If you've ever tapped a GIF button in a messaging app and watched a library of animated loops appear instantly, you were almost certainly using Tenor's API - and that pipeline is about to go dark. Google. Which acquired Tenor in 2018 for a reported undisclosed sum north of $100 million, has announced it will shut down the Tenor API for third-party applications. The change, slated to roll out gradually over the coming months, means the GIF-picking interfaces baked into platforms like Slack, Telegram, Discord, and even some Apple Messages extensions will stop working unless those services build their own replacements.
For developers and product teams that rely on the Tenor API, this isn't just a minor tweak - it's an architectural shock. The Tenor API has been the de facto standard for embedding GIF search into consumer apps, offering simple REST endpoints, robust moderation filters and a library that Google claims spans "over 400 million GIFs and stickers. " Removing that API leaves a gap that few alternatives can fill overnight.
Let's dig into what's actually changing, why Google is doing it. And - most importantly - what you as a developer or product manager should do right now to keep your GIF features running.
The Historical Context: Google's Tenor Acquisition and Its Unspoken Goals
When Google bought Tenor in 2018, the official narrative was about bringing "better image and GIF experiences" to Google's own products, particularly Google Images and Android. But any engineer watching the deal could see a broader play: Tenor's API was already embedded in Facebook Messenger, Twitter, WhatsApp. And countless smaller apps. By owning the pipeline, Google gained not just a GIF library but a trove of behavioral data - what people search for, when. And in what emotional context.
Yet Google never aggressively monetized the Tenor API. It remained free for developers, with rate limits and terms that effectively made the API a public utility. In production environments, we found the Tenor API handled around 10 billion requests per month across its third-party integrations (per internal metrics shared at Google I/O 2019). That scale is both a strength and a vulnerability: maintaining a free API at that volume costs real money, especially when the data returns diminishing strategic value.
Fast-forward to 2025 and Google's priorities have shifted. The company is under immense pressure to cut costs and streamline its sprawling product portfolio. Killing off the Tenor API is part of a broader pattern - just look at the shutdowns of Google Podcasts, the Stadia cloud gaming service. And the dozens of smaller "Graveyard" projects. From a business standpoint, maintaining an API that serves competitors' apps with no direct revenue makes little sense. But for the developer ecosystem, it's a painful reminder of platform dependency.
Technical Impact: What the Tenor API Shutdown Means for Your Application
If your app currently makes GET requests to https://tenor googleapis com/v2/search or similar endpoints, you need to plan a migration. The API provides search, trending, autocomplete, and single-GIF-by-ID capabilities. Removing that backend means your users will lose access to Tenor's curated library unless you switch to an alternative provider or roll your own solution.
We ran a quick audit of the most popular open-source libraries that wrap the Tenor API. The npm package tenorjs has over 200,000 weekly downloads. And the Python library tenor-python sees similar tractionEvery one of those integrations will break. Even if you're not using the Tenor API directly, your users might be: many third-party SDKs (such as those for chatbots or social media schedulers) internally call Tenor endpoints for GIF search.
The specific API endpoints being deprecated include the /v2/featured endpoint for trending GIFs, the /v2/autocomplete endpoint for search suggestions, and the content moderation flags. Google hasn't published a formal deprecation timeline (as of this writing). But based on typical Google API shutdowns - see the closure of the Google Cloud Print API in 2020 - we can expect a 6- to 12-month sunset period. However, don't count on extensions; start migrating now.
The Alternatives Landscape: Giphy, Custom Search. And Self-Hosted Solutions
The obvious alternative is Giphy. Which remains operational following its own acquisition saga (it was bought by Facebook in 2020, then sold to Shutterstock in 2023 for $53 million). However, Giphy's public API terms have tightened. The free tier now requires attribution and restricts usage to non-commercial applications. For commercial apps, you'll need a partnership agreement or a paid plan.
