EM

emma 5

EmmaGGUF by MrMeOrYouUpdated 2026-10-30
Download on Hugging Face
Parameters614M
Context2K
Layers28
Architecturellama
Smallest quant0.1 GB
Quick answer

emma 5 needs about 1.2 GB of memory at Q4_K_M (a 0.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 25 quants, from 0.1 GB (Q1_0) to 2.5 GB (F32).

Emma‑5: 0.614 B‑parameter Italian‑focused LLM

Emma‑5 is a 0.614 B‑parameter decoder‑only LLM optimized for Italian. It offers moderate contextual understanding and medium‑complexity content generation. The model is not intended for safety‑critical or advanced multi‑step reasoning.

Good at

  • Italian language generation
  • Medium‑complexity content creation
  • Low‑latency inference on entry‑level GPUs

Limits

  • Context limited to 2 048 tokens
  • Not designed for advanced reasoning
  • No vision support

Best for

  • Chatting in Italian
  • Generating short articles
  • Lightweight local inference

Who it’s for. Developers needing a small Italian‑focused LLM for local inference. The model is suitable for hobby projects, lightweight chatbots, and educational tools where moderate quality is acceptable.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q1_00.1 GB 0.7 GB
Fits Get file →
TQ1_00.2 GB 0.8 GB
Fits Get file →
TQ2_00.3 GB 0.8 GB
Fits Get file →
Q2_K0.3 GB 0.8 GB
Fits Get file →
Q3_K_S0.3 GB 0.9 GB
Fits Get file →
IQ3_S0.3 GB 0.9 GB
Fits Get file →
IQ3_M0.3 GB 0.9 GB
Fits Get file →
Q3_K_M0.3 GB 0.9 GB
Fits Get file →
Q3_K_L0.3 GB 0.9 GB
Fits Get file →
IQ4_XS0.4 GB 0.9 GB
Fits Get file →
Q4_00.4 GB 0.9 GB
Fits Get file →
IQ4_NL0.4 GB 0.9 GB
Fits Get file →
Q4_K_S0.4 GB 1.0 GB
Fits Get file →
Q4_10.4 GB 1.0 GB
Fits Get file →
Q4_K_M0.4 GB 1.0 GB
Fits Get file →
Q5_00.4 GB 1.0 GB
Fits Get file →
Q5_K_S0.4 GB 1.0 GB
Fits Get file →
Q5_K_M0.5 GB 1.0 GB
Fits Get file →
Q5_10.5 GB 1.0 GB
Fits Get file →
Q6_K0.5 GB 1.1 GB
Fits Get file →
MXFP4_MOE0.7 GB 1.2 GB
Fits Get file →
Q8_00.7 GB 1.2 GB
Fits Get file →
BF161.2 GB 1.8 GB
Fits Get file →
F161.2 GB 1.8 GB
Fits Get file →
F322.5 GB 3.0 GB
Fits Get file →

For 8 GB VRAM, Q4_0 (0.36 GB, 1.2 GB memory) or Q4_K_M (0.40 GB, 1.2 GB memory) are good choices. For 4 GB VRAM, Q1_0 (0.13 GB, 0.9 GB memory) or TQ1_0 (0.24 GB, 1 GB memory) fit well. Higher precision quantisations require more VRAM.

What GPU do you need for emma 5?

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 F32 (2.5 GB file, about 3.3 GB in use).

Which emma 5 quant should you choose?

Q4_K_M is the usual starting point for emma 5: it balances file size and quality well. Here it is a 0.4 GB file.

The Q8_0 file is 0.7 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q1_0 at 0.1 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters614M
Layers28
Trained context2,048 tokens
KV cache (fp16)35 KB per token

Source

Quantized byMrMeOrYou
FamilyEmma
Base modelegomnia/emma-5
Licenseopenrail
Downloads (HF)1,610
Last updated2026-10-30

Frequently asked questions about emma 5

How much VRAM does emma 5 need?

About 1.2 GB at Q4_K_M with an 8K context window. That is the 0.4 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 emma 5 on an 8 GB GPU?

Yes, with F32: the 2.5 GB file needs about 3.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of emma 5 GGUF?

It depends on the quant: Q1_0 is 0.1 GB, TQ1_0 is 0.2 GB, TQ2_0 is 0.3 GB, Q2_K is 0.3 GB, Q3_K_S is 0.3 GB, IQ3_S is 0.3 GB and 19 more.

What context length does emma 5 support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.3 GB at 8K and 0.1 GB at 2K.

Where can I download emma 5 GGUF?

The files are on Hugging Face in the MrMeOrYou/emma-5-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 emma 5 use?

Hugging Face lists the license as openrail. 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 Emma‑5 on 8 GB VRAM?

Yes, Q4_0 or Q4_K_M quantisations fit within 8 GB VRAM with about 1.2 GB memory usage.

Does Emma‑5 support multi‑step reasoning?

No, the model is not designed for advanced multi‑step reasoning tasks.

Is Emma‑5 licensed for commercial use?

It is released under OpenRAIL‑M, which allows commercial use with restrictions on harmful or illegal applications.

Similar models

Browse all →

More from MrMeOrYou

All by MrMeOrYou →

More Emma models

All Emma →

Other tiny models

All tiny models →