TR

Translategemma 4b It

TranslategemmaGGUF by SandLogicTechnologiesUpdated 2026-01-19
Download on Hugging Face
Parameters3.88B
Context128K
Layers34
Architecturegemma3
Smallest quant2.5 GB
Quick answer

Translategemma 4b It needs about 4.1 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 2 quants, from 2.5 GB (Q4_K_M) to 2.8 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 4.6 GB.

TranslateGemma-4B-IT – instruction‑tuned multilingual translation model

TranslateGemma-4B-IT is a 4 billion‑parameter decoder‑only transformer fine‑tuned for translation tasks. It supports multiple high‑resource and selected low‑resource languages and follows explicit translation instructions. The model is available in GGUF format with Q4_K_M and Q5_K_M quantizations.

Good at

  • High‑quality multilingual translation
  • Instruction‑driven control over tone and formality
  • Long‑form translation with contextual awareness

Limits

  • Text‑only modality, no vision support
  • Requires 8 k context for memory estimates
  • License is gemma, no commercial redistribution

Best for

  • Batch translation pipelines
  • Interactive translation assistants
  • Research prototyping of multilingual models

Who it’s for. Developers and researchers needing a compact, instruction‑tuned translation model for production or experimentation. The model is suitable for edge deployments and low‑VRAM GPU inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M2.5 GB 4.1 GB
Fits Get file →
Q5_K_M2.8 GB 4.5 GB
Fits Get file →

For 8 GB VRAM, Q4_K_M (2.49 GB file, ~4.1 GB memory at 8 k context) is recommended. If 12 GB VRAM is available, Q5_K_M (2.83 GB file, ~4.5 GB memory) offers slightly better quality. Memory estimates are approximate.

What GPU do you need for Translategemma 4b It?

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 Q5_K_M (2.8 GB file, about 4.5 GB in use).

Which Translategemma 4b It quant should you choose?

Q4_K_M is the usual starting point for Translategemma 4b It: it balances file size and quality well. Here it is a 2.5 GB file.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegemma3
Parameters3.88B
Layers34
Trained context131,072 tokens
KV cache (fp16)136 KB per token

Source

Licensegemma
Downloads (HF)9,932
Last updated2026-01-19

Frequently asked questions about Translategemma 4b It

How much VRAM does Translategemma 4b It need?

About 4.1 GB at Q4_K_M with an 8K context window. That is the 2.5 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 Translategemma 4b It on an 8 GB GPU?

Yes, with Q5_K_M: the 2.8 GB file needs about 4.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Translategemma 4b It GGUF?

It depends on the quant: Q4_K_M is 2.5 GB, Q5_K_M is 2.8 GB.

What context length does Translategemma 4b It support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.6 GB at 32K.

Where can I download Translategemma 4b It GGUF?

The files are on Hugging Face in the SandLogicTechnologies/translategemma-4b-it-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 Translategemma 4b It use?

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

Does the model support image input?

No, the model is text‑only and does not process visual modalities.

What license applies to this model?

The model is released under the gemma license, which restricts commercial redistribution.

Other GGUFs of translategemma-4b-it

Similar models

Browse all →

More from SandLogicTechnologies

All by SandLogicTechnologies →

More Translategemma models

All Translategemma →

Other small models

All small models →