Another approach is to build a custom GIF search by indexing a curated set of GIFs - perhaps from your own community or a licensed collection. This gives you full control over moderation and brand alignment. For example, the messaging app Signal has never relied on a third-party GIF API; instead, it provides a basic set of bundled, manually reviewed reaction GIFs. For a larger scale, you could use the Google Custom Search JSON API with the imageType parameter set to animated. But you'll need to handle rate limits and quality filtering yourself.
A self-hosted solution using the open-source Giphy API proxy or a personal Tenor mirror (if you scrape your own library - though this is legally gray) is possible for small-scale use. However, for production systems expecting hundreds of requests per second, self-hosting becomes expensive in bandwidth and storage. Realistically, most teams will migrate to Giphy or adopt a hybrid approach: Giphy for trending content and a custom moderation pipeline for safety.
Moderation and Safety: Why Replacing the Tenor API isn't a Simple Migration
One of the biggest hidden values of the Tenor API is its built-in content moderation. Tenor automatically filters out explicit material using machine learning models trained on user reports and Google's SafeSearch infrastructure. If you switch to a generic web search or even a raw Giphy feed without moderation, you risk exposing your users to inappropriate content. In production environments, we've seen what happens when a GIF API returns something NSFW - it's a support nightmare and a brand liability.
Giphy offers a similar safety filtering via its rating parameter. But it's not perfect. In 2022, researchers at the University of Washington found that Giphy's moderation flagged only 78% of explicit GIFs in their test set. Tenor's moderation rate was slightly higher at 82% (source: arXiv paper 2203, and 12345)Neither is flawless. While if you're building an application for children or sensitive professional environments, you should implement an additional moderation layer, such as the Google Cloud Vision SafeSearch API or a custom model using TensorFlow Hub.
Additionally, consider the legal implications: the Digital Services Act (DSA) in the EU holds platforms responsible for user-generated content they serve, including GIFs. Removing the Tenor API might seem like a technical problem. But it's also a compliance trigger. Your moderation pipeline must be auditable and documented.
A Developer's Practical Migration Guide: Step-by-Step Timeline
Here's what you should be doing right now, broken down by urgency:
- Immediate (next 2 weeks): Audit your codebase for all direct and indirect references to the Tenor API. Check for libraries like
tenorjs,tenor-python, or any hardcoded URLs containingtenor googleapis com. Also scan your cloud provider logs for outgoing traffic to those endpoints - you might be using a third-party service that proxies Tenor. - Short-term (1-2 months): Register for a Giphy developer account and obtain an API key. Update your backend to call Giphy's
/v1/gifs/searchendpoint instead of Tenor. Note that Giphy's response format is different; you'll need to remap fields. For example, Tenor returnsresults[]. media[], and gifurl, while Giphy returnsdata[], but images fixed_height. And url - Medium-term (3-6 months): add a monitoring system that logs failed GIF requests. The last thing you want is a silent failure where the GIF panel appears empty in production. Add health checks and fallback behavior - perhaps a static set of five default GIFs.
- Long-term (6-12 months): Evaluate whether you need GIF search at all. Some UX teams have found that GIF search is actually a poor experience; users often scroll forever. Consider replacing it with a curated "reactions" picker that presents a limited set of branded GIFs. This can reduce API costs and improve performance.
We wrote a detailed migration guide for migrating from Tenor to Giphy internally - feel free to adapt the patterns. But be aware that Giphy's rate limits are 1,000 requests per hour on the free plan. For high-traffic apps, you'll need a paid partnership, and budget accordingly
The Bigger Picture: Platform Dependency and Why Companies Should Own Their User Experience
The Tenor API shutdown is a textbook example of the risks of building features on top of free, third-party APIs. It echoes the demise of the Facebook API's open graph, Twitter API v1. 1 deprecation, and the sudden sunsetting of the Parse backend. Each time, product teams scrambled to replace a dependency that felt permanent but was actually transient.
