TO

tolkio models

TolkioGGUF by datacrunchUpdated 2026-10-27
Download on Hugging Face
Parameters1.94B
Context256K
Layers24
Architectureqwen35
Smallest quant0.6 GB
Quick answer

tolkio models needs about 2.1 GB of memory at its main file (a 1.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.6 GB to 2.1 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.

Tolkio models – 1.94B parameter Qwen‑35 architecture for offline speech translation

Tolkio models are 1.94 billion parameter Qwen‑35 models designed for offline speech translation on Android. They support reasoning and tool usage. The models are available as GGUF files with varying sizes and memory footprints.

Good at

  • Reasoning
  • Tool usage
  • Offline speech translation

Limits

  • No vision support
  • Large context requires high VRAM
  • License is other

Best for

  • Offline Android speech translation
  • Local reasoning tasks
  • Tool‑enabled interactions

Who it’s for. Developers building offline translation apps or local LLM tools that need reasoning and tool support without internet access. The models are suitable for mobile environments where bandwidth is limited.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.6 GB 1.5 GB
Fits Get file →
—1.2 GB 2.1 GB
Fits Get file →
—1.5 GB 2.4 GB
Fits Get file →
—2.1 GB 3.0 GB
Fits Get file →

For 8 GB VRAM, the 0.64 GB file uses about 1.5 GB memory at 8 k context, which is safe. If more memory is available, the 1.24 GB file uses about 2.1 GB, and the 1.49 GB file uses about 2.4 GB. The 2.08 GB file uses about 3 GB and is best for higher‑capacity systems.

What GPU do you need for tolkio models?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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 (2.1 GB file, about 3.0 GB in use).

Which tolkio models quant should you choose?

If memory is tight, the smallest file is the first one in the table at 0.6 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters1.94B
Layers24
Trained context262,144 tokens
KV cache (fp16)48 KB per token

Source

Quantized bydatacrunch
FamilyTolkio
Licenseother
Downloads (HF)4,313
Last updated2026-10-27

Frequently asked questions about tolkio models

How much VRAM does tolkio models need?

About 2.1 GB at its main file with an 8K context window. That is the 1.2 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 tolkio models on an 8 GB GPU?

Yes, with the smaller files: the 2.1 GB file needs about 3.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of tolkio models GGUF?

It depends on the quant: file is 0.6 GB, file is 1.2 GB, file is 1.5 GB, file is 2.1 GB.

What context length does tolkio models support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.

Where can I download tolkio models GGUF?

The files are on Hugging Face in the datacrunch/tolkio-models repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does tolkio models use?

Hugging Face lists the license as other. 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 Tolkio models on 8 GB VRAM?

Yes, the 0.64 GB file uses about 1.5 GB memory at 8 k context, well within 8 GB VRAM.

Do these models support vision tasks?

No, the models do not support vision.

What license do the models have?

The models have an "other" license; check the repository for details.

Similar models

Browse all →

Other tiny models

All tiny models →