Fable 5: FREE Tools Cut Tokens by 95%!

Today I'm going to show you some of the best ways to reduce the amount of tokens you actually use inside Fable 5 because Fable 5 is absolutely awesome, but if you're not careful you're going to absolutely rinse through the tokens you have available on your subscription. Bear in mind that it uses more tokens and you can only use 50% of your subscription on it and also by July the 7th everything you do doesn't go through the subscription. So now is the time to start thinking about how to reduce your token usage. The first time this came out I actually burned through the full cap in like hours and what we've got is a bunch of different ways and methods that you can use to quickly reduce the amount of tokens that you use.

And there's 10 different moves that you can use to do this. So first of all we're going to categorize this by root. So when you are using Fable 5 and you're building with it, this alone can cut your tokens by 50%. Now some of the strategies I'll show you actually claim to reduce tokens by 95%.

So we'll come on to those later and this only takes a few minutes to set up. So one of the first things that you could do is stop paying the architect to lay the bricks. And what I mean by that for example is like every task and every time you ask Claude to do something you can route it to different people or different models depending on what you want. So for example if it's like a really hard task like a plan, a judge, difficult bugs, then you can use Fable 5 the architect.

But actually for most of the general sort of automations, Opus 4.8 is pretty good and you can just use that as a builder that writes the actual code. And then you save your tokens because you assign your tasks to Opus for the hard stuff, Fable 5 for the really really difficult stuff. And then Haiku even can do like the small sub-agent stuff or you can use something even cheaper like GLM 5.2. So for example we've got GLM code here which is basically Claude code running GLM 5.2 so that if we have for example some harder tasks, some more work etc we can run it through there instead.

Now additionally move number two is that Fable can plan and Opus actually builds for you. So this way you get Fable 5 to build the blueprint but not for the actual typing, not for the actual coding. Because Opus uses less tokens you get more tokens available with it. So how does that work?

Well basically you would get Fable 5 to draw the blueprint with plan mode. Then you get the plan and this is like all the decisions forked and then Opus actually lays the bricks for you. When I say lay the bricks I mean like actually go off and do the work for you. So we actually did a playbook on this already.

But basically if you plan with Claude Fable 5 and you implement with a cheaper model based on Kilo Code's research you can produce the same service for 59% less than if you're using Claude Fable 5 for both phases. So the way this works is you have the planner, the genius which is Fable 5 then your agents can build with smaller cheaper models. So that alone can reduce your cost by 50% as well. Now you can also change the level of effort.

So you can change the effort dial based on what you need to do. Now you've got a bunch of choices here. So if you go into Claude you can change what it runs on. So it can go from like low to medium to high to extra, max and ultra code.

So you don't really need to go too high on this and it depends on the task. If it's just something simple like a blog post you can go pretty low with that. But if it's something super difficult like for example maybe you are coding out a really difficult bug or you're trying to fix something in your app then you'd go over here. But the main point here is that you can reduce the amount of effort you actually use based on the job which means you use less tokens when you're running through Fable 5.

Next up there's a really good tool called Headroom. This is free and open source. And the way this works is that this shrinks what Claude code sends automatically in every session. So this is a tool you can ask Claude just to install it for you.

It will reduce the amount of tokens you use. Now on their own tests and just try this yourself. I've tested it out myself as well. I've got another tutorial on this.

But basically Headroom can reduce your tokens by 60 to 95%. So it can save your tokens by 60 to 95% which is pretty wild. So imagine you're combining all of these different methods. The model routine.

The plan method. The effort dial. Headroom. This can literally reduce the amount of tokens you use by 2 to 3x if you have them all in combination.

Next up we also have something similar which is Ponytail. Now Ponytail is another open source plugin and it basically acts like a lazy senior dev. So it refuses to read code and code stuff it doesn't need. And so the cool thing about this is that it's very precise in terms of what it uses.

It's a skill that changes Claude's behavior. So it's great for reducing the amount of tokens you use. And even if you're just using Opus this is actually useful. It's only for coding whereas for example with Headroom it's quite different in terms of being a compressor.

But either way with Headroom and Ponytail these are both open source projects designed to quickly reduce the amount of code. And tokens you use. And you can get Ponytail on GitHub as well. So it's available right here.

It's free. And it reduces the amount of tokens used. In fact if you actually look at this stats here. 54% less code on average up to 94% less code.

Pretty wild. And it's just free and you can just literally copy and paste GitHub details. Give it to Claude and it's ready to go from there. Now also you can change your rules.

So if you actually change the Claude MD or the files that you use. And you reduce the amount of rules you have. Then it uses less tokens because of the rulebook. So you have lighter messages, lighter back and forth.

And so if you minimize the amount of rules that you give to Claude. And you just focus on the super crucial stuff. Then that reduces the amount of tokens you use as well. And that way it's lighter.

