Skip to content
Bencher

Models · EuroLLM (UTTER project consortium) · Out sinceReleased 5 Dec 2025

EuroLLM 22B Instruct (2512)

Only from the makerNot on OpenRouterAvailable nowGenerally availableCan run on your own serversOpen weights

EuroLLM 22B Instruct (2512) is made by EuroLLM (UTTER project consortium). We don't have enough test results yet to rank it. Its makers have published it, so you can run it on your own servers.

EU-funded dense 22.6B model trained from scratch on ~4T tokens covering all 24 EU languages incl. Swedish, Danish, Finnish plus Norwegian; Apache 2.0; 32K context; base EuroLLM-22B-2512 published; card includes the full Axolotl SFT config. Date = HF repo creation (no announcement post found).

Writing & creativity
—
Research & analysis
—
Coding
—
Price
Price unknown— / —
Price per 1M (blended)Blended / 1M
—
MemoryContext
33K
Longest answerMax output
—
Test resultsResults
1 (1 independent1 indep.)
Out sinceReleased
5 Dec 2025
Made byVendor
EuroLLM (UTTER project consortium)
UnderstandsInputs
text

Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.

Thinking levels

Reasoning settings

You can't choose how long this one thinks.

No adjustable reasoning setting listed.

Read more

Links

Run it yourself

Self-hosting

On your own servers,Licence, sizeunder your own control.and hardware.

Compare all open models →All open-weights models →

LicenceLicence
Apache-2.0
OK for business useCommercial use
Yes
SizeParams (total / active)
22.6B
Can be trained furtherBase model
Yes ↗

Hardware you'd need

Hardware tier & memory

Runs on a powerful laptop or workstation.

≤ 24 GB at 4-bit (one consumer GPU / 32 GB Mac). Weights ≈ 24 GB at 8-bit, 12 GB at 4-bit (+10–30% for KV cache).

Languages: Languages: All 24 EU languages incl. Swedish, Danish, Finnish + Norwegian, Catalan, Galician, Turkish, Ukrainian, Arabic, Chinese, Hindi, Japanese, Korean, Russian.

Training it further: Fine-tuning: Card publishes the full Axolotl SFT config used for the instruct model (base EuroLLM-22B-2512, EuroBlocks-SFT-2512 data, 32K seq len).

Quantisations: bf16, gptq / gguf (community). Engines: Transformers.

Download (Hugging Face)Weights on Hugging Face ↗huggingface.co ↗huggingface.co ↗

Every test result

Every result

All the numbers,with where they came from.

Every result we've found for this model, with who measured it and a link to where we read it.

All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.

All results for this model
NotesNotes
EuroEval Swedish (generative)2.3Defaultutter-project/EuroLLM-22B-Instruct-2512Independent testIndependentEuroEval (Alexandra Institute) ↗29 Sep 2026—rank tier 12; ±0.12; lower is better; base EuroLLM-22B-2512 2.52; SweDN 37.83, Skolprov 50.05, Swedish facts 30.46, ScaLA-sv 61.85