Image Save As Not Working? Check Hotlink Blocking vs Cross-Origin

PublishedToolForce-Download an Image or Video URL (with Filename)

Contents
  1. Why right-click saving acts up: what you actually see
  2. Checking and Saving with Built-In Browser Features
  3. Why the Difference Happens: Reasons on Both the Browser Side and the Server Side
  4. How it handles different quality levels, formats, and situations
  5. Where people get stuck
  6. Copyright and legal use
  7. How it differs from similar tools and platforms
  8. Summary
  9. Frequently asked questions

You've probably right-clicked an image, clicked "Save image as…," and had nothing happen — or instead of a save dialog, the photo just opened large in a new tab. Even when you try to save it with a filename, it sometimes lands in your folder as a meaningless string of characters. Your browser or computer isn't broken. This happens because the image isn't hosted on the site you're looking at, but somewhere else (a different origin), and the browser is behaving exactly as designed.

In short ① For images and videos hosted on another site (a cross-origin address), the browser sometimes ignores the download filename and just opens the file in a new tab. ② A server with hotlink protection rejects requests coming from anywhere else, so no method will get you that file. ③ Only the file's direct address works (one that ends in a file extension, or whose response Content-Type is a media type). Webpage addresses, like articles or posts, are not supported.

Why right-click saving acts up: what you actually see

When you spot a photo you like on a blog post, a community board, or a product page and try to save it with a right-click, one of three things happens.

The first is that the save dialog appears normally and the file saves under the name you want. This usually happens when the image lives on the same domain as the page you're viewing.

The second is that a new tab opens instead of a save dialog, showing nothing but the photo filling the screen. If you right-click again there and try to save, you often get the same result. In Chrome the file sometimes drops straight into your Downloads folder with no dialog, and in Firefox or Safari the menu behaves a little differently, so the symptom feels different from browser to browser.

The third is that the file does save, but as "download," "unknown," or a hash of scrambled letters and numbers. Open the folder later and you have no idea which file is which. This gets especially annoying once you've saved a batch of photos.

Add hotlink protection to the mix and you see different symptoms altogether. The image itself shows up as a broken-image icon (a gray box), or "Save image as" is grayed out in the right-click menu from the start. On sites like this, an image may display fine inside the original page but suddenly disappear when you arrive through a search result or an external link, which makes it even more confusing.

Saving a photo on a mobile browser by long-pressing works on the same principle. You get blocked the same way, or end up with the same strange filenames, just as on a PC.

Take the case of saving the featured image from a blog post you're reading. The blog platform often uploads images to a separate image server (a CDN) rather than its own server, so the page address you see and the address of the actual image file are on different domains. From your side it looks like "I'm saving from the same blog, so why?" But from the browser's side, the page and the image are different origins, so different rules apply.

The same goes for animated GIFs and memes shared on a community board. The board itself may be a Korean service, but the poster often uploaded the image to an external image host and pasted in only the link. Because the board's domain and the image's domain were never the same, right-click saving seems to work on some posts and not others. Product images on marketplace listing pages behave the same way. Sellers often paste in an address from another image server, so within a single shopping mall, saving works for some products and not others.

This is why the same person on the same browser gets different results from site to site. The problem isn't your browser settings or the state of your computer. It comes down to whether that image shares an origin with the page you're viewing. Without knowing that rule, you're stuck asking "why didn't it work this time?" over and over.

You'll also run into this when searching Wikipedia or Wikimedia Commons for photos to save for research. Those sites generally don't use hotlink protection, so the image opens fine if you go to the direct address. But it's still usually on a different domain from the article page you're viewing (a dedicated upload server), so right-clicking inside the article can save it under a filename that doesn't match the page title. In other words, no hotlink protection doesn't mean the cross-origin filename problem disappears too.

One more confusing case: the right-click menu has no "Save image as" item at all. This is common when the photo isn't actually an image inserted with an img tag, but an element laid down for design purposes, like a background (a CSS background image). The browser never treats these as "savable images," so the save option doesn't appear regardless of origin.

