Stake Weight

How stake weight describes validator influence inside Bittensor subnet consensus.

Stake weight is the subnet-level stake measure used to describe validator influence in Bittensor consensus. It is the computed stake value tied to consensus power and emissions inside a subnet (Glossary: Stake Weight, Yuma Consensus).

The term links staking vocabulary to consensus evaluation. It answers how much influence a submitted evaluation carries after stake context is applied, while validator weights answer what was signaled about miner work.

Consensus Influence

In Yuma Consensus, validators submit evaluation signals and the consensus process weights those signals before producing subnet outcomes (Yuma Consensus, Glossary: Validator Weights).

This measure is the influence side of that path. It does not replace the evaluation signal; it scales how strongly the submitted signal participates in the subnet consensus process (Glossary, Yuma Consensus).

This is why the term matters in plain English: two validators can submit signals into the same subnet process, but their stake context affects how those signals count when consensus is computed.

Components and Neighboring Terms

The measure combines alpha stake, TAO stake, and TAO weight in subnet context (Glossary, Glossary: TAO Weight).

The staking-total term and the validation-eligibility term have different jobs. One describes backing, while the other concerns eligibility in a subnet (staking total, validation eligibility).

Together, these terms give readers a compact map: backing, eligibility, TAO contribution, and the consensus measure are related but separate concepts.

Signal Path

Validator weights are the evaluation values. Weight vectors organize one validator’s submissions, and weight matrices collect those vectors for the subnet-level consensus process (Glossary: Weight Vector, Glossary: Weight Matrix).

The measure sits beside that input structure as the influence value applied around validator submissions. Later result terms such as rank, trust, and validator trust describe what happens after the signals are filtered and interpreted (Glossary: Rank, Glossary: Trust, Glossary: Validator Trust).

Subnet and Environment Scope

The measure belongs inside the subnet context that uses it. Validators evaluate miner work according to subnet-specific standards, and Yuma Consensus interprets those signals inside that setting (Understanding Subnets, Yuma Consensus).

Bittensor separates mainnet, testnet, and localnet environments. Examples from one environment should not be treated as evidence for another because subnet tasks, validators, and stake conditions can differ (Bittensor Networks, Introduction to Bittensor: Subnet development).

Localnet examples are isolated development examples. Testnet examples are shared non-production examples. Mainnet interpretation belongs to the mainnet subnet and consensus context being discussed.

Development Stage Context

The Introduction to Bittensor describes subnet development as moving from localnet to testnet and then mainnet. For stake weight on a subnet such as netuid 1, that sequence changes how readers should interpret validator influence readings and metagraph stake fields.

In localnet, stake-weight examples can be exercised in an isolated environment. Local validator influence values reflect local chain state rather than production subnet consensus.

On testnet, stake weight can be observed in a shared, non-production network. Testnet stake-weight readings on a selected netuid are separate from mainnet metagraph state (Subnet Metagraph).

On mainnet, stake weight is the live subnet-level validator influence measure used in production consensus context on the connected Bittensor network (Glossary: Stake Weight).

The Bittensor Networks reference separates mainnet, testnet, and localnet. A stake-weight example from one environment should not be read as representing production validator influence on another network.

Relationship to Yuma Consensus

Stake Weight and Yuma Consensus describe related parts of Bittensor’s incentive system. Yuma Consensus is the on-chain process that aggregates validator weight signals within a subnet into miner incentives and validator dividends, applying consensus clipping, bonding, and emission calculation (Yuma Consensus).

For readers, stake weight names a specific part of that incentive picture, while Yuma Consensus names the consensus process that turns validator weights into the resulting incentives and dividends.

Reader Boundary

Stake weight should not be read as a payout amount, a validator trust score by itself, or proof that a miner received a specific reward. It names validator influence inside subnet consensus—the computed stake value that scales how strongly a submitted evaluation counts (Glossary: Stake Weight, Yuma Consensus).

Validator weights say what a validator signaled about miner work. Stake weight explains how strongly that signal participates when Yuma Consensus filters and combines validator submissions (Glossary: Validator Weights).

Stake Weight Differs From Effective Stake

Effective stake and validator permit vocabulary describe backing and eligibility. Stake weight describes consensus influence after those staking contexts are already in view (Glossary: Effective Stake, Glossary: Validator Permit).

Keeping those terms separate avoids reading a permit flag or staking total as the same thing as a consensus influence measure.

Stake Weight Sits Between Weight Inputs and Rank Outputs

Validator weights, weight vectors, and weight matrices describe the evaluation inputs Yuma Consensus receives. Rank, trust, and validator trust describe filtered outcomes after those inputs are interpreted. Stake weight names the influence value applied along that path rather than replacing either side (Glossary: Weight Matrix, Glossary: Rank, Yuma Consensus).

Rank Sums Post-Clip Weights by Stake

Yuma Consensus: Miner emissions documentation calculates each miner’s aggregate rank as the stake-weighted sum of post-clip validator weights. Stake weight therefore scales how much each submitted evaluation can move the merged miner result inside a subnet (Glossary: Rank).

That formula keeps evaluation and influence separate. Validator weights express what a validator scored; stake weight expresses how strongly that score counts when those signals are combined.

Validator Permits Follow Stake-Weight Ranking

The Glossary: Validator Permit describes the flag as awarded to the top neurons by stake weight on a subnet. Stake weight is therefore tied both to consensus influence and to which neurons hold validation rights in that subnet (Understanding Subnets: Validator stake weight).

Permit vocabulary answers whether weight submission rights exist. Stake weight explains the ranking context that determines who can receive those rights.

Clipping Benchmarks Use a Stake Fraction

Yuma Consensus clipping uses a stake-weighted benchmark for each miner. Weights that exceed that benchmark are clipped, so neither the miner nor the validator receives emissions for the excess portion. The kappa hyperparameter sets what fraction of total stake must support a weight level before it becomes the benchmark (Subnet Hyperparameters).

Stake weight therefore appears twice in the filtering path: it helps set the benchmark and it scales how surviving weights combine into rank after clipping.

Development Stage Context

The Introduction to Bittensor describes subnet development as moving from localnet to testnet and then mainnet. For stake weight, that sequence changes how readers should interpret validator-influence values and consensus examples.

In localnet, stake-weight behavior can be exercised in an isolated environment. Local stake-weight values reflect local chain configuration rather than production consensus influence.

On testnet, stake-weight behavior can be observed in a shared, non-production network. Testnet stake-weight values are separate from mainnet consensus state (Glossary: Stake Weight).

On mainnet, stake weight is the live measure of validator influence applied inside production subnet consensus on the Bittensor network.

The Bittensor Networks reference separates mainnet, testnet, and localnet. A stake-weight value from one environment should not be read as representing production consensus influence in another environment.

Further Reading

Topics ConsensusStaking
Last updated (UTC) 7 min read