---
title: sentence-transformers/all-MiniLM-L6-v2
description: "This is a sentence-transformers model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like. BERT, 23M parameters."
canonical_url: https://superlinked.com/models/sentence-transformers-all-minilm-l6-v2
last_updated: 2026-05-25
---

# sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search.

Source: [sentence-transformers/all-MiniLM-L6-v2 on HuggingFace](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2)

## Overview

| Field | Value |
|-------|-------|
| Architecture | BERT |
| Parameters | 23M |
| Tasks | Encode |
| Outputs | Dense |
| Dimensions | Dense: 384 |
| Max sequence length | 256 tokens |
| License | apache-2.0 |
| Inputs | text |
| Languages | en |

## Benchmarks

### CQADupstackPhysicsRetrieval

Domain: scientific · Task: retrieval · Language: en

Duplicate question retrieval from StackExchange Physics

Corpus: 38,314 · Queries: 1,039

**Quality:** ndcg at 10: 0.4698 · map at 10: 0.4073 · mrr at 10: 0.4632

**Performance (A10G b1 c16):** Corpus 1.7K tok/s · Corpus p50 1.1s · Query 512 tok/s · Query p50 327.0ms

**Performance (L4 b1 c16):** Corpus 37.0K tok/s · Corpus p50 53.3ms · Query 2.8K tok/s · Query p50 49.8ms

[Reference](http://nlp.cis.unimelb.edu.au/resources/cqadupstack/)

### CosQA

Domain: technology · Task: retrieval · Language: en

Code search with natural language queries

Corpus: 6,267 · Queries: 500

**Quality:** ndcg at 10: 0.3288 · map at 10: 0.2577 · mrr at 10: 0.2885

**Performance (A10G b1 c16):** Corpus 1.1K tok/s · Corpus p50 750.0ms · Query 299 tok/s · Query p50 296.9ms

**Performance (L4 b1 c16):** Corpus 17.1K tok/s · Corpus p50 50.1ms · Query 1.8K tok/s · Query p50 45.5ms

[Reference](https://arxiv.org/abs/2105.13239)

### FiQA2018

Domain: finance · Task: retrieval · Language: en

Financial opinion mining and question answering

Corpus: 57,599 · Queries: 648

**Quality:** ndcg at 10: 0.3687 · map at 10: 0.2914 · mrr at 10: 0.4451

**Performance (A10G b1 c16):** Corpus 1.9K tok/s · Corpus p50 1.3s · Query 587 tok/s · Query p50 345.4ms

**Performance (L4 b1 c16):** Corpus 46.3K tok/s · Corpus p50 52.4ms · Query 3.7K tok/s · Query p50 43.8ms

[Reference](https://sites.google.com/view/fiqa/)

### LegalBenchConsumerContractsQA

Domain: legal · Task: retrieval · Language: en

Question answering on consumer contracts

Corpus: 153 · Queries: 396

**Quality:** ndcg at 10: 0.6560 · map at 10: 0.5883 · mrr at 10: 0.5874

**Performance (L4 b1 c16):** Corpus 128.8K tok/s · Corpus p50 58.9ms · Query 4.3K tok/s · Query p50 59.1ms

[Reference](https://huggingface.co/datasets/nguha/legalbench)

### NFCorpus

Domain: medical · Task: retrieval · Language: en

Biomedical literature search from NutritionFacts.org

Corpus: 3,593 · Queries: 323

**Quality:** ndcg at 10: 0.3160 · map at 10: 0.1105 · mrr at 10: 0.5040

**Performance (L4 b1 c16):** Corpus 81.3K tok/s · Corpus p50 54.7ms · Query 1.5K tok/s · Query p50 44.7ms

[Reference](https://www.cl.uni-heidelberg.de/statnlpgroup/nfcorpus/)

### NanoFiQA2018Retrieval

Domain: finance · Task: retrieval · Language: en

Smaller subset of the FiQA financial QA dataset

**Quality:** ndcg at 10: 0.4774 · map at 10: 0.3931 · mrr at 10: 0.5476

**Performance (L4 b1 c16):** Corpus 44.2K tok/s · Corpus p50 56.1ms · Query 2.8K tok/s · Query p50 49.9ms

[Reference](https://sites.google.com/view/fiqa/)

### SCIDOCS

Domain: scientific · Task: retrieval · Language: en

Citation prediction, document classification, and recommendation for scientific papers

Corpus: 25,656 · Queries: 1,000

**Quality:** ndcg at 10: 0.2164 · map at 10: 0.1294 · mrr at 10: 0.3594

**Performance (L4 b1 c16):** Corpus 55.3K tok/s · Corpus p50 51.8ms · Query 4.1K tok/s · Query p50 43.1ms

[Reference](https://allenai.org/data/scidocs)

### SciFact

Domain: scientific · Task: retrieval · Language: en

Scientific claim verification using research literature

Corpus: 5,183 · Queries: 300

**Quality:** ndcg at 10: 0.6451 · map at 10: 0.5959 · mrr at 10: 0.6047

**Performance (L4 b1 c16):** Corpus 75.7K tok/s · Corpus p50 52.8ms · Query 5.8K tok/s · Query p50 44.5ms

[Reference](https://github.com/allenai/scifact)

### StackOverflowQA

Domain: technology · Task: retrieval · Language: en

Programming question answering from Stack Overflow

Corpus: 19,931 · Queries: 1,994

**Quality:** ndcg at 10: 0.8396 · map at 10: 0.8117 · mrr at 10: 0.8117

**Performance (L4 b1 c16):** Corpus 58.9K tok/s · Corpus p50 56.0ms · Query 65.7K tok/s · Query p50 61.5ms

[Reference](https://arxiv.org/abs/2407.02883)