Top 4 Bittensor TAO Subnets

Up 10% in the last month, Bittensor is helping lead the market back to normalcy. It’s one of our favorite AI projects, and favorites overall.

But you don’t have to invest just in the $TAO token.  There are 126 subnets to choose from. All of them do something different in AI or AI infrastructure.  You can allocate your TAO to any one of them for even greater growth opportunities. So today, we look at 4 of our favorite Bittensor Subnets.

1) Chutes

We start today with Subnet #64, Chutes. It is arguably the largest of all the subnets. It’s also in one of our favorite areas of AI. Computing power. Chutes offers serverless compute for AI, at scale. So it grows as you do. AI companies and those training models love this as they are only paying for the computing power they use. That’s a win for them.

And the decentralized AI infrastructure is a win for Chutes and for us. They already support common models like:

• OpenAI.
• Deepseek.
• Qwen.
• And MSFT AI (that’s Co-Pilot).

There are many small, emerging companies taking one of these low-cost less GPU intensive models like Deepseek and training it. Then they can use it for specific industry tasks like sales, marketing or supply chain. There are currently 2100 nodes around the world giving their extra compute to the network and earning $$ for it.

You can even pick which GPUs you want to use all from different NVIDIA chip options. The Deepseek V3 is the most popular AI model on the network right now, dominating token use. So many are testing and training it to launch agents and products of their own.

Those who contribute their compute get paid in $TAO, since that’s the Bittensor network token. But from there, you can decide to allocate your $TAO to Chutes. You do that here at TaoStats. You can see that Chutes is the #1 subnet by value with a value of 356,000 TAO or just under $100 million dollars.

So that means all the other subnets, including the last 3 on today’s list, are smaller and firmly in the small-cap market value category.

2) Lium

Next, we have lium, which is Subnet #51. Like Chutes, lium is a GPU marketplace. And like Chutes, much of the opportunity is the same.

You can pick which NVIDIA GPUs you want to process your AI data and even rent just by the hour if you want to run a short term test. That type of flexibility is invaluable to a newer project.

Like Chutes, those that rent out their compute, since it is a p2p network, get paid in $TAO. It has one important difference from Chutes if you are bullish on GPU marketplaces and how they apply to AI and machine learning projects. Value.

Lium offers many of the same things and it’s half the market value of Chutes. Instead of $100 million, lium’s value is $51 million.

3) Ridges

Ridges is Subnet #62. Its specialty is training AI agents. This one is agent specific. The training here is  not of an AI model like Claude to do what you want it to. It’s specific to training autonomous AI agents.

In the last week, 32 new agents have come online for testing and evaluation by the Ridges team. By the way, there is an 18k prize pool that agents can win for top performance. Ridges also has outside investment from one of Bittensor’s biggest DAT or digital asset treasury companies. The Stillmark Fund invested in Ridges just last month.

Ridges is THE leader in Bittensor for agent training. Its current value is just over $45 million, placing it 3rd in value among all subnets. Complex topics like agent training are one of the reasons why people avoid Bittensor and why they avoid Ridges. Their complexity is your opportunity.

4) Targon

Our last subnet today is Targon. Targon is Subnet #4. So it’s been around a while. Targon is in the compute marketplace too. But there are differences here and all 3 of the marketplaces in today’s list can be winners.  You can rent computing power or even your own entire server cheaply. There are 2 major differences we can hit on right away.

First, they have their own network the TVM. That’s Targon Virtual Machine. In it, you can build and test privately. That’s without others seeing or knowing what you are doing or with how many GPUs. This is great for stealth projects, side projects, and those not ready to deploy to the public yet.

And that leads to the 2nd big difference. Privacy. Here you can do Confidential AI. This is something you CANNOT do elsewhere. A reminder, confidential compute means the computations can be done privately or done off-chain bringing the results back on-chain. Either way, it’s much more private than seeing everyone’s transaction and computation activity.

This is a union of 2 of our industry’s hottest sectors. AI and Privacy. And the fact that it’s only valued at $45 million looks like a steal. Osmosis, the Cosmos DEX, is running in maintenance mode now and is basically a dead project.

And it’s valued the same as Targon right now. That seems crazy. Targon should 5-10x here in an AI bull market. We would be more surprised if that doesn’t happen given its potential and how well they are executing.

Disclaimer

The information provided by Altcoin Buzz is not financial advice. It is intended solely for educational, entertainment, and informational purposes. Any opinions or strategies shared are those of the writer/reviewers, and their risk tolerance may differ from yours. We are not liable for any losses you may incur from investments related to the information given. Bitcoin and other cryptocurrencies are high-risk assets; therefore, conduct thorough due diligence. Copyright Altcoin Buzz Pte Ltd.