Exa through Squiid
Neural web search that finds pages by meaning, with full content. Exa is next up on Squiid, being built now.
What Exa is
Exa indexes the web semantically. Instead of matching keywords, it embeds your query and finds pages whose meaning is close, which makes it good at requests phrased the way a person would describe what they want rather than the words a page would contain. Results can include the full cleaned text, highlights of the relevant passages, or a summary, and the find-similar endpoint takes a URL and returns pages like it. Filters cover date, domain and category such as research papers or company pages. Results can be restricted to a date range, which is how a monitoring feature avoids re-reading the same articles every morning.
In AI-coded products this is the retrieval layer for research-shaped features. Finding companies that do something specific, tracking down papers on a topic, building a list of comparable products, or gathering sources for a report are all tasks where keyword search returns SEO pages and semantic search returns the actual thing. The websets and content endpoints also make it usable as a lightweight alternative to running your own crawler. Because the index is built on embeddings, a longer and more descriptive query usually works better than a short keyword one, which is the opposite of the habit most developers have.
Why connect it through Squiid
Squiid provisions Exa instantly and holds the key. Requests go to https://api.squiid.io/v1/exa/... with Authorization: Bearer $SQUIID_API_KEY, and the official SDKs work by pointing their base URL at the gateway (gateway convention). Searches and content retrieval are metered separately against prepaid credits at Exa's list price, which is useful because asking for full text on twenty results costs more than the search itself. Everything settles on one monthly invoice with your models and databases.
- One account, one key, one bill, one dashboard. Exa sits beside every other service your project uses, on the same invoice and the same credit balance.
- Key custody. Squiid holds the Exa credentials. Your code carries
SQUIID_API_KEY, so EXA_API_KEY, EXASEARCH_API_KEY never land in a repository, a preview deployment or an agent’s context. - Metering you can see. Every call is counted per service and per day, so you know what Exa costs before the month ends.
- Pause on zero. When prepaid credits run out the gateway stops authorizing usage instead of running up a bill you did not fund.
- One rotation. Rotating the Squiid key cuts off every agent, worker and deployment at once, with no provider key to hunt down.
What you get
- Semantic search over an embeddings index
- Full page content, highlights or summaries
- Find-similar search from a URL
- Category and date filters for research use
- Exa list pricing, no markup
How to connect Exa
One variable replaces the provider credentials this service would normally need. Requests go to the gateway and the official SDKs work by pointing their base URL at it (gateway convention).
# .env
# replaces EXA_API_KEY, EXASEARCH_API_KEY
SQUIID_API_KEY=sq_live_xxxxxxxxxxxxxxxxxxxxxxxx
# every request: https://api.squiid.io/v1/exa/...
# header: Authorization: Bearer $SQUIID_API_KEY
Add it in your dashboard
Open the Squiid catalog, add Exa and fund credits. Access is instant and metered per request.
Put the one key in your env
Add SQUIID_API_KEY to .env and delete EXA_API_KEY, EXASEARCH_API_KEY. Nothing else from Exa needs to exist in your project.
Point your agent at it
Claude Code, Cursor and Codex read the same variable and call https://api.squiid.io/v1/exa/. Agents use Exa when the query describes a kind of page rather than containing the words that page uses.
Pricing through Squiid
Exa charges per search and per page of content retrieved, with different rates by search type, at the provider's listed price. Squiid passes that through at par: one dollar of prepaid credit is one dollar of Exa usage, with no markup and no per-request margin.
Exa has no subscription attached on Squiid today. If a plan is required for a feature you need, it passes through at the provider’s listed price with no Squiid fee.
Squiid adds no markup to provider prices. There is a free plan at $0 a month with a 15% fee when you top up credits, Solo at $19.97 a month with a top-up fee that starts at 8% and falls to 3.5% as your spend grows, Team from $29.97 a seat a month with the same ladder one point lower, and Custom with a rate of 3% or less. Credits are prepaid, so Exa usage stops when the balance reaches zero rather than producing an invoice you did not fund. Auto top-up with a cap keeps production running without handing anyone an open-ended card.
If you leave
Nothing to move: Exa is usage-metered, so leaving means creating your own Exa key and pointing the client at it. Squiid never owns your code, your schema or your domain, and the gateway speaks Exa's own API, so the client change is a base URL. The handover page lists what happens for every provider in the catalogue.
Works with every coding agent
Exa through Squiid is just an HTTP call with a bearer token, so every agentic coding tool can use it with the single key already in your environment:
- Claude Code
- Cursor
- ChatGPT
- Codex
- Grok
- Gemini CLI
- GitHub Copilot
- Windsurf
- Perplexity
- Devin
- Replit
- v0
- Lovable
- Cline
Vibe coding a feature on Exa usually means asking for it in plain language and letting the agent write the integration. Because the key is Squiid’s rather than the provider’s, it can do that without ever holding a credential that could be leaked, logged or committed.
Related services
Tavily
Web search and scrapingSearch API designed for LLM agents, returning clean ranked context
LiveInstantPerplexity
LLM APIsSonar models that answer from live web results with citations
RoadmapInstantFirecrawl
Web search and scrapingTurn any website into clean markdown or structured data for LLMs
LiveInstant BSBrave Search
Web search and scrapingIndependent web search index with a privacy-focused API
Next upInstant PIPinecone
Vector and searchManaged vector database for retrieval augmented generation at scale
Next upProvisionedQuestions people ask
How is Exa different from Tavily?
Exa searches by meaning over its own embeddings index and is strong at find-similar queries. Tavily is optimised to return prompt-ready context for agent loops. Both are on one Squiid key.
Can I get the full text of results?
Yes. Content, highlights and summaries can be returned with the search, and they are metered separately at Exa's listed price.
Does Squiid mark up Exa?
No. Searches and content bill at Exa's list price against your prepaid credits.
Can Cursor use Exa in a research feature?
Yes. Cursor writes the client call with SQUIID_API_KEY from the environment and the Exa key stays in Squiid's custody.
Is it good for finding companies or papers?
Those are two of its strongest categories, because semantic matching handles descriptive queries that keyword search mishandles.
Connect Exa with one key.
Exa searches the web with embeddings rather than keywords, returns page content directly, and can find pages similar to a given URL. Metered per search through Squiid at list price.