Table of Contents
Key takeaways
- Emissions are the newly minted TAO the protocol hands out every block, and subnets compete over how that daily flow gets divided.
- Under Dynamic TAO, the market decides the split, because capital staked into a subnet lifts its alpha token and pulls a larger share of emissions toward it.
- The allocation formula has changed more than once, moving from price to staking flow and back to a price-based model that now favors the strongest subnets.
- A subnet also has to defend its slot, since the network caps active subnets and a stronger newcomer can force out the weakest one.
- A high emission share signals where capital is flowing inside Bittensor, not proof that anyone outside the network is paying for the work.
Every block on Bittensor mints new TAO, and none of it is promised to any subnet in advance. That single design choice turns the network into an ongoing contest, because more than a hundred subnets are all reaching for part of the same daily emission, and the amount they receive is decided by the market rather than by a fixed schedule. If you stake into subnets, build one, or invest around the ecosystem, knowing how that money gets divided is what separates chasing a yield number from understanding what the network is actually paying for.
How Emissions Get Allocated
Bittensor runs on a steady issuance clock, producing a block about every twelve seconds and paying rewards out each tempo, a 360-block cycle. After the first halving on December 14, 2025, daily issuance dropped from roughly 7,200 TAO to about 3,600, which means the prize subnets are fighting over is smaller now than it was for most of the network's life.
The harder question is who gets that new TAO. Before Dynamic TAO, a small set of root validators set the weights that steered emissions, and that approach held up while the network was young and had only a handful of subnets. dTAO replaced that committee with a market. When a TAO holder stakes into a subnet, the capital enters that subnet's automated market maker pool and pushes up the price of its alpha token against TAO, and the protocol then sends more of the daily emission toward the subnets where capital is collecting. In plain terms, capital votes and emissions follow the vote.
What Wins A Subnet More Emissions
A subnet earns a larger share by attracting staked capital and then keeping it, so the whole game runs on demand for the subnet's alpha token. Rewards inside a subnet are paid in that alpha rather than in TAO directly, and they divide on a fixed split of about 41% to miners, 41% to validators and their stakers, and 18% to the subnet owner. Because everyone is paid in alpha, everyone has a reason to care whether the token holds value, and that shared incentive is what a subnet has to protect.
Holding capital is harder than winning it once. Emissions constantly create fresh alpha supply, and if miners, validators, or the owner sell that alpha faster than new stakers buy in, the token can slide even while the subnet is still collecting a healthy emission share. The subnets that stay near the top tend to ship a product people actually use, communicate honestly with the stakers funding them, and keep their pool liquid enough that capital can move in and out without a violent price swing. A good story can attract capital for a while, but retention is what compounds into a durable emission share.
The Fight For A Slot
Winning a bigger share is only half the contest, since a subnet also has to keep its seat at the table. Bittensor caps the number of active subnets, currently 128 slots with an expansion to 256 planned, and when the slots are full, registering a new subnet can force out the weakest existing one, measured by the market value of its token. That makes the design feel a lot like a sports league with promotion and relegation, where a place has to be earned and then defended rather than owned outright. A subnet drifting near the bottom earns less and risks being replaced by a newcomer with a stronger token and a better story, so the pressure to keep performing never really lets up.
How The Allocation Rules Keep Tightening The Race
The signal the protocol uses to split emissions has not stayed still. The original dTAO design leaned on price, so a subnet with a higher moving-average alpha price attracted a larger share of the daily TAO. Around November 2025 the network moved to a flow-based model sometimes called Taoflow, which allocated by net TAO staking inflows instead of price. Then in late June 2026 it largely returned to a price-based approach built on EMA prices, layered with extra factors like miner-burn penalties, active-mining gates, and a root proportion.
As of August 2026 the system is once again primarily price-oriented, now paired with mechanisms that concentrate emissions toward the strongest subnets. The label on the formula has moved back and forth, but the underlying logic has held throughout, because the network keeps trying to route capital toward the subnets participants value most and away from the ones they do not.
That narrowing was deliberate. The December 2025 halving had already cut new supply in half, and across May and June the network shipped a run of emission changes that blocked payouts to inactive or exploitative subnets and pulled the number of high-emission subnets down toward the range of 30 in later analyses. Further gating in July pushed emissions even harder toward the top performers, which cut down the indiscriminate selling that used to come from weaker subnets dumping alpha.
What A High Emission Share Really Means
The most important thing to understand about this whole contest is what winning does and does not prove. A large emission share shows that capital is flowing toward a subnet inside Bittensor, and that is genuinely useful information, but it is not evidence that anyone outside the network is paying for the work. The gap can be wide. Analysts have pointed out that the largest subnet drew around 52 million dollars in annual emissions while producing far less than that in external revenue, a reminder that emissions currently subsidize a lot of activity that has not yet shown it can pay for itself.
The subnets worth watching are the ones turning their share into products with demand beyond the staking loop, because emissions can fund a real business on the way up and can just as easily mask one that has no customers. Competing for emissions keeps the network honest about where capital wants to go, and the open question that will define the next stretch is whether the work being funded grows into revenue that justifies the spend.
