MO

moss transcribe.cpp

MossGGUF by mudlerUpdated 2026-07-09
Download on Hugging Face
Parameters909M
Context—
Layers—
Architecturemoss_transcribe_diarize
Smallest quant0.5 GB
Quick answer

moss transcribe.cpp needs about 1.2 GB of memory at Q5_K (a 0.6 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 7 quants, from 0.5 GB (Q4_0) to 1.8 GB (F16).

MOSS-Transcribe-Diarize 0.9B: One‑pass audio transcription with speaker diarization

MOSS-Transcribe-Diarize 0.9B is an end‑to‑end audio understanding model that transcribes long‑form multi‑speaker recordings, provides timestamps, and assigns anonymous speaker labels. It supports over 50 languages and can process audio up to 90 minutes in a single pass.

Good at

  • Long‑form transcription
  • Speaker‑aware diarization
  • Acoustic event annotation

Limits

  • No vision support
  • Requires CPU or GPU ggml backend
  • License is Apache‑2.0

Best for

  • Meeting and call transcription
  • Podcast and interview captioning
  • Video subtitle generation

Who it’s for. Developers building transcription pipelines, content creators needing speaker‑aware captions, and researchers working with multi‑speaker audio. The model is suitable for CPU‑only inference via llama.cpp or ggml backends.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_00.5 GB 1.1 GB
Fits Get file →
Q4_K0.5 GB 1.1 GB
Fits Get file →
Q5_00.6 GB 1.2 GB
Fits Get file →
Q5_K0.6 GB 1.2 GB
Fits Get file →
Q6_K0.8 GB 1.4 GB
Fits Get file →
Q8_01.0 GB 1.6 GB
Fits Get file →
F161.8 GB 2.6 GB
Fits Get file →

For 8 GB VRAM, use Q4_0 or Q4_K (file size 0.54 GB, memory ~1.1 GB). For 12 GB VRAM, Q5_0 or Q5_K (0.65 GB, ~1.2 GB). For 16 GB VRAM, Q6_K (0.77 GB, ~1.4 GB). Larger budgets can use Q8_0 (0.99 GB, ~1.6 GB) or F16 (1.83 GB, ~2.6 GB).

What GPU do you need for moss transcribe.cpp?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is F16 (1.8 GB file, about 2.6 GB in use).

Which moss transcribe.cpp quant should you choose?

Q5_K (0.6 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.

If memory is tight, the smallest file is Q4_0 at 0.5 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturemoss_transcribe_diarize
Parameters909M

Source

Quantized bymudler
FamilyMoss
Licenseapache-2.0
Downloads (HF)1,725
Last updated2026-07-09

Frequently asked questions about moss transcribe.cpp

How much VRAM does moss transcribe.cpp need?

About 1.2 GB at Q5_K with an 8K context window. That is the 0.6 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run moss transcribe.cpp on an 8 GB GPU?

Yes, with F16: the 1.8 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of moss transcribe.cpp GGUF?

It depends on the quant: Q4_0 is 0.5 GB, Q4_K is 0.5 GB, Q5_0 is 0.6 GB, Q5_K is 0.6 GB, Q6_K is 0.8 GB, Q8_0 is 1.0 GB and 1 more.

Where can I download moss transcribe.cpp GGUF?

The files are on Hugging Face in the mudler/moss-transcribe.cpp-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does moss transcribe.cpp use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run moss transcribe.cpp on 8 GB VRAM?

Yes, the Q4_0 or Q4_K quantizations use about 1.1 GB of VRAM at 8k context, fitting within 8 GB.

Does the model support languages other than English?

It supports transcription and diarization across 50+ languages, including English, French, German, Italian, and more.

Is the model licensed for commercial use?

The model is released under the Apache‑2.0 license, which allows commercial use with attribution.

Other GGUFs of MOSS-Transcribe-Diarize

Similar models

Browse all →

More from mudler

All by mudler →

More Moss models

All Moss →

Other tiny models

All tiny models →