cohere asr ja needs about 2.7 GB of memory at Q6_K (a 2.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 5 quants, from 1.5 GB (Q4_K) to 4.1 GB (F16).
Japanese ASR model optimized for general and anime domains
cohere asr ja is a 2.07B parameter model fine‑tuned for Japanese speech recognition. It supports a 8k token context and is released under Apache‑2.0. The model is intended for use with CrispASR’s cohere backend.
Good at
- Japanese speech recognition
- Anime domain audio
- General dialogue
Limits
- No vision support
- Requires CrispASR cohere backend
- No VAD integration
Best for
- Transcribing Japanese podcasts
- Anime subtitle generation
- Voice‑to‑text for Japanese apps
Who it’s for. Developers building Japanese speech‑to‑text solutions who need a lightweight, open‑source model. Ideal for projects that can run on a GPU with 8k context support.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K | 1.5 GB | 2.2 GB | Fits | Get file → | |
| Q5_0 | 1.7 GB | 2.4 GB | Fits | Get file → | |
| Q6_K | 2.0 GB | 2.7 GB | Fits | Get file → | |
| Q8_0 | 2.4 GB | 3.2 GB | Fits | Get file → | |
| F16 | 4.1 GB | 5.1 GB | Fits | Get file → |
For 8GB VRAM, use Q4_K (1.51 GB, ~2.2 GB memory). For 12GB VRAM, Q5_0 (1.74 GB, ~2.4 GB memory) or Q6_K (1.98 GB, ~2.7 GB memory). For 16GB VRAM, Q8_0 (2.42 GB, ~3.2 GB memory) or F16 (4.14 GB, ~5.1 GB memory).
What GPU do you need for cohere asr ja?
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 Q8_0 (2.4 GB file, about 3.2 GB in use).
- 6 GB VRAM: the best quant that fits is F16 (4.1 GB file, about 5.1 GB in use).
Which cohere asr ja quant should you choose?
Q6_K (2.0 GB) is the middle option in this repository. Very close to the original quality for most uses, at a clearly larger size than 4-bit.
If memory is tight, the smallest file is Q4_K at 1.5 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
Source
Frequently asked questions about cohere asr ja
How much VRAM does cohere asr ja need?
About 2.7 GB at Q6_K with an 8K context window. That is the 2.0 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 cohere asr ja on an 8 GB GPU?
Yes, with F16: the 4.1 GB file needs about 5.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of cohere asr ja GGUF?
It depends on the quant: Q4_K is 1.5 GB, Q5_0 is 1.7 GB, Q6_K is 2.0 GB, Q8_0 is 2.4 GB, F16 is 4.1 GB.
Where can I download cohere asr ja GGUF?
The files are on Hugging Face in the CKHO/cohere-asr-ja-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 cohere asr ja 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 cohere asr ja on 8 GB VRAM?
Yes, the Q4_K quantized file is 1.51 GB and requires about 2.2 GB VRAM at 8k context.
Does this model support English?
No, the model is fine‑tuned for Japanese and does not support other languages.
What license applies to this model?
The model is released under the Apache‑2.0 license.