Saving a photo straight from a search engine's image results is another familiar situation. The results screen belongs to the search engine's domain, but the photo files themselves live on entirely different sites where they were originally posted. So if you right-click right there in the results, you may end up saving the thumbnail the search engine generated, and to get the original quality you may need to visit the page where the original was posted and find the image address there. Either way, the screen you're clicking and the place the file actually lives are not the same.

Checking and Saving with Built-In Browser Features

Here's the step-by-step rundown of how far you can get with nothing but your browser's built-in features.

  1. Right-click the image and choose "Copy image address" (Chrome and Edge) or "Copy image link" (Firefox). What you're getting is the address of the image file itself, not the page it's displayed on.

  2. Paste that address into a new tab's address bar and open it. If the image loads on its own, it's a real file URL. If you get a connection refused or error screen instead, the server has hotlink protection in place.

  3. Right-click the image in that new tab and try "Save image as." If the domain in the address bar matches the domain of the page the image came from, this step usually saves the file under the name you want without any trouble.

  4. If the domains differ (cross-origin), results vary by browser. Sometimes the save dialog never appears and the image just sits there open. Other times it saves, but under whatever the end of the URL happens to be — some random hash or "index" — instead of the name the page intended.

  5. If you want a more accurate look at the file type and actual size, open the Network tab in your browser's developer tools (F12, or right-click → Inspect), reload the image, and check the Content-Type and response size of the actual request. This also tells you whether you're looking at a real image file or a page doing something else behind the scenes.

  6. On mobile, long-pressing a photo and choosing "Save image" does the same job as right-clicking on a PC. Cross-origin rules apply there too, though, so a long press may not save anything, or the image may only open inside the browser instead of going to your photo gallery.

  7. Some browsers let you drag an image straight to your desktop or a folder to save it. This method is also subject to cross-origin rules, so it works on some sites and not others. Even when it does work, the browser often assigns a filename on its own, just like right-click saving.

  8. You might also think of taking a screenshot and saving that, but that's not the same as getting the original file. You're re-photographing the pixels on screen, so the image quality is lower than the original, and the file size and format (JPG, PNG, etc.) differ from the source. If you need the file exactly as it is, a screenshot isn't a good substitute.

  9. If the right-click menu has no "Save image" option at all, you can inspect that area with the element inspector in developer tools. If it turns out to be an element inserted as a background-image style rather than an img tag, you'll have to dig the image URL out of the style properties. And once you find it, if it's on a different domain, you hit the same cross-origin problem described above.

After repeating this a few times, you reach a point where copying an address, opening a new tab to check it, and then trying to save again starts to feel tedious. That goes double when you want to save several images with tidy filenames. This is where our tool comes in: paste in the address and it identifies the file type and size first, then lets you set the filename you want and finish the save in one go.

Force-Download an Image or Video URL (with Filename)

Free · no signup · paste a file URL, check its format and size, then save it under the filename you want

If you paste in the direct URL of a photo on Wikimedia Commons (a path starting with upload.wikimedia.org), it's recognized right away as a 2.70MB original JPG file, as shown above. Check the format and size on screen, press the save button, and it downloads under the filename you chose — not whatever name the browser would have made up.

There's a simple reason for showing the format and size before you press save. When you know up front what format the file is (JPG or PNG, or video or audio) and how big it is, you're less likely to accidentally download something far larger than you expected. And when you need to check several URLs in a row, you can use it just to scan formats and sizes without downloading anything at all.

Why the Difference Happens: Reasons on Both the Browser Side and the Server Side

This problem comes from two causes overlapping: one on the browser side and one on the server side. Breaking them apart makes the confusion much easier to clear up.

When a web page provides a save link or button, it usually adds a download attribute to an anchor tag so that clicking it saves the file right away under a specified filename. But this download attribute only works fully when the image is on the same site as the current page (the same origin). If the image sits on a different domain (cross-origin), the browser ignores that filename instruction under its security policy. As a result, clicking no longer opens a save dialog — the image just opens in a new tab instead. The same reasoning explains why "Save image as…" from the right-click menu behaves differently from situation to situation: the browser treats the image differently depending on whether it lives on the same site as the page you're viewing or somewhere else.

