CA

canary ctc aligner

CanaryGGUF by cstrUpdated 2026-08-02
Download on Hugging Face
Parameters626M
Context—
Layers—
Architecturecanary-ctc
Smallest quant0.4 GB
Quick answer

canary ctc aligner needs about 1.0 GB of memory at Q5_0 (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.4 GB (Q4_K) to 1.3 GB.

Multilingual CTC aligner for 25 European languages

The canary ctc aligner is a 0.626‑billion‑parameter model derived from nvidia/canary-1b-v2. It provides word‑level and segment‑level timestamps for transcripts in 25 European languages. The model is released under a CC‑BY‑4.0 license.

Good at

  • Accurate forced alignment
  • Multilingual support for 25 European languages
  • Compatible with any ASR transcript

Limits

  • No vision support
  • Requires 8k context for memory estimates
  • No explicit language list beyond 25 European languages

Best for

  • Generating timestamps for ASR transcripts
  • Aligning transcripts from other ASR systems
  • Post‑processing speech transcription outputs

Who it’s for. Developers needing precise word‑level timestamps for multilingual speech data. Researchers and engineers working with Canary or other ASR models who require forced alignment without retraining.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K0.4 GB 0.9 GB
Fits Get file →
Q5_00.5 GB 1.0 GB
Fits Get file →
Q8_00.7 GB 1.2 GB
Fits Get file →
—1.3 GB 1.9 GB
Fits Get file →

For 8 GB VRAM, use the Q4_K quant (0.39 GB file, ~0.9 GB memory). For 12 GB VRAM, Q5_0 (0.46 GB file, ~1 GB memory) is suitable. If more memory is available, Q8_0 (0.67 GB file, ~1.2 GB memory) offers higher precision.

What GPU do you need for canary ctc aligner?

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 the main file (1.3 GB file, about 1.9 GB in use).

Which canary ctc aligner quant should you choose?

Q5_0 (0.5 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_K at 0.4 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

Architecturecanary-ctc
Parameters626M

Source

Quantized bycstr
FamilyCanary
Licensecc-by-4.0
Downloads (HF)4,339
Last updated2026-08-02

Frequently asked questions about canary ctc aligner

How much VRAM does canary ctc aligner need?

About 1.0 GB at Q5_0 with an 8K context window. That is the 0.5 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 canary ctc aligner on an 8 GB GPU?

Yes, with the smaller files: the 1.3 GB file needs about 1.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of canary ctc aligner GGUF?

It depends on the quant: Q4_K is 0.4 GB, Q5_0 is 0.5 GB, Q8_0 is 0.7 GB, file is 1.3 GB.

Where can I download canary ctc aligner GGUF?

The files are on Hugging Face in the cstr/canary-ctc-aligner-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 canary ctc aligner use?

Hugging Face lists the license as cc-by-4.0. 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 the canary ctc aligner on 8 GB VRAM?

Yes, the Q4_K quant uses about 0.9 GB of VRAM at 8k context, fitting within an 8 GB budget.

Does the model support languages outside the 25 European ones?

No, the model is trained for 25 European languages only.

What license applies to this model?

The model is released under a CC‑BY‑4.0 license.

Other GGUFs of canary-1b-v2

Similar models

Browse all →

More from cstr

All by cstr →

More Canary models

All Canary →

Other tiny models

All tiny models →