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

← Catalog

Salesforce/SFR-Embedding-2_R

Open comparison →

Primitive: /encode · Encode · Mistral

Salesforce/SFR-Embedding-2_R

Long contextDense

Overview

Hardware: — drives latency, throughput & cost

Size7.1B params
Tasks /encode
Licensecc-by-nc-4.0
Languagesen
Latency682 ms
Throughput2.9K tok/s
Cost$0.076 /1M tok

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

Embedding

Output typesDense
Dimensionsdense: 4,096
Max sequence length32,768
Inputstext

Benchmarks

NFCorpus

medical retrieval en

Biomedical literature search from NutritionFacts.org

Corpus: 3,593 Queries: 323
Quality
ndcg at 10 0.4285
map at 10 0.1692
mrr at 10 0.6300
Reference →

NanoFiQA2018Retrieval

finance retrieval en

Smaller subset of the FiQA financial QA dataset

Quality
ndcg at 10 0.6468
map at 10 0.5714
mrr at 10 0.6532
Performance L4 b1 c16
Corpus 2.9K tok/s
Corpus p50 682.5ms
Query 457 tok/s
Query p50 238.4ms
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 2.3K

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.