This principle affects not just the filename but whether saving happens at all. In some cases a cross-origin image can still be saved from the right-click menu, with only the filename ignored; in other cases the save action itself turns entirely into opening a new tab. The symptoms you see vary, but the root in every case comes down to one fact: the page you're viewing and the image are on different origins.

The second reason is on the server side. Some servers are set up to reject image and video requests that come from anywhere other than their own site. This is commonly called hotlink protection. Here the issue isn't how the browser behaves — the server receiving the request decides where the request came from and chooses not to return a response. So even if you switch browsers or try another method, you can't get the file unless that server changes its decision. The "access denied" message on screen is just telling you that situation exactly as it is.

Because these two causes are completely different, whether they can be fixed differs too. The browser-side problem (a cross-origin filename being ignored) can be worked around by changing the way you fetch the file. The server-side problem (hotlink protection) is a decision the server made not to hand out the file in the first place, so no method on the client side can get around it. Knowing this distinction means that when a save fails, instead of wondering "what did I do wrong," you can start by figuring out which side the problem is on.

Understanding why these restrictions exist at all also helps you understand the workarounds. Cross-origin rules were created for web security in the first place. If any site could save another site's resources under whatever filename it wanted, malicious sites could abuse that to label an executable like a photo and trick users. That's why browsers are designed not to hand filename control to a page when the resource is cross-origin.

Hotlink protection is a defense measure in a similar spirit. From a server operator's point of view, they want to stop images and videos posted on their site from being embedded directly in other sites and burning through traffic and resources. So they configure the server to check where a request came from and reject responses for requests that don't come from their own site. This isn't a shared rule set by browsers — each server operator decides it individually, which is why one site of a given kind may have it while another doesn't.

The value servers commonly use as the basis for that judgment is referrer information sent along with the request — that is, a value that tells the server which page the request started from. If that value isn't the server's own domain, the usual hotlink protection behavior is to filter the request out by returning an error or a placeholder image instead of the image. That's why the same image URL can display fine when viewed inside its original page, yet get blocked when you copy just that URL and open it in a new tab or on another site.

It's worth looking a little closer at how file formats get identified, too. If the URL ends with an extension like .jpg or .mp4, guessing the format is easy, but plenty of URLs have no extension at all or end in a meaningless string. In that case the tool checks the Content-Type header the server sends with its response (a value that says what kind of data this response is), and if it confirms the value is an image, video, or audio type, the URL is recognized as a file. Conversely, if the URL is actually returning an HTML page, the Content-Type will say page, so it won't be recognized as a file and will be filtered out.

Our free tool, Force-Download an Image or Video URL (with Filename), handles the browser-side problem of the two. First it determines whether the URL you enter is actually an image, video, or audio file — by checking whether the URL ends with an extension, or whether the Content-Type in the server's response confirms it's media. Once it's confirmed to be a file, if the server allows cross-origin requests, the tool pulls the entire file as a Blob (a binary data object held in the browser's memory). Because that data already exists inside the browser, saving it is no longer a matter of going off to another site and specifying a filename — it's simply saving local binary data as-is. That's what lets you choose the filename you want.

When the server doesn't allow cross-origin requests at all, the tool takes a different path. It fetches the file through a signed proxy and attaches a Content-Disposition header to the response (a header that tells the browser "don't open this response on screen — save it as a file") to get around the restriction. However, when the server refuses the request outright, as with hotlink protection, even this proxy approach can't fetch it — because the server never returns a file response in the first place. These two branches are the real reason behind the distinction mentioned earlier: "cross-origin is workable, hotlink protection isn't." On the surface both look like "it won't save," but the judgment happening underneath is running along completely different tracks.

For reference, if a site offers its own "Download" button from the start, you'll almost never run into this problem. Those sites have the server attach a Content-Disposition header to the response ahead of time, so the save dialog appears right away without the browser needing to consider whether it's cross-origin. The cases where right-click saving gets murky are mostly images and videos that were never designed for downloading in the first place — they were just uploaded to be shown on screen.

