Company

Community

Check out our blog post on the future of search

Connect your AI to the internet's knowledge

Feed your AI the highest quality information from across the internet, powered by Metaphor's neural search algorithm.

Get 1000 requests/month for free

Find the hottest AI agent startups and what they do

Metaphor

from metaphor_python import Metaphor
metaphor = Metaphor("your_api_key")

results = metaphor.search("AI agent startups", use_autoprompt=True)
contents = results.get_contents()

Response

Here are hottest AI agent startups and what they do:

Name	Description
Adept	We’re building a machine learning model that can...
aiagent	Aiagent is an LLM pipeline that listens to calls...

example content from contents endpoint

Features designed for LLMs

An intelligent search algorithm for your intelligent LLM

Metaphor uses a transformer architecture to predict links from text. That means you can search with all the expressiveness of natural language.

FIND

Metaphor

adept.ai

Adept: Useful General Intelligence

generallyintelligent.com

generally intelligent

paradox.ai

Paradox: The AI Assistant for Recruiting

fixie.ai

Fixie.ai: Build on LLMs

agentrunner.ai

AgentRunner.ai

Google

blog.hubspot.com

40 AI Startups To Watch Out For in 2023

bloomberg.com

These Are the 10 AI Companies to Watch Right Now

startupsavant.com

30 Top AI Startups to Watch in 2023

crn.com

The 10 Hottest AI Startups Of 2023 (So Far)

technologymagazine.com

Top 10 AI technology startups to watch in 2023

Instant parsed HTML for any page

Don’t worry about web scraping - we instantly return rich, parsed HTML for any webpage.

Adept.ai

URL

https://adept.ai

Title

Adept: Useful General Intelligence

Content

<main>
    <div>
      <h2>A new way to use computers</h2>
      <p>We're building a machine learning model that can interact with everything on your computer.</p>
      <button>Join the waitlist</button>
      <header>
        <nav>
          <a href="/">
          </a>
          <ul>
            <li>
              <a href="/sustainable-food">
                  Mission
              </a>
            </li>
          </ul>
        </nav>
      </header>
    </div>
  </main>

Search over any subset of the internet

You can add filters to search over a set of domains and within a certain time period.

Find the latest AI agent projects

metaphor.search

includeDomains: [“github.com”, “huggingface.co”, “arxiv.org”]
startPublishDate: "2023-06-12"

github.com/llSourcell/Unity_ML_Agents/2023.06.23

Unity_ML_Agents

https://huggingface.co/spaces/camel-ai/camel-agents/2023.07.12

CAMEL: Communicative Agents for “Mind”gents

https://arxiv.org/abs/2307.02485/2023.06.28

Building Cooperative Embodied Agents Modularly with Large Language Models

https://github.com/manekinekko/minecraft-openai/2023.06.23

minecraft-openai

Integrate in 5 minutes

We have easy to use Node and Python SDKs. And you can get GPT to do all the work for you here.

OPTIONS


import Metaphor from 'metaphor-node';
const metaphor = new Metaphor('your_api_key');

await metaphor.search('a blog post about the future of search from the early days of the internet');

