Model reference · open weights
LLaMmlein2Vec is an open-weight embedding model from LSX-UniWue. LLaMmlein2Vec_120M (BF16) weighs 200 MB; the smallest configuration that runs it is RTX 3060 12 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the LSX-UniWue/LLaMmlein2Vec_120M model card, 2026-10-01
What it is
| Released by | LSX-UniWue |
|---|---|
| Released | 2025-07-14 |
| Parameters | 100M |
| VRAM | 200 MB for the weights |
What it runs on
| Card | Runs | Memory |
|---|---|---|
| RTX 3060 12 GB | fits | 11.6 GB |
| RTX 4060 Ti 16 GB | fits | 15.4 GB |
| RTX 3090 24 GB | fits | 23.4 GB |
| RTX 4090 24 GB | fits | 23.4 GB |
| RTX 5090 32 GB | fits | 31.0 GB |
| L40S 48 GB | fits | 44.0 GB |
| A100 80 GB | fits | 78.2 GB |
| H100 80 GB | fits | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | fits | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | fits | 107 GB |
| H200 141 GB | fits | 138 GB |
| B200 180 GB | fits | 176 GB |
From the model card
LLäMmlein2Vec 120M is a German encoder language model derived from our German decoder-only model LLäMmlein 120M via LLM2Vec.
We provide three transformed models:
LLäMmlein2Vec 120M ← You are here
Find more details in our preprint!
You can use LLäMmlein2Vec with the llm2vec library.
import torch
from llm2vec import LLM2Vec
model_id = "LSX-UniWue/LLaMmlein2Vec_120M"
l2v = LLM2Vec.from_pretrained(
model_id,
device_map="cuda" if torch.cuda.is_available() else "cpu",
torch_dtype=torch.bfloat16,
)
We release the ModernGBERT models under a research-only RAIL-M license. See license.md for details. Data Take Down
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.