So what you see on screen is ultimately the final output of this decision process. If the file format checks out and the only problem was a cross-origin restriction, you end up with a save under the filename you want; if the server refused the request from the start, nothing more happens and it ends in access denied. Even when the result is the same "it won't save," the judgment that took place inside is a different one each time.

How it handles different quality levels, formats, and situations

Here's how right-click saving and Force-Download an Image or Video URL (with Filename) differ depending on the file type and the situation. Keep in mind that the tool doesn't change image quality at all. It reads whatever the original the URL points to and turns it into something you can save.

SituationRight-click save in the browserResult with the toolWhy
Image on the same site (same domain)Usually saves fine, filename preservedSaves fine (you don't need the tool)Same origin, so the filename you specify applies as-is
Direct image URL on another site (CDN, etc.)Opens in a new tab, or the filename gets ignoredSaves with whatever filename you wantUses a Blob or a signed proxy to get around cross-origin restrictions
Image on a server with hotlink blockingThe image doesn't show at all, or shows up brokenCan't process it — the connection is refusedThe server rejects the request itself
Webpage address (article, post, etc.)You have to right-click every image one by oneNot supported (it isn't a file URL)It's a page address, not a file
URL with no extension where only the Content-Type says it's mediaMight save, but sometimes with a weird extensionSaves in the format detected from the Content-TypeFormat comes from the response header, not the extension
Large video file over 600MBThe browser may fail partway through the downloadNot possible — it's over the processing limitThe limit is 600MB per file
Low-resolution thumbnail URL instead of the originalSaves the thumbnail as-isSaves the thumbnail as-is (not the original)If the URL you enter is low-res, the result is low-res too
Private image that only shows up after you log inRequires a logged-in session to view or saveCan't process it (it isn't a public URL)Only public URLs are handled; anything requiring authentication is out of scope
Direct audio file URL (podcast, interview, etc.)Saving is unclear when the player has no save buttonSaves fine — it reads the format (e.g. mp3) and the file size instead of the durationSame criteria as images and video (extension, Content-Type)

As the table shows, the tool doesn't boost quality or create data that wasn't there. It recognizes the file the URL actually points to, exactly as it is, and removes only the restrictions the browser runs into at the final saving step. So if the original is low-resolution or large, it stays that way, and if the original is high quality, it saves at that quality.

Compare the second and third rows and you'll hit a spot that's easy to confuse. Both are "an image on another site," but the outcomes are completely different. The second row (a CDN image on another site) is a case where the server accepts the request. The third row (a server with hotlink blocking) is a case where the server rejects the request outright. You can't tell them apart just by looking, but if you open the URL directly in a new tab, you find out: either the image appears, or you get a connection-refused screen. If the image shows up, you're in the second row and the tool has something to work with. If you get a connection-refused screen, you're in the third row and nothing will get around it.

Where people get stuck

When people actually use the tool, the places they get stuck come down to about six cases. Symptoms, causes, and fixes, all in one place.

SymptomCauseFix
It says the connection was refusedThat server has hotlink blocking turned onThere's no way to download it this way. Try saving directly from the page on the original site
You enter a URL and it isn't recognizedYou entered a webpage address (an article, for example) instead of a file URLEnter the direct URL you get from "Copy image address" on the image, or use a bulk webpage image downloader tool instead
It says the file is too largeThe file is over 600MBOpen the original link in a new tab and save it with your browser's own save function
The URL is right but the format isn't recognizedIt has no extension, and the response's Content-Type doesn't confirm it's mediaOpen that URL directly in a new tab first and check that the file actually opens
It saved, but the quality is lower than you expectedThe URL you entered was a scaled-down thumbnail URL to begin withGo back to the original page, find the direct URL of the high-resolution image, and enter that
Pasting the URL does nothingSpaces, quotation marks, or other characters got mixed in before or after the URLCopy just the URL again, precisely, and paste it

If your problem isn't on this list, the fastest check is to paste the URL straight into a new browser tab and open it. If the file won't open properly there either, the problem is most likely with the URL itself or the original server, not the tool. On the other hand, if it opens fine in a new tab and only fails in the tool, it may be a temporary network issue at that moment, so it's worth waiting a bit and trying again.

