nyralabs/CrisperWhisper2.0_medium, nyralabs/CrisperWhisper2.0_turbo, nyralabs/CrisperWhisper2.0_large
- **Tags**: speech-recognition, verbatim, disfluency, whisper, word-timestamps, cpp, ggml, cuda, cpu, arxiv:2607.18934
- **Downloads**: 50
- **Stars**: 0
- **Last updated**: 2026-09-21

Source: https://www.modelscope.ai/models/drbaph-org/CrisperWhisper2.0-GGML

---

# CrisperWhisper 2.0 GGML

Unofficial FP16 GGML conversions of the public CrisperWhisper 2.0 checkpoints
for [CrisperWhisper.cpp](https://github.com/Saganaki22/CrisperWhisper.cpp), a
native Windows/Linux C++ runtime with CPU, NVIDIA CUDA, supervised word
timestamps, and long-form stitching. Python is not required during inference.

> This is an unofficial community conversion. It is not affiliated with,
> endorsed by, or maintained by Nyra Health or Nyra Labs.

## Files and exact upstream revisions

| Model | GGML file | Size | Upstream checkpoint | Revision |
| --- | --- | ---: | --- | --- |
| Small | `ggml-crisperwhisper-small-f16.bin` | 488 MB | [nyralabs/CrisperWhisper2.0_small](https://huggingface.co/nyralabs/CrisperWhisper2.0_small) | `57750c47fde52dc1b016ec2bd4bf4704944cf3df` |
| Medium | `ggml-crisperwhisper-medium-f16.bin` | 1.53 GB | [nyralabs/CrisperWhisper2.0_medium](https://huggingface.co/nyralabs/CrisperWhisper2.0_medium) | `cc01b03e6d9486e6d15688bac3077380c0775c61` |
| Turbo | `ggml-crisperwhisper-turbo-f16.bin` | 1.62 GB | [nyralabs/CrisperWhisper2.0_turbo](https://huggingface.co/nyralabs/CrisperWhisper2.0_turbo) | `831f87e1d69c3799fa05fd5e21c01fbb0b6957c0` |
| Large | `ggml-crisperwhisper-large-f16.bin` | 3.09 GB | [nyralabs/CrisperWhisper2.0_large](https://huggingface.co/nyralabs/CrisperWhisper2.0_large) | `f4334f6e8193f2691212d49b20fa12d370e13896` |

These are FP16 conversions, not integer-quantized models. Each `.bin` has a
matching `.bin.metadata.json` sidecar containing its immutable source revision,
revision-specific supervised alignment heads, and generation suppression list.
Keep both files together; CrisperWhisper.cpp v1.2.0 discovers the sidecar
automatically.

The gated
[CrisperWhisper2.0_large_pro](https://huggingface.co/nyralabs/CrisperWhisper2.0_large_pro)
checkpoint is not included. It requires approval and its applicable commercial
model license.

## Download

```bash
hf download drbaph/CrisperWhisper2.0-GGML \
  ggml-crisperwhisper-turbo-f16.bin \
  ggml-crisperwhisper-turbo-f16.bin.metadata.json \
  --local-dir models
```

Replace `turbo` with `small`, `medium`, or `large`.

## Native runtime

Download source and prebuilt Windows/Linux CPU, CUDA, and Linux ARM64 packages
from:

- [CrisperWhisper.cpp](https://github.com/Saganaki22/CrisperWhisper.cpp)
- [CrisperWhisper.cpp releases](https://github.com/Saganaki22/CrisperWhisper.cpp/releases)

The CUDA v1.2.0 packages contain native code for every CUDA 13 target from
SM 86 through SM 121 and retain a CPU fallback. Models are not duplicated in
the GitHub release archives.

## Usage

Windows:

```powershell
.\crisper-whisper.exe `
  --model .\models\ggml-crisperwhisper-turbo-f16.bin `
  --file .\audio.wav `
  --mode verbatim `
  --language en
```

Linux:

```bash
./crisper-whisper \
  --model ./models/ggml-crisperwhisper-turbo-f16.bin \
  --file ./audio.wav \
  --mode verbatim \
  --language en
```

Word timestamps and boundary-safe repeated-speech stitching:

```bash
./crisper-whisper \
  --model ./models/ggml-crisperwhisper-turbo-f16.bin \
  --file ./long-audio.wav \
  --longform-strategy chunked-lcs \
  --word-timestamps \
  --json
```

The CLI accepts WAV, MP3, FLAC, and Ogg Vorbis and automatically decodes,
downmixes, and resamples input to mono 16 kHz in memory. No FFmpeg subprocess
or preprocessing script is required.

Normal transcription keeps GGML Flash Attention enabled. The optional
supervised timestamp path uses an additional teacher-forced no-Flash alignment
pass because fused Flash Attention does not expose the required decoder/audio
cross-attention matrix.

## Validation

Every refreshed model passed native CPU transcription and supervised
word-timestamp validation on the bundled JFK sample. Turbo and Large also
passed native CUDA inference on an RTX 5090. The `chunked-lcs` 55-second,
five-copy JFK regression returned 110 words across two chunks with zero invalid
timestamps and zero adjacent overlaps.

## Original project and paper

- [Official CrisperWhisper repository](https://github.com/nyrahealth/CrisperWhisper)
- [CrisperWhisper paper](https://arxiv.org/abs/2607.18934)
- [Native C++ implementation](https://github.com/Saganaki22/CrisperWhisper.cpp)
- [whisper.cpp](https://github.com/ggml-org/whisper.cpp)

## Limitations

- These are conversions of upstream checkpoints, not separately trained models.
- Output can differ slightly from Python/Transformers because the numerical
  backend and decoding implementation differ.
- FP16 files use more storage and memory than integer-quantized models.
- Speculative draft-model decoding, forced alignment, and the complete upstream
  hallucination-recovery stack are not yet implemented in the C++ runtime.
- Large Pro is not included.

## License

These conversions do not change or replace the original model license.

Small, Medium, Turbo, and Large weights, configuration, tokenizers, parameters,
and model outputs remain governed by the
[Nyra Health Non-Commercial Research License](https://huggingface.co/nyralabs/CrisperWhisper2.0_large/blob/main/LICENSE.md).
Review that license before downloading, using, modifying, or redistributing
these converted files. The original models, architecture, research, and related
intellectual property belong to their respective rights holders.