Why did we open-source our inference engine? Read the post

← Catalog

gliner-community/gliner_small-v2.5

Open comparison →

Primitive: /extract · Extract · DeBERTa

GLiNER v2.5 is the community-maintained release of the GLiNER zero-shot NER family, identifying arbitrary entity types with a bidirectional transformer encoder

MultilingualEntities

Overview

Hardware: — drives latency, throughput & cost

Size60M params
Tasks /extract
Licenseapache-2.0
Languagesmultilingual
Latency
Throughput
Cost /1M tok

Cost is approximate — computed from list GPU prices; your actual price depends on the provider you deploy SIE with.

Extraction

Output kindsEntities
Inputstext
Max sequence length

Benchmarks

CoNLL-2003

news ner en

Named entity recognition on Reuters newswire text

Corpus: 3,453 Queries: 3,453
Quality
f1 0.6691
precision 0.6481
recall 0.6916
Reference →

Open source inference for agents

Open-source inference for the models behind your agents. Run it yourself, or let us run it for you.

Github 2.8K

Contact us

Tell us about your use case and we'll get back to you shortly.

Apply for an inference grant

Free capacity on our hosted cluster for selected projects. Tell us what you run and we reply by email.