whatmodel.sh / EuroLLM 22B Instruct (2512)

EuroLLM 22B Instruct (2512)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

EuroLLM 22B Instruct (2512) is an open-weights language model from UTTER (EuroLLM), released December 2025. Its published size is 22.64B parameters and a 33k-token context window. Licence: apache-2.0.

Catalogue identity notes

utter-project/EuroLLM-22B-Instruct-2512 @8c6ed5c (created 2025-12-05, apache-2.0). EuroLLM 22B instruction model, December 2025 release (distinct from the earlier 22B-Instruct-Preview). Independent rows: EuroEval English (both datasets).

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 22,637,328,384 parameters.

52.5±6.0#387 of 949 LLMsfrom 2 benchmark results across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of EuroLLM 22B Instruct (2512), out of 1132 with a score, then more from UTTER (EuroLLM).

#modelscorewhereRAMreleased
501LFM2.5 2.6B Liquid AI53±1local~3.2GB2026-07
502Bielik 11B v2.1 Instruct SpeakLeash53±3local~8.5GB2024-08
503EuroLLM 22B Instruct (2512) UTTER (EuroLLM)53±6local~15GB2025-12
504Llama 3.1 70B Instruct Meta52±1api $0.72/M +local~27GB2024-07
505Reka Flash 3 (21B) Reka AI52±6local~14GB2025-03
more from UTTER (EuroLLM)
682EuroLLM 9B Instruct UTTER (EuroLLM)47±3local~6.9GB2024-11
1061EuroLLM 1.7B Instruct UTTER (EuroLLM)27±3local~2.4GB2024-08

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests2 not in a task · 84% of its score
testresultvs the others in its rundifficultycountswhy
EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy71.5% ✎▃·▁▁▂▁▂▅█▇▄▁ beats 55/100medium · avg 68%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 53%at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy87.7% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 108/146easy · avg 83%▇▇▇▇▇▇▇▇▇▇▇▇▇ 31%noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run EuroLLM 22B Instruct (2512) · needs ~13.7 GB · from huggingface.co/bartowski
    chat in the terminal
    llama cli -hf bartowski/utter-project_EuroLLM-22B-Instruct-2512-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf bartowski/utter-project_EuroLLM-22B-Instruct-2512-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run EuroLLM 22B Instruct (2512) · needs ~13.7 GB · from huggingface.co/bartowski

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/bartowski/utter-project_EuroLLM-22B-Instruct-2512-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run EuroLLM 22B Instruct (2512) · needs ~13.7 GB · from huggingface.co/bartowski
    chat in the terminal
    ollama run hf.co/bartowski/utter-project_EuroLLM-22B-Instruct-2512-GGUF:Q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run EuroLLM 22B Instruct (2512) · needs ~12.7 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/EuroLLM-22B-Instruct-2512-mlx-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/EuroLLM-22B-Instruct-2512-mlx-4bit

Other ways to run it

  1. Run it as published · needs ~45.27 GB, NVIDIA only · from huggingface.co/utter-project (official)
    vllm serve utter-project/EuroLLM-22B-Instruct-2512

Official weights and documentation: huggingface.co/utter-project ↗

$ how the score was worked out · EuroEval leaderboards (result… 84%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EuroEval leaderboards (results collection), English generative leaderboard2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 84%MIT
prior · pull toward the middle▇▇▇▇▇ 16%
weighted average= 52.5

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json