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

← Catalog

mixedbread-ai/mxbai-edge-colbert-v0-32m

Open comparison →

Primitive: /encode · Encode · ModernBERT

The crispy, lightweight ColBERT family from Mixedbread.

Long contextMulti-vector

Overview

Hardware: — drives latency, throughput & cost

Size32M params
Tasks /encode · /score
Licenseapache-2.0
Languagesen
Latency60 ms
Throughput45.9K tok/s
Cost$0.0048 /1M tok

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

Embedding

Output typesMulti-Vec
Dimensionsmultivector: 64
Max sequence length8,192
Inputstext

Benchmarks

CQADupstackPhysicsRetrieval

scientific retrieval en

Duplicate question retrieval from StackExchange Physics

Corpus: 38,314 Queries: 1,039
Performance L4 b1 c16
Corpus 28.3K tok/s
Corpus p50 60.5ms
Query 3.0K tok/s
Query p50 55.7ms
Reference →

CosQA

technology retrieval en

Code search with natural language queries

Corpus: 6,267 Queries: 500
Performance L4 b1 c16
Corpus 15.0K tok/s
Corpus p50 51.8ms
Query 1.9K tok/s
Query p50 48.5ms
Reference →

FiQA2018

finance retrieval en

Financial opinion mining and question answering

Corpus: 57,599 Queries: 648
Performance L4 b1 c16
Corpus 38.9K tok/s
Corpus p50 57.5ms
Query 3.7K tok/s
Query p50 48.0ms
Reference →

LegalBenchConsumerContractsQA

legal retrieval en

Question answering on consumer contracts

Corpus: 153 Queries: 396
Performance L4 b1 c16
Corpus 87.3K tok/s
Corpus p50 76.5ms
Query 5.4K tok/s
Query p50 48.3ms
Reference →

NFCorpus

medical retrieval en

Biomedical literature search from NutritionFacts.org

Corpus: 3,593 Queries: 323
Quality
ndcg at 10 0.3376
map at 10 0.1285
mrr at 10 0.5432
Performance L4 b1 c16
Corpus 67.7K tok/s
Corpus p50 58.6ms
Query 1.4K tok/s
Query p50 53.0ms
Reference →

SCIDOCS

scientific retrieval en

Citation prediction, document classification, and recommendation for scientific papers

Corpus: 25,656 Queries: 1,000
Performance L4 b1 c16
Corpus 36.4K tok/s
Corpus p50 70.2ms
Query 2.7K tok/s
Query p50 61.8ms
Reference →

SciFact

scientific retrieval en

Scientific claim verification using research literature

Corpus: 5,183 Queries: 300
Performance L4 b1 c16
Corpus 58.8K tok/s
Corpus p50 61.2ms
Query 5.4K tok/s
Query p50 49.0ms
Reference →

StackOverflowQA

technology retrieval en

Programming question answering from Stack Overflow

Corpus: 19,931 Queries: 1,994
Performance L4 b1 c16
Corpus 52.8K tok/s
Corpus p50 58.9ms
Query 68.3K tok/s
Query p50 66.6ms
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 2.1K

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.