EU

EuroLLM 22B Instruct 2512

EurollmGGUF by mradermacherUpdated 2025-12-15
Download on Hugging Face
Parameters22.6B
Context32K
Layers54
Architecturellama
Smallest quant0.0 GB
Quick answer

EuroLLM 22B Instruct 2512 needs about 16.0 GB of memory at Q4_K_M (a 13.7 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 24 quants, from 0.0 GB to 18.6 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 7.2 GB.

EuroLLM 22B Instruct 2512 – a 22B parameter multilingual LLM

EuroLLM 22B Instruct 2512 is a 22.6B parameter transformer model trained on 4 trillion tokens. It supports instruction following and machine translation across many European languages. The model is released under Apache‑2.0.

Good at

  • Instruction following
  • Multilingual translation
  • General NLP tasks

Limits

  • No vision support
  • Requires large VRAM for full precision
  • License is Apache‑2.0

Best for

  • Chatbots in EU languages
  • Translation services
  • Research on multilingual models

Who it’s for. Developers and researchers needing a large multilingual LLM for instruction‑based tasks. The model is suitable for applications that can handle the memory requirements of a 22B parameter model.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 2.3 GB
Fits Get file →
IQ1_S5.3 GB 7.6 GB
Tight Get file →
IQ1_M5.7 GB 8.0 GB
Tight Get file →
IQ2_XXS6.4 GB 8.7 GB
Too big Get file →
IQ2_XS7.0 GB 9.3 GB
Too big Get file →
IQ2_S7.4 GB 9.7 GB
Too big Get file →
IQ2_M8.0 GB 10.3 GB
Too big Get file →
Q2_K_S8.0 GB 10.4 GB
Too big Get file →
Q2_K8.7 GB 11.0 GB
Too big Get file →
IQ3_XXS9.0 GB 11.3 GB
Too big Get file →
IQ3_XS9.6 GB 11.9 GB
Too big Get file →
Q3_K_S10.0 GB 12.4 GB
Too big Get file →
IQ3_S10.1 GB 12.4 GB
Too big Get file →
IQ3_M10.4 GB 12.7 GB
Too big Get file →
Q3_K_M11.1 GB 13.4 GB
Too big Get file →
Q3_K_L12.1 GB 14.4 GB
Too big Get file →
IQ4_XS12.3 GB 14.6 GB
Too big Get file →
Q4_013.0 GB 15.3 GB
Too big Get file →
Q4_K_S13.0 GB 15.3 GB
Too big Get file →
Q4_K_M13.7 GB 16.0 GB
Too big Get file →
Q4_114.3 GB 16.6 GB
Too big Get file →
Q5_K_S15.7 GB 18.0 GB
Too big Get file →
Q5_K_M16.0 GB 18.4 GB
Too big Get file →
Q6_K18.6 GB 20.9 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (5.25 GB, 7.6 GB memory at 8k context) or IQ1_M (5.67 GB, 8 GB). For 12 GB VRAM, Q4_K_M (13.66 GB, 16 GB) or Q5_K_M (16.04 GB, 18.4 GB) are suitable. Larger VRAM allows higher‑quality quants such as Q6_K (18.57 GB, 20.9 GB).

What GPU do you need for EuroLLM 22B Instruct 2512?

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 (0.0 GB file, about 2.3 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.0 GB file, about 2.3 GB in use).
  • 8 GB VRAM: the best quant that fits is IQ1_M (5.7 GB file, about 8.0 GB in use).
  • 12 GB VRAM: the best quant that fits is IQ3_XS (9.6 GB file, about 11.9 GB in use).
  • 16 GB VRAM: the best quant that fits is Q4_K_M (13.7 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q6_K (18.6 GB file, about 20.9 GB in use).

Which EuroLLM 22B Instruct 2512 quant should you choose?

Q4_K_M is the usual starting point for EuroLLM 22B Instruct 2512: it balances file size and quality well. Here it is a 13.7 GB file.

If memory is tight, the smallest file is the first one in the table at 0.0 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

Architecturellama
Parameters22.6B
Layers54
Trained context32,768 tokens
KV cache (fp16)216 KB per token

Source

Quantized bymradermacher
FamilyEurollm
Licenseapache-2.0
Downloads (HF)801
Last updated2025-12-15

Frequently asked questions about EuroLLM 22B Instruct 2512

How much VRAM does EuroLLM 22B Instruct 2512 need?

About 16.0 GB at Q4_K_M with an 8K context window. That is the 13.7 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 EuroLLM 22B Instruct 2512 on an 8 GB GPU?

Yes, with IQ1_M: the 5.7 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of EuroLLM 22B Instruct 2512 GGUF?

It depends on the quant: file is 0.0 GB, IQ1_S is 5.3 GB, IQ1_M is 5.7 GB, IQ2_XXS is 6.4 GB, IQ2_XS is 7.0 GB, IQ2_S is 7.4 GB and 18 more.

What context length does EuroLLM 22B Instruct 2512 support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.8 GB at 8K and 7.2 GB at 32K.

Where can I download EuroLLM 22B Instruct 2512 GGUF?

The files are on Hugging Face in the mradermacher/EuroLLM-22B-Instruct-2512-i1-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 EuroLLM 22B Instruct 2512 use?

Hugging Face lists the license as apache-2.0. 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 EuroLLM 22B Instruct 2512 on 8 GB VRAM?

Yes, using IQ1_S or IQ1_M quants will fit within 8 GB VRAM, with estimated memory usage of 7.6 GB or 8 GB at 8k context.

Does the model support vision tasks?

No, EuroLLM 22B Instruct 2512 is a text‑only model with no vision capabilities.

What license does the model use?

The model is released under the Apache‑2.0 license.

Other GGUFs of EuroLLM-22B-Instruct-2512

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Eurollm models

All Eurollm →

Other medium models

All medium models →