---
title: opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini
description: "The model should be selected considering search relevance, model inference and retrieval efficiency(FLOPS). We benchmark models' zero-shot p. BERT, 23M parameters."
canonical_url: https://superlinked.com/models/opensearch-project-opensearch-neural-sparse-encoding-doc-v2-mini
last_updated: 2026-06-15
---

# opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini

The model should be selected considering search relevance, model inference and retrieval efficiency(FLOPS). We benchmark models' zero-shot performance on a subset of BEIR benchmark: TrecCovid,NFCorpus,NQ,HotpotQA,FiQA,ArguAna,Touche,DBPedia,SCIDOCS,FEVER,Climate FEVER,SciFact,Quora.

Source: [opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini on HuggingFace](https://huggingface.co/opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini)

## Overview

| Field | Value |
|-------|-------|
| Architecture | BERT |
| Parameters | 23M |
| Tasks | Encode |
| Outputs | Sparse |
| Dimensions | Sparse: 30,522 |
| Max sequence length | 512 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

**Performance (L4 b1 c16):** Corpus 36.4K tok/s · Corpus p50 49.4ms · Query 4.6K tok/s · Query p50 37.1ms

[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

**Performance (L4 b1 c16):** Corpus 15.0K tok/s · Corpus p50 48.7ms · Query 2.3K tok/s · Query p50 40.4ms

[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

**Performance (L4 b1 c16):** Corpus 39.9K tok/s · Corpus p50 54.3ms · Query 4.7K tok/s · Query p50 38.5ms

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

### LegalBenchConsumerContractsQA

Domain: legal · Task: retrieval · Language: en

Question answering on consumer contracts

Corpus: 153 · Queries: 396

**Performance (L4 b1 c16):** Corpus 110.6K tok/s · Corpus p50 61.1ms · Query 7.0K tok/s · Query p50 36.7ms

[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: 0.3267 · map at 10: 0.1263 · mrr at 10: 0.5384

**Performance (L4 b1 c16):** Corpus 61.6K tok/s · Corpus p50 67.8ms · Query 1.8K tok/s · Query p50 42.1ms

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

### SCIDOCS

Domain: scientific · Task: retrieval · Language: en

Citation prediction, document classification, and recommendation for scientific papers

Corpus: 25,656 · Queries: 1,000

**Performance (L4 b1 c16):** Corpus 48.4K tok/s · Corpus p50 53.4ms · Query 4.6K tok/s · Query p50 38.3ms

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

### SciFact

Domain: scientific · Task: retrieval · Language: en

Scientific claim verification using research literature

Corpus: 5,183 · Queries: 300

**Performance (L4 b1 c16):** Corpus 63.3K tok/s · Corpus p50 57.3ms · Query 6.8K tok/s · Query p50 38.1ms

[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

**Performance (L4 b1 c16):** Corpus 53.8K tok/s · Corpus p50 54.7ms · Query 97.8K tok/s · Query p50 45.7ms

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