"The connection was refused" from the first table and "the address isn't recognized" from the second table feel similar on screen, but the causes are different. Connection refused means you found the right URL and the server blocked the request. A recognition failure means the URL was never pointing to a file in the first place. To tell them apart, paste the URL into a new tab directly, the way described above. If an image or video fills the screen and opens on its own, the file URL is correct. If you get an error or refusal message inside that tab instead, it's hotlink blocking.

The tool only does one thing: it fetches image, video, and audio files that are already public on the internet, using a URL you already know. It isn't built to get around logins or pull in content that's set to private.

Downloading a file doesn't make its copyright go away. The copyright in an image or video you download still belongs to whoever made it, or to whoever the rights were transferred to. It's worth being clear that saving something doesn't transfer ownership.

So the rule is to limit your use to personal, lawful purposes — material you shot or made yourself, or material you have explicit permission from the copyright holder to use. Keeping a copy for reference, or using it personally within the scope you're allowed, is the safe zone.

The same copyright applies to an image you save as a screenshot, no differently than to the original. You've just re-recorded the pixels, but the expression of the original work is still there, so taking a screenshot doesn't turn protected material into something you can use freely.

In particular, re-uploading a downloaded file to another website, blog, or social media account, or redistributing it without crediting the creator, can count as copyright infringement. Even content with a free-use license, like on Wikimedia Commons, comes with conditions set by that license (attribution, share-alike, and so on), and you need to check them on the license page for that specific file. Using it commercially without checking the license terms, or distributing it with the credit stripped out, can cause problems regardless of the fact that the file was publicly available.

For example, not every photo on Wikimedia Commons is under the same terms. Some are public domain because their copyright has expired, so anyone can use them freely. Others carry specific conditions, like a Creative Commons license that says "you can reuse it if you credit the source." Checking the license section on the page where the photo is posted before you save it is the surest way to know where and how you can use it later. That check is a step you always need when working with public content, no matter where or how you downloaded the file.

On the flip side, you might be trying to recover a photo or video you uploaded yourself long ago after losing the original file. In that case, you're the copyright holder retrieving your own material, so you can use it freely without worrying about scope of use. Even then, it's safer to separately confirm that the material really is yours, or whether it's a joint work with someone else who holds rights to it.

How it differs from similar tools and platforms

Tools with a similar purpose fall into two broad groups. One installs as a browser extension and automatically scrapes images every time you open a page. The other takes a whole web page address, finds every image inside it, and shows them as a list.

This tool is neither. There's nothing to install, and it doesn't crawl a page to hunt for images. You give it the direct address of one file you've already seen and verified, and it focuses on checking that one file's format and size and turning it into something you can save.

Browser extensions often ask for permission to read the content of every page you visit, which makes some people uneasy about installing them at all. A tool that runs in the web page has no install or permission baggage: open it when you need it, close it when you're done.

One more thing worth pointing out: if right-click saving already works fine on a site, you don't need this tool at all. Its value sits in exactly one narrow gap — "the file is on a different site, so the browser ignores the filename you specified." It is not a universal fix for problems outside that gap, such as a request being rejected outright by hotlink blocking.

So the use cases split. If you want to gather many images from one page at once, a bulk downloader that takes a page address fits better. If you already know the exact address of the file you want and just need to save that one file cleanly under the filename you choose, this is simpler and faster.

It also differs from dedicated downloaders built for a specific platform (say, a particular social network or video site). Those tools analyze a platform's page structure to dig out the hidden video address inside it. This tool skips that analysis entirely and concentrates on taking a direct file address you already have and processing it as-is. Its support range is narrower, but its behavior is simple and its results are predictable.

In the end, the choice comes down to "what do I already know?" If you only know the page address and have no idea which images are on it or how many, you need a tool that scans the whole page and lists them. If you know the direct address of one file you've already seen, this tool that processes that single address is the shorter path. The two aren't competing replacements so much as different tools you pick from depending on the situation.

