Embeddings

Multilingual, multimodal embeddings for your search.

APIPricing

Embeddings API

Explore the API. Make a real request. Use the code in your application.

API Key & UsageMore

Select embeddings

Downstream task

task

Optimize embeddings for your application.

Truncate at maximum length

truncate

Truncate inputs that exceed the model context window.

Output dimensions

dimensions

Smaller vectors use less storage and bandwidth.

L2 normalization

normalized

Normalize output vectors to unit length.

Output data type

embedding_type

Choose float or a compressed representation.

Process a batch

curl 'https://chatsax.com/v1/embeddings' \
  -X 'POST' \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $CHATSAX_API_KEY" \
  --data-binary '{
  "model": "",
  "input": [
    "",
    "",
    ""
  ],
  "normalized": true,
  "embedding_type": "float",
  "task": "retrieval.query",
  "dimensions": 1024,
  "truncate": false
}'

Input

Your response will appear here.

API keyNot connected

Meaning, represented.

Connect multilingual text and multimodal content to the same retrieval workflow, with model-specific controls for tasks and dimensions.

Multilingual retrieval

Match questions and documents across supported languages.

Multiple modalities

Choose text encoders or omni models for images, audio, video, and PDFs.

Storage choices

Use supported dimensions and encoding formats to fit your vector index.

From playground to production.

Use the same request contract in the browser, an HTTP client, or your agent tools. A Chatsax key keeps access and usage separate from the upstream provider account.

Open dashboard Connect an agent Read the docs

Usage & access

Know what your requests use.

Chatsax currently uses invitation-based access. Request limits are assigned to each key; provider usage is recorded when available. No public checkout or standard price plan is offered here.

Your limits

View the current key’s scopes, minute limit, and daily request limit in the dashboard.

Recorded usage

Inspect requests, response status, latency, and available token counts.

Provider costs

Different services count usage differently. A cancelled or long-running request may still consume upstream resources.

Prepare integration requirements

Frequently asked questions

Which model should I start with?

v5-text-nano is a compact text option; v5-text-small provides a larger text encoder. Choose omni when your inputs include other modalities.

Can I switch an existing index to a new model?

Check vector-space compatibility first. A different model or task may require re-embedding your documents.

Can I self-host the weights commercially?

Hosted API access and weight licensing are separate. Review each model’s license and commercial authorization.

Is Chatsax operated by Jina AI?

Chatsax is independently operated. Jina supplies the current underlying APIs; model names and source links identify that provider.