[mg]

As far as I can tell, that makes for 5 independent image search engines on the web:

    Baidu
    Bing
    Brave
    Google
    Yandex

You can compare their results on this search comparison page I maintain:

https://www.gnod.com/search/?engines=p,o,br,n,q&nw=1

(If you want to also search image libraries like Flickr and Pexels, click on "more engines" to select all places you want to search)

[esperent]

I accidentally found a good test string for image search a few months ago:

Bamboo sign

Give it a try on Google images. You'll see that nearly all the results are x-rays of people with ankylosing spondylitis, a form of which is commonly referred to as "bamboo spine".

I tested out brave search - it correctly shows 90% signs however it does also show a few spines.

Google still shows incorrect results months later. It's by far the worst of all the search engines in the list for this simple and obvious search.

[firejake308]

Actually, the Google result is not incorrect. In radiology, an X-ray of a patient with ankylosing spondylitis is said to display the "bamboo sign", a real medical term describing the appearance of the backbones. Google is showing the result that is more relevant to thousands of medical students (such as myself) who are trying to understand a radiology report, but that may not be relevant to someone who just wants a sign made out of bamboo

[silisili]

Nice!

I actually like Brave here for my test better than Google. I typed in a few cities, just wanting to see the skylines and such.

Brave gives me good photos, some stock photos, etc. Google gives me pictures from recent news articles, which isn't overly helpful IMO.

[isaacremuant]

Brave is the one that censors less, from all those. Specially doesn't censor for political motives that I'm aware of.

That already makes it worth of support.

But Google having become so bad of late has made switching quite easy, even if brave is not getting better super fast, Google unfortunately is getting worse and making up for it.

[yreg]

Interesting, I regularly use both and I find Google to perform better for me than Brave (in text search).

[jorvi]

I have a deeeeeeep dislike for Google’s “must include: duck | missing: duck”.

[magicalist]

> Specially doesn't censor for political motives that I'm aware of

What are the censored image searches you found?

[reitanqild]

Try to search for the 1989 Tiananmen Square protests and massacre.

That tends to upset some engines including Bing I think.

[the_third_wave]

I just did and found that Baidu is the only one to censor all images related to the protests and massacre. Bing may filter out these images when used in China (I tried searching in Chinese with the browser set to request Chinese-language results, this did not affect the search results) but it does not seem to do so for me (Sweden, Linux, Firefox).

[yreg]

Baidu doesn't show anything relevant as expected, but Bing, Brave, Google and Yandex show similar results. Not overly graphic, but the photos are there.

[jaystraw]

I'm sure this isn't in the spirit or meaning of what you said, but to respond jokingly, I haven't found many censored anythings :-P turns out...

[isaacremuant]

I meant for text. Haven't really used the image feature yet.

[ColinHayhurst]

You missed out Mojeek. For an independent take: https://seirdy.one/posts/2021/03/10/search-engines-with-own-...

[SomeoneOnTheWeb]

Shouldn't Kagi (https://kagi.com) also be on that list?

[MildRant]

Someone will correct me if I'm wrong but Kagi uses Google search results. I'm sure it's more complicated than that and they have their own secret sauce but it is not an independent search engine.

See: https://help.kagi.com/kagi/why-kagi/kagi-vs-google.html

[Terretta]

> Someone will correct me if I'm wrong but Kagi uses Google search results.

Click two links down in the same menu:

https://help.kagi.com/kagi/why-kagi/kagi-vs-brave.html

Kagi Search includes anonymized requests to traditional search indexes like Google and Bing as well as sources like Wikipedia, DeepL, and other APIs. We also have our own non-commercial index (Teclis), news index (TinyGem), and an AI for instant answers. Teclis and TinyGem are a result of our crawl through millions of domains, focusing primarily on non-commercial, high-quality content.

Our unique results combined from all of these sources help you discover the best content you can possibly find online, sometimes from the quieter places on the web.

[MildRant]

I would say using Google and Bing makes it not independent but they could invest in further developing their own index if they wanted to go independent.

[anderber]

I believe you're correct. Kagi just uses Google's API and makes some changes on top of it.

[fsflover]

One more, which is self-hosted, peer-to-peer and FLOSS: https://yacy.net

[vGPU]

Yacy search results/ranking are worse than useless

[fsflover]

Because too few people are participating.

[vGPU]

I run a senior peer.

The issue is the metadata ranking is garbage, the documentation extremely lacking, and the support is nonexistent. The software is confusingly laid out, with various options strewn all over the place and absolutely no clue given as to what they do.

If you want a peer driven search engine, perhaps the quick start guide shouldn’t involve having to read the source code.

[NooneAtAll3]

does searx count as independent?

[esperent]

It's not a search engine, I think? It's a search aggregator - it combines results from other search engines.

[ShrigmaMale]

ahrefs

c.f. yep.com