Summary

Here it is in short:

  • When right-click saving fails or the filename is ignored, it's usually because the image lives on another site (cross-origin), and that's normal browser security behavior.
  • A server with hotlink blocking rejects the request itself, so no method will let you pull a file from that server.
  • This tool only handles direct addresses of files. It does not support web page addresses such as articles or posts.
  • It first determines whether something is a file by its extension or Content-Type, then checks format and size up to 600MB so you can save it under the filename you want.
  • Copyright in a file you save stays with the original author, so always check the license and source before redistributing it.
Force-Download an Image or Video URL (with Filename)

Free · no signup · paste a file URL, check its format and size, then save it under the filename you want

Frequently asked questions

Can I paste a web page (article) address?

No. This tool only handles direct addresses of image, video, and audio files themselves. If you paste a web page address that mixes many elements, like an article or a post, it won't be recognized as a file. Right-click the image and choose "Copy image address" and you'll get the file's own address rather than the page address. If you want to collect all the images on a page at once, use a web page image bulk download tool instead.

I get "connection refused." Is there another way?

No. That means the server is hotlink blocking — it's blocking requests coming from other sites. The server itself is rejecting the request, so no method will let you pull the file from that server directly. Once you see that message, retrying won't change the result. The only thing left is to try saving from the original site's own screen.

How do I get an image's direct address?

Right-click the image and choose "Copy image address" (Chrome and Edge) or "Copy Image Link" (Firefox). Paste that address into a new tab. If only the image opens by itself, with no other elements, you've got the file's direct address. If the whole page opens along with it, you haven't found the file address yet.

Can I download video or audio files too?

Yes. Direct addresses of video and audio files are handled the same way as images. If the address ends in an extension or the response's Content-Type identifies it as media, the tool shows you the format and size first, then lets you save it under the filename you want. It does have to be an address pointing directly at a single file, not the address of a playback page.

What's the maximum file size?

Up to 600MB per file. Anything larger can't be handled by this tool, so open the original link in a new tab and save it with your browser's own feature. Most images and short videos fall within that limit, but for long, high-resolution video files it's worth estimating the size first.

Can I repost a file I saved on my blog or social media?

Be careful. Downloading a file doesn't make the copyright go away — it stays with the original author. Unless it's your own material or you have explicit permission to use it, reuploading or redistributing it may infringe copyright. Even for content under a free-use license, check the conditions it requires (such as attribution) first.

Why does right-click saving fail in the browser but this tool works?

For files on another site, the browser ignores the filename specified in the save link and just opens the file in a new tab. This tool fetches the file as binary data, or attaches a Content-Disposition header through a signed proxy, to get around that restriction and save it under the filename you want. That said, when the server rejects the request itself — as with hotlink blocking — this method can't get around it either.

This article was adapted from our Korean original. Read the Korean original

Related articles

  1. Favicon without favicon.ico: why tabs still show an iconBrowsers read the icon tags in a page's head first and only try /favicon.ico when nothing is declared. Here's how to find every icon a site hides.
  2. Why PM2.5 "Bad" Levels Differ by App: Korea vs. WHO vs. US AQIKorea's "Bad" PM2.5 grade starts at 36, WHO's 24-hour guideline is 15, and US AQI hits Moderate at 9.1. Why the tables differ — and PM10 where Korea is stricter.
  3. Link Preview Not Showing? Check og Tags and Clear the CacheLink preview blank or showing an old thumbnail? It's missing og tags or cached results. How to check the source and reset the cache.
  4. Keyword Density Explained: The Real Formula and Its LimitsLearn how keyword density is calculated, why the 1–2% rule isn't a real ranking factor, and how to use it to spot awkward repetition.
  5. Is It Better to Draw First or Last? Drawing Odds ExplainedSee why your draw order doesn't change your odds, and how to work out live odds from the cards already flipped. Simple permutation math, counted by hand.
  6. Name Compatibility Hash Test: The Formula Behind the ScoreSee how two names turn into a 40–100 compatibility score with a 31 polynomial hash, Unicode handling, and deterministic results you can repeat.

Related tools