Your Search FoundationSupercharged.

API Key & UsageMore

Chatsax API Key

Authorization

Connect your key to run requests with your assigned limits.

Browser Engine (Quality/Speed)

X-Engine

Choose how the target page is fetched and rendered.

Content Format

X-Return-Format

Select the format that best fits your downstream workflow.

JSON Response

Accept

Include the title, source URL, content and metadata.

Timeout (seconds)

X-Timeout

Maximum page-load wait. Enable to set a custom limit.

Token Budget

X-Token-Budget

Stop the request if its token budget is exceeded.

Use ReaderLM-v2

X-Respond-With

Model-based extraction for complex pages. Uses more tokens.

Extract Only (CSS Selector)

X-Target-Selector

For example: article, .main-content, #post-body.

Wait For (CSS Selector)

X-Wait-For-Selector

Wait for content that is added by JavaScript.

Exclude (CSS Selector)

X-Remove-Selector

Remove navigation, advertisements or other unwanted elements.

Remove All Images

X-Retain-Images

Omit image references from the extracted output.

OpenAI Citation Format

X-Retain-Links

Use citation markers for compatible web agents.

Links Summary Section

X-With-Links-Summary

Add a navigable list of links after the content.

Images Summary Section

X-With-Images-Summary

Collect image references in a separate section.

Browser Viewport Size

viewport

Set the viewport width and height. Requires POST.

Forward Cookie

X-Set-Cookie

Send a cookie to the target page. Do not share request examples containing credentials.

Image Caption

X-With-Generated-Alt

Generate descriptions for images without alt text.

Use a Proxy Server

X-Proxy-Url

Optional public HTTP(S) proxy URL.

Use a Country-Specific Proxy Server

X-Proxy

A two-letter country code, or auto.

Bypass Cached Content

X-No-Cache

Fetch the latest page instead of a cached copy.

Cache Tolerance (seconds)

X-Cache-Tolerance

Maximum age of an acceptable cached response.

Page Ready Timing

X-Respond-Timing

Choose when extraction should begin.

Custom User-Agent

X-User-Agent

Override the browser user agent used at the target.

Custom Referer

X-Referer

Set the referring URL seen by the target website.

Preserve Base64 Images

X-Keep-Img-Data-Url

Keep inline image data URLs in the output.

Do Not Cache or Track

DNT

Ask the upstream service not to cache this content.

GitHub Flavored Markdown

X-No-Gfm

Control extensions such as tables and task lists.

Stream Mode

Accept

Receive results incrementally through server-sent events.

Customize Browser Locale

X-Locale

For example: en-US or zh-CN.

Respect robots.txt

X-Robots-Txt

Name of the crawler whose rules should be checked.

Include iframe Content

X-With-Iframe

Extract text from embedded frames.

Include Shadow DOM

X-With-Shadow-Dom

Read content inside web components.

Use Final URL as Base

X-Base

Resolve relative links against the final page URL.

Run JavaScript Before Extraction

injectPageScript

JavaScript evaluated on the target page before extraction. Requires POST.

Heading Style

X-Md-Heading-Style

Select Markdown heading syntax.

Horizontal Rule Style

X-Md-Hr

Markdown characters for horizontal rules.

Bullet Point Style

X-Md-Bullet-List-Marker

Character used to begin list items.

Emphasis Style

X-Md-Em-Delimiter

Delimiter for emphasized text.

Strong Emphasis Style

X-Md-Strong-Delimiter

Delimiter for bold text.

Link Style

X-Md-Link-Style

Inline or reference-style Markdown links.

EU Residency

X-Chatsax-Region

Use the upstream experimental EU processing endpoint.

Local PDF/HTML file

POST

Upload a PDF or HTML file up to 5 MiB. HTML uses the URL as its base.

Request
curl 'https://chatsax.com/r/' \
  -H "Authorization: Bearer $CHATSAX_API_KEY"
Your response will appear here.

Your API key

Use this key right away. Sign in to manage your account.

For Better Search

Read the web. Understand your data. Find what matters.

Our Publications

Research by Jina AI. Titles and links identify the original publications.

arXiv

jina-reranker-v3.5: An Efficient Listwise Reranker with Hybrid Attention and Self-Distillation

SIGIR 2026

jina-embeddings-v5-omni: Geometry-preserving Embeddings via Locked Aligned Towers

SIGIR 2026

jina-embeddings-v5-text: Task-Targeted Embedding Distillation

ICLR 2026

Embedding Compression via Spherical Coordinates

arXiv

Vision Encoders in Vision-Language Models: A Survey

ICLR 2026

Jina-VLM: Small Multilingual Vision Language Model

AAAI 2026

jina-reranker-v3: Last but Not Late Interaction for Document Reranking

NeurIPS 2025

Efficient Code Embeddings from Code Generation Models

EMNLP 2025

jina-embeddings-v4: Universal Embeddings for Multimodal Multilingual Retrieval

ICLR 2025

ReaderLM-v2: Small Language Model for HTML to Markdown and JSON

ACL 2025

AIR-Bench: Automated Heterogeneous Information Retrieval Benchmark

ICLR 2025

jina-clip-v2: Multilingual Multimodal Embeddings for Text and Images

ECIR 2025

jina-embeddings-v3: Multilingual Embeddings With Task LoRA

SIGIR 2025

Late Chunking: Contextual Chunk Embeddings Using Long-Context Embedding Models

EMNLP 2024

Jina-ColBERT-v2: A General-Purpose Multilingual Late Interaction Retriever

WWW 2025

Leveraging Passage Embeddings for Efficient Listwise Reranking with Large Language Models

ICML 2024

Jina CLIP: Your CLIP Model Is Also Your Text Retriever

arXiv

Multi-Task Contrastive Learning for 8192-Token Bilingual Text Embeddings

arXiv

Jina Embeddings 2: 8192-Token General-Purpose Text Embeddings for Long Documents

EMNLP 2023

Jina Embeddings: A Novel Set of High-Performance Sentence Embedding Models