Claude isn't carrying a suitcase every time it tries to do stuff. Because it's reading through the Claude MD and all the rulebooks that you have. Another option is you can actually turn off web search by default. This is pretty interesting because.

You know one search might take up a lot of tokens. If it's automatically doing web search. So if you want to reduce the amount of tokens used. You can switch off web search.

Use less tokens. And then you can give it context when you need to. Or you can switch it back on for when you need to. So for a lot of coding tasks you don't actually need web search on.

Unless it's connecting to external APIs. And then you reduce the amount of tokens used. Which is awesome. And then when you need research.

You can fetch it once and just go with that. But switch it off by default. Now we also have forward slash compact. Most people don't use it.

But when you're using Claude. You can go into Claude like so. You can type forward slash compact. Like this.

It does take a few minutes to run that. But basically it frees up context. By summarizing the conversation so far. What does that mean?

Well basically it's cheaper on every turn. You get cleaner summaries. And you don't over compact. So if you wait for the auto compact.

You're using more tokens on each turn. With forward slash compact. You're squashing your conversation into a summary. Now most people let it fire automatically.

But that's when they get right to the ceiling. And they've already used too many tokens. It becomes inefficient. Waiting means like dozens of turns near full price.

Whereas for example you can compact at about 60%. And keep every turn cheaper. And the summary sharper. Now what you can also do after that.

Is you can use the post compact brief. So after every compact. You can auto inject like a five line brief. Of what was going on.

Give it the summary. And then the agent still knows the tasks. The files, the rules. But you don't need to re-explain.

And you don't lose too much context. So a lot of people when they compact. They're worried about losing context. But if you have that five line brief.

The add in. Automatically it has more context straight away. So it doesn't just forget everything. That you've been working on.

And then finally we have the two hour clear. This is actually something I learned from. An absolute legend Johan. So credit to him.

Run Sindra. Gave out some great tips on reducing tokens. Especially with Fable 5. So you can do a two hour clear.

It's free to use. And the thing to note here. Is like if you're wondering what it is. Well basically long sessions.

Collect a lot of junk right. Old tool output. Abandoned directions. And so this costs more tokens.

Each time you use it. And the more the conversation goes on for. The more tokens it's going to use up. If you use forward slash clear.

It wipes the room for free. And then you can just give it. Like a three line handoff note. Carrying over the things that actually matter.

What you're doing. What you try. What's next. And then every two hours.

You can write that handoff note. And then do forward slash clear. And paste in the note. So that's basically it.

Now you might say okay. Does this actually reduce tokens. By 95%. I would say if you combined.

All of these different methods. You would save even more tokens. But that's a great way to just keep using. Fable 5.

Reduce the amount of tokens you use. You can keep an eye on it over here. In terms of how much usage you've gone through. And if you want to be efficient.

You can just use all 10 of these methods. Now. Minimizing the amount of tokens. You use is one piece of this.

The agent operating system. Is a machine that you can plug into. So the token. Minimize a playbook.

Keeps the best models. Like Fable 5. Affordable and usable. The boardroom.

Is where the whole agent operating system. That I personally use lives. And we have token efficiency playbooks. We have more details on headroom.

We show you how to have every agent in one place. This is our agent operating system. The great thing about this is. When you're running so many CLIs.

If you're using these sort of. Token minimization playbooks. You can reduce the amount of tokens you use. But have a system.

That's even more powerful right. We're building out new features all the time. Into the agent operating system. Especially during this window.

Because Fable 5. Will go off the subscription. On the 7th of July. And then you can see for example.

Like we've built out. Hermes Apollo. Which is a voice activated agent for Hermes. We have Hermes Oracle.

We even have Hermes Astros. That can do competitive research. And search the web. And get us all the latest results.

So that we can get ideas for our content right. Super powerful stuff. So if you want to get this full setup. Check out the agent operating system.

Inside the AI Profit Boardroom. Which is my community. To help you save time. Grow and learn.

And scale with AI automation. Inside the community. You can ask questions. Get help and support in real time.

Inside the classroom. You can get access to all of my best trainings. I personally answer these questions. By the way.

Every single day. With a video tutorial. And then also inside the classroom. You can get access to all of my best trainings.

And if you want the agent OS. It is right here. Waiting for you. We update it daily.

With a video tutorial. A full guide. And a zip file to quickly install it. And then we show you.

All the best stuff. Based on what's just dropped. So feel free to get that. Also inside the calendar.

You get four weekly coaching calls. So you can share your screen. Ask questions in real time. Etc.

And inside the map. You can meet people. In your local area. Who are building with AI agents like you.

So feel free to get that. Link in the comments description. Or just go to the AI Profit Boardroom. Thanks for watching.