New Tiny Fish AI Just Changed Everything! 🤯

Tinyfish AI just made web search and page fetching completely free, no credit card, no trial, just sign up and go. And I want to break down exactly what that means because this is one of those updates that looks small on the surface but changes a lot about how AI actually works in the real world. Here's the thing about AI agents right now, they're smart, really smart, but most of them have a massive blind spot, they can't see the live web. They're working off training data that has a cut off date.

They can reason, they can write, they can plan, but the second you ask them to check what's happening on a website right now, they hit a wall. Tinyfish is the fix for that. It gives AI agents four things, search, fetch, browser, and agent. Search pulls structured results from the live web in under 500 milliseconds, that's faster than most humans even register a page loading.

Fetch takes any URL and converts it into clean, readable text that an AI can actually process. Browser gives the agent a real Chrome session to navigate sites, fill out forms, log in, interact. And agent ties it all together into multi-step workflows that run on their own. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency.

Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. Julian Goldie reads every comment, so make sure you comment below. Up until now, search and fetch both cost credits. That's been the barrier, not a huge one, but enough that a lot of people building small tools, testing ideas, or just getting started would hesitate.

That barrier is gone. Starting now, Tinyfish web search and fetch are free, free for every developer, free for every agent, free across every surface you build on, the REST API, the MCP server, Python and TypeScript SDKs, and integrations like N8n, Lanchain, Crew AI, and more. One signup, one API key, no credit card. Tinyfish, the free tier, includes five queries per minute for search and 25 URLs per minute for fetch.

For anyone building a tool, testing a workflow, or automating something for a client, that's more than enough to get serious work done. Tinyfish. So let me explain why search and fetch specifically are the ones that matter here. Think about how you use the internet.

When you want to know something, you search, then you click a link and read the page. That's it. That's the entire loop. Search finds it.

Fetch reads it. Those two steps are the foundation of everything. Tinyfish's position is that these are the two most fundamental things agents and even humans do on the web. And fundamental things shouldn't cost you anything to start with.

Tinyfish, that's a clear statement of intent. They're not giving away a feature. They're removing a toll on the most basic web behavior. Now here's where it gets genuinely useful for people watching this.

Let's talk about what you can actually do with this. Say you're running a business and you want to know every time a competitor updates their pricing page, you could set up a Tinyfish Fetch workflow that checks our page daily, pulls a clean text, and flags any changes. That would have cost credits before. Now it's free.

Or let's say you want to build a tool for the AI profit boardroom, something that pulls the latest AI news every morning, summarizes it, and sends a digest to members. Tinyfish Search handles the live web lookup. Tinyfish Fetch pulls the actual article content. The whole pipeline runs automatically.

No manual checking. No outdated info. And here's the part most people don't think about. Search engines index less than 5% of online content.

The remaining 90 plus percent exists behind authentication walls, interactive workflows, and dynamic interfaces. Tinyfish is one of the only tools built to access that part of the web at scale. Tinyfish, look, if you want to actually build workflows like this, not just watch someone talk about them, come check out the AI profit boardroom. We've built a full 30-day roadmap specifically around tools like Tinyfish.

Inside, we run four weekly AI automation coaching calls where members are building real pipelines, competitor monitoring, lead research, automated content using exactly this kind of infrastructure. There are 2,800 business owners in there right now. A lot of them already using Tinyfish style web automation to save hours every week. We've got step-by-step tutorials, a prompt library, and a member map so you can connect with people near you who are building the same stuff.

Link in the description or the comments. Come join us. Now let me go a bit deeper on the technical side because there's something here that most people covering this are missing. When a standard AI agent fetches a web page, it pulls the entire thing, all the code, all the ads, all the navigation, all the hidden scripts, and dumps it into the model's context window.

This is a problem. When an AI agent uses a standard web fetch tool, it typically pulls thousands of tokens of navigation elements, ad code, and boilerplate markup, and puts all of it into the model's context window before reaching the actual content. By pointer, Tinyfish Fetch doesn't work that way. It renders a page in a real browser, strips everything irrelevant, and sends back only the clean content.

Their benchmarks show CLI-based operations using approximately 100 tokens per operation versus roughly 1,500 tokens when routing the same workflow over MCP, an 87% reduction per operation. By pointer, there's not just a performance improvement. There's a cost improvement. Every token costs compute.

When you're running thousands of fetch operations, the difference between 100 tokens and 1,500 tokens adds up extremely fast. And Tinyfish built all of this in-house. For example, Browserbase uses Exa to power its search endpoint, meaning that layer is not proprietary. Firecrawl offers search crawl and an agent endpoint, but the agent endpoint has reliability issues on many tasks.

Tinyfish owns every layer of the stack. That matters when something breaks. When your agent fails halfway through a multi-step workflow, and every layer is built by the same team, they can see exactly what happened and fix it. When you're stitching together three different providers, that visibility disappears.

MarkTech posts, when Tinyfish's agent succeeds or fails using its own search and fetch, the company gets end-to-end signal at every step, what was searched, what was fetched, and exactly where failures occurred. That's a meaningful reliability advantage. By pointer, let's talk about the search performance specifically because the numbers here are real. Tinyfish web search returns structured JSON results using a custom Chromium engine with a P50 latency of approximately 488ms.

Competitors in this space average over 2,800ms for the same operation. That's roughly 6x faster. When you're building an agent that needs to search dozens of queries in a workflow, that speed compounds. By pointer.

Now let's zoom out and talk about what this free tier actually signals about where Tinyfish is going. Companies don't make core infrastructure free without feeling generous. They do it because they want to become the default. Google made search free because they wanted every person on the planet using it.

Obvious made its free tier generous because they wanted every startup learning their stack. Tinyfish making search and fetch free is the same play. They want every AI agent, every framework, every developer to default to their infrastructure. The platform works with Claude, Cursor, Codex, Chat, GPT, Desktop, and any MCP compatible client.

That's every major AI coding tool. Drop one URL into your config, one key, done. Tinyfish, who is this actually for? If you're building AI automations for clients, this is your infrastructure.

You can now offer live web research as part of any automation package without paying for every search query. That changes your cost structure entirely. If you're running a business and you want to automate research, competitor tracking, news monitoring, lead enrichment, Tinyfish gives you the tools to build that without stitching together 5 different services. And when you're ready to go deeper, when you want to build full automation workflows that bring in leads, save you time, and actually run without you, come inside the AI Profit Boardroom.

We have a specific 30-day roadmap built around web automation tools like Tinyfish walking you through exactly how to set up your first live web AI pipeline, how to use it for lead research, how to build workflows that monitor the web for your business while you sleep. 4 coaching calls every week where you can ask questions live about your specific setup. Over 2,800 members, a lot of them building exactly this kind of web automation right now. Daily tutorials covering the tools that actually matter as they ship, a prompt library and a member map so you can find and connect with other AI automation builders near you.

Link in the comments and description or go to aprofitboardroom.com And if you want the full breakdown, every tool, every workflow, every use case from this video, the AI Success Lab is the free version. 67,000 members in there. All the video notes, links and resources from today's episode posted there for free. Link is in the comments and description too.

Come join us.