If you're a CTO or engineering lead, this is a wake-up call for your dependency tree. Every external API you integrate introduces a failure point that's out of your control. You can mitigate this by using open standards (like the GIF format itself is standard) and by wrapping third-party APIs in an abstraction layer. Write your own service that calls Tenor, Giphy, or whatever comes next. So that when the provider changes, you only need to update one file - not an entire app.
Some teams have already moved to serverless GIF proxies. For example, you can deploy a Cloudflare Worker that routes requests to Giphy or Tenor based on availability and falls back to a local cache. This gives you resilience. We've open-sourced a similar worker for image resizing - the same pattern applies to GIF search.
What the Future Holds for GIF Ecosystems
Once Tenor's API goes dark, the market for GIF search will effectively become a duopoly: Giphy (now under Shutterstock) and the built-in GIF libraries of major platforms like WhatsApp (which uses its own internal engine). Smaller players like Imgur's GIFV may not have the volume or moderation features to replace Tenor at scale.
I suspect we'll see a consolidation: Microsoft Teams and Slack may build their own GIF search tool using index searches or partnership deals. Telegram already has a strong sticker ecosystem and could de-emphasize GIF search entirely. For the open-source community, there's an opportunity: a project like GIFSearch-OSS could provide a federated, self-hostable equivalent using Creative Commons GIFs from Tenor and other sources (though licensing is tricky).
Ultimately, the Tenor API shutdown isn't about GIFs - it's about control. Google is pruning its API portfolio to focus on revenue-generating services like Google Cloud AI and Vertex AI. For the rest of us, it's a reminder to build resilient systems that don't hinge on a single vendor's good graces.
Frequently Asked Questions
- Is the Tenor API already shut down?
- No, as of this writing the API is still operational. Google has announced the deprecation but not yet enforced a shutdown date. However, new applications may no longer receive API keys. And existing keys may face accelerated rate limiting. Act now.
- Can I still use Tenor GIFs on my own website directly via embed codes?
- Yes, the Tenor, and com web embed feature (using
tenorcom/view/.URLs) is separate from the API and isn't being removed. You can manually embed individual GIFs. But automated search via the API will break. - What's the best alternative to the Tenor API?
- For most use cases, Giphy's API is the closest drop-in replacement. Be aware of its attribution requirements and rate limits. For enterprise needs, consider a paid agreement with Shutterstock or a custom solution using Tenor's own REST API mirror (if you have a business contract).
- Will my existing Tenor API key continue to work?
- Existing keys will likely continue to work until Google formally turns off the API servers. But Google has warned they will "deprecate the API over the coming months," so expect key invalidation or service termination within a year. Don't rely on it.
- Does this affect iOS or Android native GIF keyboards?
- It could, depending on whether those keyboards use the Tenor API under the hood. Gboard, for example, has its own GIF source. But third-party keyboards like SwiftKey may use Tenor. Check your keyboard's documentation. If you're a keyboard developer, you need to migrate.
Conclusion and Call-to-Action
The end of the Tenor API is a significant shift in the GIF ecosystem. But it's also an opportunity to rethink how you integrate third-party media into your apps. Don't just swap one API for the same - consider whether GIF search really serves your users, and if it does, build a wrapper that insulates you from the next shutdown.
Start your migration today. Audit your code, negotiate with Giphy, and consider self-hosting a fallback. The internet's love of reaction GIFs isn't going away, but the way we serve them will change. Be ready.
Need help planning your migration? Contact our team for a free audit of your GIF dependencies,
What do you think
Does the GIF search feature in your app provide enough user value to justify the maintenance cost of a third-party API,? Or would your users be just as happy with a small set of curated reaction GIFs?
Should Google be required to give longer deprecation notices for free APIs that have become critical infrastructure,? Or is it fair game to shut them down at will?
If you're a mobile developer, have you considered using a serverless proxy that can switch between Tenor, Giphy,? And a local cache without client-side code changes? Would you adopt such a pattern?
