GE

gemma 3 4B reasoning multimodal en ft v2

GemmaGGUF by CALISTA-INDUSTRYUpdated 2025-06-26
Download on Hugging Face
Parameters3.88B
Context128K
Layers34
Architecturegemma3
Smallest quant7.8 GB
Quick answer

gemma 3 4B reasoning multimodal en ft v2 needs about 9.4 GB of memory at its main file (a 7.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 4.6 GB.

Gemma 3 4B reasoning multimodal fine‑tuned for English VQA and image captioning

Gemma 3 4B reasoning multimodal en ft v2 is a fine‑tuned Gemma3 model for multimodal reasoning. It accepts text and image inputs and outputs text. The model is trained on a 128K context window and is tailored for English.

Good at

  • Multimodal reasoning
  • Visual question answering
  • Image captioning

Limits

  • Performance may degrade on non‑English inputs
  • May not generalize to very different domains
  • No vision support beyond image input

Best for

  • VQA systems
  • Multimodal dialogue with visual context
  • Instruction following with images

Who it’s for. Developers building lightweight multimodal AI applications that require reasoning over text and images in English. The model is suitable for laptops, desktops, or modest cloud deployments.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—7.8 GB 9.4 GB
Too big Get file →

The only available GGUF quant is unknown. The file is 7.77 GB and requires about 9.4 GB of VRAM at 8 k context. For 8 GB VRAM, it is borderline; consider using a higher‑capacity GPU or reducing context size.

What GPU do you need for gemma 3 4B reasoning multimodal en ft v2?

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: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: the best quant that fits is the main file (7.8 GB file, about 9.4 GB in use).

Specifications

Architecture from the GGUF header

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

Source

Quantized byCALISTA-INDUSTRY
FamilyGemma
Licensegemma
Downloads (HF)873
Last updated2025-06-26

Frequently asked questions about gemma 3 4B reasoning multimodal en ft v2

How much VRAM does gemma 3 4B reasoning multimodal en ft v2 need?

About 9.4 GB at its main file with an 8K context window. That is the 7.8 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 gemma 3 4B reasoning multimodal en ft v2 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 9.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of gemma 3 4B reasoning multimodal en ft v2 GGUF?

It depends on the quant: file is 7.8 GB.

What context length does gemma 3 4B reasoning multimodal en ft v2 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 gemma 3 4B reasoning multimodal en ft v2 GGUF?

The files are on Hugging Face in the CALISTA-INDUSTRY/gemma_3_4B_reasoning_multimodal_en_ft_v2 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does gemma 3 4B reasoning multimodal en ft v2 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.

Can I run this model on 8 GB VRAM?

The model requires about 9.4 GB of VRAM at 8 k context, so 8 GB is insufficient unless context is reduced.

Does it support languages other than English?

Performance may degrade on non‑English inputs; the model is fine‑tuned for English.

Is the model open source?

Yes, it is released under the Apache 2.0 license and available on the Hugging Face hub.

Other GGUFs of gemma-3-1b-it-unsloth-bnb-4bit

Similar models

Browse all →

More from CALISTA-INDUSTRY

All by CALISTA-INDUSTRY →

More Gemma models

All Gemma →

Other small models

All small models →