Skip to the page
Get started free

Exa Deep search alternative in 2026

A cited report from $0.10.

Exa's deep search type costs $12 per 1,000 requests and answers in 4 to 15 seconds. deep-reasoning is a legacy type at $15, and Exa sends long-running research to Exa Agent, at $0.012 to $1.00 a request.

SpecStream is the live web data engine for teams and agents. POST /v1/research returns a written report with the page, read time and passage behind each cited sentence, at $0.10, $0.30 or $1.00. Exa's /search call costs $1 per 1,000 on the same key.

Get an API key

New accounts get $10 of credit, no card. It covers 100 quick reports, 33 standard reports or 10 deep reports.

SpecStream and Exa's research options side by side

SpecStreamExa
A written research reportPOST /v1/research: $0.10 quick, $0.30 standard, $1.00 deep, one flat price per depthExa Agent for long-running research: $0.012, $0.025, $0.10, $0.50 or $1.00 a request at a fixed effort, or metered up to a $5 cap on auto
A search with a synthesized answerSearch API at $5 per 1,000 searches of up to 10 results, and $1 on the fast tierdeep and deep-lite at $12 per 1,000 requests, deep-reasoning (legacy) at $15
Your Exa /search codeSet the base address to https://api.specstream.ai/exa. POST /exa/search answers in Exa's shape at $1 per 1,000 searches/search at $4 per 1,000 requests on lite and $7 on auto
How long it takesA report: about one minute quick, four minutes standard, five minutes deepdeep-lite about 4 seconds, deep 4 to 15 seconds, deep-reasoning 12 to 40 seconds
What comes backreport.content with [1] markers, report.citations with the page, the read time and the passage of each cited sentence, and report.sources with every page readThe selected results under results. With an outputSchema, output.content and field-level citations in output.grounding
JSON in place of text"output": {"type": "json", "schema": {...}} with your JSON SchemaoutputSchema on the /search request
The price before the runThe reply to the POST shows price_usd. On MCP, start_research runs only with that price approved as confirm_max_centsA price per request by type. A metered Agent run bills actual usage up to its cap
When you payOnce, when the report is delivered. A cancelled report is not chargedPer request, from credits you load
Pages and rows on the same keyExtract API at $0.80 per 1,000 pages, and Deep Search rows at 5¢, 15¢ or $1 eachContents at $1 per 1,000 pages. Agent adds $0.02 per email and $0.07 per phone number
Starting credit$10 of credit on a new account, no card: 100 quick reports or 10,000 searches through the Exa address$10 in credits on the Free Tier the day you sign up

Why Exa Deep users move to SpecStream

A report where a deep request returns results

A deep request is a search: it returns the selected results, and a synthesized answer when you add an outputSchema. For long-running research Exa points to Exa Agent, a second product with its own prices. On SpecStream the research answer is one call. POST /v1/research reads the web for about one to five minutes and returns the written report with its citations.

Three prices, known when the report starts

Exa Agent has five fixed efforts from $0.012 to $1.00 a request, and its default auto mode bills usage up to a $5 cap. A SpecStream report costs $0.10, $0.30 or $1.00 by depth, and the reply to the POST shows it as price_usd. The price is taken once, when the report is delivered.

Each cited sentence shows its page and when it was read

Exa returns output.grounding with citations per field. SpecStream's report.citations gives each cited sentence the URL of its page, the time SpecStream read it and the passage that supports it, and report.sources lists every page read in the run. Your reviewer opens the page and finds the passage in seconds.

The same /search call at $1 per 1,000

Requests that need ranked pages keep their code. Set the base address to https://api.specstream.ai/exa, and POST /exa/search answers in Exa's shape for $1 per 1,000 searches of up to 10 results, against $12 on deep and $7 on auto. The same key runs the Search API, which scores 97.9 F-score on all 1,000 SimpleQA Verified questions, and the Extract API at $0.80 per 1,000 pages. The benchmarks page shows each run. See research reports and Exa pricing.

Other Exa Deep search alternatives

Linkup Research

SpecStream writes a cited report for $0.10 to $1.00 in about one to five minutes. Linkup Research costs $0.25, $0.50, $1.50 or $2.50 a request by depth, and its pricing page gives 1 to 10 minutes a request. A SpecStream standard report costs $0.30, under Linkup's M depth at $0.50.

Parallel Task API

Parallel prices Task runs per 1,000 by processor, from $5 to $2,400. Its deep research runs on pro at $100 per 1,000 runs, in 2 to 10 minutes, and on ultra at $300, in 5 to 25 minutes. A SpecStream quick report costs the same $0.10 as a pro run and takes about a minute.