RESPONSE


[{
   "url": "https://idlewords.com/2003/01/permalinks_aren_t.htm",
   "title": "Permalinks Aren't (Idle Words)"
},
{
   "url": "https://googleblog.blogspot.com/2008/09/future-of-search.html",
   "title": "The future of search"
},
{
   "url": "https://kottke.org/03/04/the-anatomy-of-a-largescale-hypertextual-web-search-engine",
   "title": "The Anatomy of a Large-Scale Hypertextual Web Search Engine"
},
{
   "url": "https://fishbowl.pastiche.org/2003/10/14/google_weblogs_and_the_end_of_the_world",
   "title": "Google, Weblogs, and the End of the World - The Fishbowl: Charles Miller’s Weblog"
},
{
   "url": "https://fishbowl.pastiche.org/2003/10/09/the_internet_and_understanding_users",
   "title": "The Internet, and Understanding Users - The Fishbowl: Charles Miller’s Weblog"
},
{
   "url": "https://ideas.4brad.com/kayak-sailing-san-juan-islands",
   "title": "Kayak sailing "san juan islands""
},
{
   "url": "https://idlewords.com/2002/12/distributed_search_engines.htm",
   "title": "Distributed Search Engines (Idle Words)"
},
{
   "url": "https://avc.com/2003/10/search/",
   "title": "Search - AVC"
},
{
   "url": "https://dannysullivan.com/why-search-sucks-you-wont-fix-it-the-way-you-think-203",
   "title": "Why Search Sucks & You Won’t Fix It The Way You Think"
},
{
   "url": "https://www.wired.com/1996/05/indexweb/",
   "title": "Seek and Ye Shall Find (Maybe)"
},
{
   "url": "https://www.mrmarkets.com/1998/11/the_new_search_.html",
   "title": "THE NEW SEARCH GENERATION: Google, AskJeeves, NewHoo, MiningCo"
},
{
   "url": "http://infolab.stanford.edu/~backrub/google.html",
   "title": "The Anatomy of a Large-Scale Hypertextual Web Search Engine"
},
{
   "url": "https://searchtools.com/slides/baychi2002/index.html",
   "title": "Web Search Engines: Users Surprising the Experts"
},
{
   "url": "http://www.wiley.com/compbooks/sonnenreich/history.html",
   "title": "A History of Search Engines"
},
{
   "url": "http://blogoscoped.com/archive/2005-10-02-n67.html",
   "title": "None"
},
{
   "url": "https://www.jinfo.com/go/blog/2505",
   "title": "Jinfo Blog: Search Trails: Back to the Future"
},
{
   "url": "http://blogoscoped.com/archive/2007-02-02-n25.html",
   "title": "None"
},
{
   "url": "http://www.wiley.com/legacy/compbooks/sonnenreich/history.html",
   "title": "A History of Search Engines"
},
{
   "url": "https://thehistoryoftheweb.com/postscript/what-happened-to-the-webmaster/",
   "title": "What Happened to the Webmaster - The History of the Web"
},
{
   "url": "https://lists.w3.org/Archives/Public/www-talk/1992NovDec/0212.html",
   "title": "Searching"
},
{
   "url": "https://mmry.io/blog/20210807-the-second-coming-of-the-search-engine",
   "title": "The Second Coming of the Search Engine"
},
{
   "url": "https://web.archive.org/web/20111007073429/http://www.tbray.org/ongoing/When/200x/2003/06/15/OnSearch",
   "title": "On Search: Backgrounder"
},
{
   "url": "http://distributedsearch.blogspot.com/2008",
   "title": "Distributed Search"
},
{
   "url": "http://paulgraham.com/nsearch.html",
   "title": "Search"
},
{
   "url": "https://web.archive.org/web/20090515125503/http://searchlores.org/tadimens.htm",
   "title": "tadimens.htm: How to search the web, by fravia+: tadimens"
},
{
   "url": "https://jonudell.net/bytecols/2000-09-06.html",
   "title": "Tangled in the Threads"
},
{
   "url": "http://www.ibiblio.org/history/1sthunt.txt",
   "title": ""
},
{
   "url": "http://www.searchenginehistory.com/",
   "title": "History of Search Engines: From 1945 to Google Today"
},
{
   "url": "https://idlewords.com/2003/04/peer_to_peer_search.htm",
   "title": "Peer-to-Peer Search (Idle Words)"
},
{
   "url": "https://www.w3.org/History/1994/WWW/Journals/PhysicsWorld/PW.ascii",
   "title": ""
},
...

Join the community on the search frontier

✷ϵ∞ yayo/acc 🖤🛡️

cannot recommend metaphor.systems enough, just gets me exactly what i want as i describe it from articles to random indie github apps, no bs

0

This is _nuts_. Real-time partiful event search?!?!? As a side effect of a broader system? So much better than Google, which returns *nothing*. No words. Huge congrats to @metaphorsystems for the incredible achievement.

My wknd hack project using the (beta) @metaphorsystems and @CohereAI APIs - Search for related content from within a website. Highlight text, right click, semantic search. Even works in Twitter. Still tinkering with the algo but I've already discovered a lot!

Prasanna Srikhanta

metaphor.systems is becoming my favorite way to discover new papers that I wouldn’t normally find using my usual sources like Twitter etc😍

1

Been playing w/ LLM-powered search engine metaphor.systems by @wangzjeff + team and it's very cool! Some good use cases: - finding github repos - finding lists of particular things - "tip of my tongue" type queries of a website - bagel reviews (seriously) Check it out!

1

Metaphor makes web search actually usable again

6

FAQ

What's different about Metaphor Search?

Metaphor search is fully neural - when you search, we use a transformer-based model to predict the links most relevant to your search. The biggest difference is that in Metaphor, the results you get much more closely match the query you use, letting you for instance search by entity or descriptors.

For example, a Google search for “AI podcasts” returns almost all links like “The 11 Best AI Podcasts” whereas in Metaphor the results are all actual podcasts, neurally sorted by quality and relevance.

How often is the index updated?

How does Metaphor compare to other search APIs?

When should you not use Metaphor?

We think that if you need a web search API, Metaphor is the best option. If you need semantic, neural search, Metaphor search returns much higher quality results. And if you just need traditional search, we wrap traditional search providers with our "keyword" search type and can also get you HTML content from those pages. If you need neither page contents nor neural search, then some other providers may be cheaper.

How can Metaphor be used in an LLM?

Is it free?

What's our roadmap?

absolutely hilarious wikipedia page

companies working on fusion

recent news about interest rates

podcasts discussing biohacking

The world's knowledge at your LLM's fingertips

(metaphorically of course)

Get 1000 requests/month for free

© Copyright 2023 Metaphor