Most AI tasks no longer need a frontier model Read launch post

opensearch neural sparse encoding doc v2 mini

An open sparse model that turns a text into weighted terms for search. opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini

  • Accuracy

    8 retrieval sets, mean nDCG@10 · not measured

  • Time per request

    Not timed one request at a time

  • Serverless price

Call it

from sie_sdk import SIEClient
​
client = SIEClient("http://localhost:8080")
​
res = client.encode(
"modelopensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini",
[{"text": "textIs there a discount for students on a yearly plan?"}],
output_types=["sparse"],
)
sparse = res[0]["sparse"] # indices = term ids, values = weights
pairs = zip(sparse["indices"], sparse["values"])
top = sorted(pairs, key=lambda p: -p[1])[:12]
for term_id, weight in top:
print(f"{float(weight):.2f} {term_id}")
import { SIEClient } from '@superlinked/sie-sdk';

const client = new SIEClient('http://localhost:8080');

const [res] = await client.encode(
  'opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini',
  [{ text: "Is there a discount for students on a yearly plan?" }],
  { outputTypes: ['sparse'] },
);
// sparse: { indices, values }, mapping learned term ids to weights
const sparse = res!.sparse!;
const top = [...sparse.indices]
  .map((id, i) => ({ id, weight: sparse.values[i] }))
  .sort((a, b) => b.weight - a.weight);
console.log(top.slice(0, 12));
curl http://localhost:8080/v1/encode/opensearch-project%2Fopensearch-neural-sparse-encoding-doc-v2-mini \
  -H "Content-Type: application/json" \
  -d "{\"items\":[{\"text\":\"Is there a discount for students on a yearly plan?\"}],\"params\":{\"output_types\":[\"sparse\"]}}"
Build the "Sparse embeddings" capability into my app using the Superlinked Inference Engine (SIE).

Context
- SIE is an OpenAI-style inference API. Python SDK: `from sie_sdk import SIEClient`; TypeScript: `@superlinked/sie-sdk`.
- Base URL: http://localhost:8080, my own SIE server, which takes no API key.
- Model: opensearch-project/opensearch-neural-sparse-encoding-doc-v2-mini (SIE primitive: /encode). Keep the model id configurable.

Task
- Input: a block of text.
- Behaviour: return the learned sparse term ids and weights, ordered by strongest activation
- Call the selected SIE primitive once per request and map the response into your domain type.

Deliverables
- A typed client wrapper, an application-level function for this task, error handling for timeouts/empty input, and unit tests with a stubbed client.
- Wire it into my existing stack (ask me which framework if unclear) and add a short usage example.

Next to the other sparse models

Next to the other sparse models
Model 8 retrieval sets, mean nDCG@10 Time per request Serverless price Size
OpenSearch doc-v3-gte opensearch-project Not measured Not timed $0.0292 per 1M tokens 137M
opensearch neural sparse encoding v1 opensearch-project 0.458 Not timed Run in your cloud 133M
splade v3 naver 0.456 Not timed Run in your cloud 110M
bge m3 BAAI Not measured Not timed $0.0408 per 1M tokens 568M
All 12 sparse models

Facts

Size
23M parameters
License
Apache-2.0
Reads
Text, up to 512 tokens
Endpoint
encode
Evaluation
Accuracy
Run on SIE against the public dataset named in each figure
8 retrieval sets
The mean over NFCorpus, FiQA2018, SciFact, SCIDOCS, CQADupstackPhysicsRetrieval, CosQA, StackOverflowQA and LegalBenchConsumerContractsQA
Prices
Serverless list prices, effective 10 Nov 2026

Run opensearch neural sparse encoding doc v2 mini in your own cloud, or start on Serverless

Native Inference

Kubernetes in your own cloud, backed by our engineers.

  • Terraform modules for your cloud
  • Apache-2.0, same engine as Serverless
  • GPU workers scale to zero between jobs
  • Separate worker pools for each tenant
Agent prompt Deploy SIE to our AWS account with the superlinked/sie/aws Terraform module. Docs: superlinked.com/docs/deployment.Deploy SIE to our Google Cloud project with the superlinked/sie/google Terraform module. Docs: superlinked.com/docs/deployment.Deploy SIE to our Azure subscription with the superlinked/sie/azure Terraform module. Docs: superlinked.com/docs/deployment.Deploy SIE to our Alibaba Cloud account with the superlinked/sie/alicloud Terraform module. Docs: superlinked.com/docs/deployment.

SIE Local

The open-source server on your own machine.

  • Runs on your NVIDIA GPU
  • One container, no cluster
  • Process your data fully offline
  • Same SDK and model IDs as Serverless
docker run --gpus all -p 8080:8080 ghcr.io/superlinked/sie-server:latest-cuda12-default

Talk to 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.