HiNow LogoHiNow

HiEmbed

Model Information

Display Name: HiEmbed

API Model ID: hinow/hiembed

Category: Text To Embedding

Description: HiEmbed is the HiNow embeddings model: one stable name for multilingual text vectors, served over redundant routes so indexing never goes dark. **Key Features:** - 1024-dimension dense vectors - 100+ languages, including strong cross-lingual matching (a query in one language retrieves documents in another) - 8192-token context — embeds long chunks without pre-splitting - Redundant upstream routing with automatic failover **Capabilities:** - Semantic search and retrieval (RAG) - Clustering, deduplication and similarity scoring - Cross-language document matching **Best For:** - Knowledge bases mixing Portuguese and English content - RAG pipelines that need a stable, long-lived vector space **Technical Specs:** - Dimensions: 1024 - Max input: 8192 tokens - Encoding: float or base64

Context Window: 8,192 tokens

How to Use This Model

To use HiEmbed via the HInow.ai API, use the model ID: hinow/hiembed

API Request Example (Chat/Text)


POST https://api.hinow.ai/v1/chat/completions
Authorization: Bearer YOUR_API_KEY
Content-Type: application/json

{
  "model": "hinow/hiembed",
  "messages": [
    {"role": "user", "content": "Your message here"}
  ]
}

Pricing

  • input: $0.05

Available Parameters

  • encoding_format: Format of embeddings (Options: float, base64)

Quick Reference

To use this model, set: "model": "hinow/hiembed"

Featured: No

Documentation: https://hinow.ai/models/hinow/hiembed

API Endpoint: https://api.hinow.ai/v1

Back to Models
HiEmbed

HiEmbed

hinow/hiembed

$0.050
input

About

HiEmbed is the HiNow embeddings model: one stable name for multilingual text vectors, served over redundant routes so indexing never goes dark.

Key Features:

  • 1024-dimension dense vectors
  • 100+ languages, including strong cross-lingual matching (a query in one language retrieves documents in another)
  • 8192-token context — embeds long chunks without pre-splitting
  • Redundant upstream routing with automatic failover

Capabilities:

  • Semantic search and retrieval (RAG)
  • Clustering, deduplication and similarity scoring
  • Cross-language document matching

Best For:

  • Knowledge bases mixing Portuguese and English content
  • RAG pipelines that need a stable, long-lived vector space

Technical Specs:

  • Dimensions: 1024
  • Max input: 8192 tokens
  • Encoding: float or base64

Capabilities

Text To Embedding
Context8K tokens

Parameters

encoding_format

Format of embeddings

floatbase64

Code Examples

curl -X POST https://api.hinow.ai/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $HINOW_API_KEY" \
  -d '{
    "model": "hinow/hiembed",
    "input": "The quick brown fox jumps over the lazy dog",
    "parameters": {
      "encoding_format": "float"
    }
  }'