Tavily Research

Tavily Research bills 4 to 110 credits a request on mini and 15 to 250 on pro, at $0.008 a credit pay as you go. That is $0.032 to $0.88 on mini and $0.12 to $2.00 on pro, known after the run. A SpecStream report has one price per depth, shown as price_usd when the report starts.

How SpecStream scores

The same numbers as the benchmarks page, from the official tasks and scorers.

SpecStreamParallelExaTavilyPerplexity
Web researchWideSearch, GPT-5.6 Luna60.945.553.047.947.0
Factual accuracySimpleQA Verified, GPT-5.6 Luna9694919494
Factual accuracySimpleQA Verified, GPT-5.6 Sol9997919295

How to move from Exa Deep

1. Create a key under Agents & API in the workspace. A new account gets $10 of credit, no card.

2. For a research answer, send your query as input to POST /v1/research and choose depth: quick, standard or deep. The call answers 202 at once with the report's id.

curl -X POST https://api.specstream.ai/v1/research \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"input": "Compare how major database vendors support vector, keyword, and hybrid retrieval", "depth": "standard"}'

# 202 at once, with the report's id and price_usd. Then read it:
curl https://api.specstream.ai/v1/research/<id> \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY"

3. Read status until it is completed, or follow GET /v1/research/{id}/events. The text is in report.content and the cited sentences are in report.citations. Where your request has an outputSchema, send it as "output": {"type": "json", "schema": {...}}.

4. For plain search calls, keep the Exa SDK and change one argument:

from exa_py import Exa

exa = Exa(api_key=SPECSTREAM_KEY, base_url="https://api.specstream.ai/exa")
result = exa.search("vector database hybrid retrieval", num_results=10)

From an agent, connect the MCP server at https://api.specstream.ai/mcp: research_price shows the price of each depth, start_research starts the report, and get_research reads it. The research reports guide has every field, and the Exa migration guide has each step for the Exa address.

Get an API key

Frequently asked questions

Something we didn't answer? Talk to a human

How much does Exa Deep search cost?
$12 per 1,000 requests on deep and deep-lite, which is $0.012 a request, and $15 per 1,000 on the legacy deep-reasoning type. The base price covers up to 10 results.
Is Exa deep-reasoning deprecated?
It is a legacy type since October 19, 2026, and it keeps its price of $15 per 1,000 requests. Exa points new requests to deep, or to Exa Agent for long-running research, list building and multi-hop enrichment.
What is the best Exa Deep alternative for research?
SpecStream, at $0.10, $0.30 or $1.00 a report. Each report comes back in about one to five minutes, and each cited sentence carries the URL of its page, the time it was read and the passage.
How long does a SpecStream research report take?
About 1 minute for a quick report, 4 minutes for a standard report and 5 minutes for a deep report. The call answers 202 at once, and GET /v1/research/{id}/events streams the progress.
What does a failed or cancelled report cost on SpecStream?
$0. The price is taken once, when the report is delivered, and a cancelled report is not charged.
Does SpecStream answer Exa's search call?
Yes, at $1 per 1,000 searches of up to 10 results. Set the base address to https://api.specstream.ai/exa, and POST /exa/search answers in Exa's shape with the key in x-api-key or as a Bearer token.
Does SpecStream have an MCP server for research?
Yes, at https://api.specstream.ai/mcp, with 3 tools for reports: research_price, start_research and get_research. start_research runs only after your person approves the price as confirm_max_cents.

Sources

SpecStream wrote this page. Each figure about another provider comes from that company's own page, read on the date shown.

  1. Exa pricingexa.ai/docs/admin/pricing.md, read Nov 13, 2026
  2. Exa Deep Searchexa.ai/docs/search/deep-search.md, read Nov 13, 2026
  3. Exa changelogexa.ai/docs/changelog/pricing-update.md, read Nov 13, 2026
  4. Linkup pricinglinkup.so/pricing, read Nov 13, 2026
  5. Linkup pricing docsdocs.linkup.so/pages/documentation/development/pricing.md, read Nov 13, 2026
  6. Parallel pricingdocs.parallel.ai/getting-started/pricing.md, read Nov 13, 2026
  7. Tavily API creditsdocs.tavily.com/documentation/api-credits.md, read Nov 13, 2026
  8. SpecStream research reportsspecstream.ai/docs/research, read Nov 13, 2026
  9. SpecStream pricingspecstream.ai/pricing, read Nov 13, 2026
  10. SpecStream benchmarksspecstream.ai/benchmarks, read Nov 13, 2026
  11. SpecStream for agentsspecstream.ai/agents.md, read Nov 13, 2026