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

Linq-AI-Research/Linq-Embed-Mistral

Linq-AI-Research/Linq-Embed-Mistral

Overview

Architecture
Mistral
Parameters
7.1B
Tasks
Encode
Outputs
Dense
Dimensions
Dense: 4,096
Max Sequence Length
32,768 tokens
License
cc-by-nc-4.0
Languages
en

Benchmarks

NFCorpus

medical retrieval en

Biomedical literature search from NutritionFacts.org

Corpus: 3,593 Queries: 323
Quality
ndcg at 10 0.4066
map at 10 0.1566
mrr at 10 0.6287
Performance L4 b1 c16
Corpus 3.0K tok/s
Corpus p50 968.2ms
Query 210 tok/s
Query p50 231.6ms
Reference →

NanoFiQA2018Retrieval

finance retrieval en

Smaller subset of the FiQA financial QA dataset

Quality
ndcg at 10 0.6402
map at 10 0.5716
mrr at 10 0.6722
Performance L4 b1 c16
Corpus 2.9K tok/s
Corpus p50 667.7ms
Query 459 tok/s
Query p50 259.6ms
Reference →

Self-hosted inference for search & document processing

Cut API costs by 50x, boost quality with 85+ SOTA models, and keep your data in your own cloud.

Github 2.0K

Contact us

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