GA

GalTransl v4 4B 2601

GaltranslGGUF by SakuraLLMUpdated 2026-02-04
Download on Hugging Face
Parameters4B
Context32K
Layers36
Architectureqwen3
Smallest quant2.8 GB
Quick answer

GalTransl v4 4B 2601 needs about 4.5 GB of memory at Q5_K_S (a 2.8 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 2.8 GB (Q5_K_S) to 3.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.

GalTransl v4 4B 2601 – a 4‑billion‑parameter Qwen3 model for instant translation

GalTransl v4 4B 2601 is a 4‑billion‑parameter Qwen3 architecture model. It is designed for real‑time translation scenarios such as Luna Translator. The model supports a context length of 32768 tokens and is released under a CC‑BY‑NC‑SA‑4.0 license.

Good at

  • Real‑time translation
  • High context length support
  • Stable output format

Limits

  • No vision support
  • Requires at least 2 GB VRAM for Q5_K_S quant
  • CC‑BY‑NC‑SA‑4.0 license limits commercial use

Best for

  • Instant translation in chat applications
  • Batch translation of long documents
  • Language‑pair testing

Who it’s for. Developers building translation tools or chatbots that need quick, accurate translation. Suitable for hobbyists and research projects with non‑commercial constraints.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q5_K_S2.8 GB 4.5 GB
Fits Get file →
—3.3 GB 5.0 GB
Fits Get file →

For 4 GB VRAM, use the Q5_K_S quantized file (2.82 GB, ~4.5 GB memory at 8 k context). If 6 GB VRAM is available, the Q6K quantized file (size not listed but recommended) can be used. Memory estimates are approximate.

What GPU do you need for GalTransl v4 4B 2601?

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: no quant fits at an 8K context.
  • 6 GB VRAM: the best quant that fits is the main file (3.3 GB file, about 5.0 GB in use).

Which GalTransl v4 4B 2601 quant should you choose?

Q5_K_S (2.8 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.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen3
Parameters4B
Layers36
Trained context32,768 tokens
KV cache (fp16)144 KB per token

Source

Quantized bySakuraLLM
FamilyGaltransl
Downloads (HF)12,871
Last updated2026-02-04

Frequently asked questions about GalTransl v4 4B 2601

How much VRAM does GalTransl v4 4B 2601 need?

About 4.5 GB at Q5_K_S with an 8K context window. That is the 2.8 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 GalTransl v4 4B 2601 on an 8 GB GPU?

Yes, with the smaller files: the 3.3 GB file needs about 5.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of GalTransl v4 4B 2601 GGUF?

It depends on the quant: Q5_K_S is 2.8 GB, file is 3.3 GB.

What context length does GalTransl v4 4B 2601 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.

Where can I download GalTransl v4 4B 2601 GGUF?

The files are on Hugging Face in the SakuraLLM/GalTransl-v4-4B-2601 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does GalTransl v4 4B 2601 use?

Hugging Face lists the license as cc-by-nc-sa-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 GalTransl v4 4B 2601 on 4 GB VRAM?

Yes, using the Q5_K_S quantized file requires about 4.5 GB of VRAM at 8 k context.

Does the model support image input?

No, GalTransl v4 4B 2601 does not support vision input.

What license does the model use?

The model is released under a CC‑BY‑NC‑SA‑4.0 license, restricting commercial use.

Similar models

Browse all →

More from SakuraLLM

All by SakuraLLM →

Other small models

All small models →