SAVRN
Search Contact SAVRN

Open-weight model

LIBE

by Ca' Foscari NLP cafoscari-nlp/LIBE

basemodel: - sentence-transformers/LaBSE - language-identification - langID - sentence-embedding - similarity-search - low-resource - low-resource-languages

Parameters471M
Context512
Weights1.9 GB
Licensecc-by-nc-sa-4.0
AccessOpen weights
Monthly Downloads

Runs On

What it takes to serve LIBE (471M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

PrecisionWeightsMemory neededCheapest setupPer hourAlso fits
16-bit 0.9 GB 1.1 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
8-bit 0.5 GB 0.6 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00
4-bit 0.2 GB 0.3 GB 1x MI300X (192 GB)
Vultr
$1.85 1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

basemodel: - sentence-transformers/LaBSE - language-identification - langID - sentence-embedding - similarity-search - low-resource - low-resource-languages

Excerpt from the card by Ca' Foscari NLP, licensed cc-by-nc-sa-4.0.

Configuration

Architecture
BertModel
Context length (tokens)
512
Layers
12
Hidden size
768
Feed-forward size
3,072
Attention heads
12
Vocabulary size
501,153
Stored precision
float32
Model type
bert

Identity and Version

Repository
cafoscari-nlp/LIBE
Publisher
Ca' Foscari NLP
Task
Not stated by the source
Modality
Other
Library
Not stated by the source
Parameters
471M parameters
Languages
low-resource
Revision
129e2a1de5beae561d293e268847a22d5548c629
First published
2026-09-18
Last updated
2026-09-18

Files and Weights

12 files, 1.9 GB in total. The weights are 1 file totalling 1.9 GB in safetensors.

Weights1 file · 1.9 GB
Configuration6 files · 2.3 KB
Tokenizer3 files · 18.9 MB
Documentation1 file · 225 B
Repository1 file · 1.6 KB
Every file
FileTypeSizeSHA-256
model.safetensorsWeights1.9 GB de003d1b5e48
1_Pooling/config.jsonConfiguration296 B
config.jsonConfiguration860 B
config_sentence_transformers.jsonConfiguration201 B
modules.jsonConfiguration229 B
sentence_bert_config.jsonConfiguration53 B
special_tokens_map.jsonConfiguration695 B
README.mdDocumentation225 B
.gitattributesRepository1.6 KB
tokenizer.jsonTokenizer13.6 MB e160a07b3b76
tokenizer_config.jsonTokenizer1.3 KB
vocab.txtTokenizer5.2 MB

License and Download

License
cc-by-nc-sa-4.0
Access
Open weights, no gate
Download size
1.9 GB
Download from Ca' Foscari NLP

Released by Ca' Foscari NLP through its official repository on Hugging Face.

Built From

Memory Requirements

PrecisionWeights in memory
As published1.9 GB
16-bit0.9 GB
8-bit0.5 GB
4-bit0.2 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About LIBE

How much GPU memory does LIBE need?

About 1.1 GB at 16-bit and 0.3 GB at 4-bit: the weights (471M parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run LIBE on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use LIBE commercially?

Not without separate permission. LIBE is released under Creative Commons Attribution-NonCommercial-ShareAlike 4.0. CC BY-NC-SA 4.0 permits non-commercial sharing and adapting with credit, and requires adaptations to use the same license. Commercial use needs separate permission.

What is LIBE's context length?

512 tokens, from the maximum position embeddings in its published configuration.