Metagraph
The metagraph is Bittensor’s structured view of state for a selected subnet. The official glossary defines it as a data structure containing comprehensive information about a subnet’s neurons and notes that it helps calculate emissions.
References: Glossary: Metagraph, The Subnet Metagraph
Subnet Context
A metagraph belongs to a specific subnet context. The subnet metagraph documentation uses netuid 1 as an example of that context and describes the metagraph as a snapshot of subnet state at a particular block. This makes the term different from a general network summary: it is about the state of the selected subnet.
Reference: The Subnet Metagraph
Neuron State Fields
Official references describe the metagraph as carrying information about neurons in the selected subnet. That state includes consensus and incentive-related values: stake-related state, weights, ranks, trust, consensus, incentives, dividends, emissions, activity state, and bonds. These values describe chain-derived subnet state, not editorial ratings.
References: The Subnet Metagraph, Understanding Neurons
Subnet State Role
The metagraph gives readers a common term for subnet-level state used around Bittensor incentives. Yuma Consensus explains how validator evaluations aggregate into miner incentives and validator dividends; the metagraph is where those subnet-level results are represented as a structured snapshot.
References: Yuma Consensus, Glossary: Metagraph
Development Stage Context
The Introduction to Bittensor describes subnet development as moving from local testing to testchain and then mainchain. For a metagraph, that sequence changes what kind of subnet-state evidence a reader is seeing.
In local testing, a metagraph can show whether a subnet’s participant state, weights, and incentive-related fields are being represented in a controlled environment. That is useful for understanding the shape of subnet state, but it remains isolated development evidence.
On testchain, a metagraph can reflect subnet state in a shared testing network. This gives readers stronger evidence about how participants and validator signals appear together than a private local run, while still keeping the snapshot separate from production mainchain state.
On mainchain, a metagraph belongs to live Bittensor subnet state. The Subnet Metagraph documentation describes the metagraph as a snapshot at a particular block, so production readings should keep both the selected netuid and the block-height context attached.
The Bittensor Networks reference separates mainnet, testnet, and localnet. A metagraph observed in one environment should therefore not be read as the same kind of evidence as a metagraph observed in another environment.
That distinction also protects block-height language. A local block height, a testchain block height, and a mainchain block height can each locate a metagraph snapshot, but they locate it inside different network histories.
For readers, this keeps subnet-state examples from sounding stronger than their environment supports. Local, testchain, and mainchain metagraphs can all describe subnet state, but they do not carry the same interpretive weight.
Snapshot at a Particular Block
The Subnet Metagraph documentation describes the metagraph as a snapshot of subnet state at a particular block height. That timing detail matters because metagraph readings are block-specific, not timeless summaries of a subnet.
A metagraph statement should therefore keep both the selected netuid and the block context attached. Without that pair, the snapshot can be misread as current live state when it is only state as of one recorded block.
Later blocks can change neuron fields, so comparing metagraphs requires comparing like block heights or noting that the snapshots come from different points in chain history.
References: The Subnet Metagraph, Glossary: Metagraph
Emission Calculation Context
The Glossary: Metagraph notes that the metagraph helps calculate emissions. It carries neuron-level incentive fields such as weights, ranks, trust, consensus, incentives, dividends, and emissions that feed subnet reward processing.
That makes the metagraph an accounting snapshot for incentive flow, not a miner-quality rating by itself. The fields describe chain-derived subnet state at the snapshot block.
Those values are produced by subnet consensus and validator evaluation flow rather than by editorial ranking language.
References: Glossary: Metagraph, Understanding Neurons
Netuid Selects the Subnet View
A metagraph belongs to one subnet at a time. Official examples use a specific netuid, such as netuid 1, to show which subnet’s neuron state is being described (The Subnet Metagraph).
Netuid therefore selects the subnet lens for the snapshot. A metagraph for one netuid does not describe neuron state across every subnet at once.
Subnet comparisons should use separate metagraph readings per netuid rather than mixing fields from different subnet snapshots.
Each netuid keeps its own neuron list and incentive fields in metagraph vocabulary.
References: The Subnet Metagraph, Glossary: Netuid
Relationship to Multiple Mechanisms
A metagraph snapshot usually describes one subnet context at a time, but subnets with multiple incentive mechanisms can carry mechanism-specific incentive state. The official documentation describes additional metagraph fields for per-mechanism weights and incentives when a subnet runs more than one evaluation path.
For readers, the metagraph is still a subnet state view rather than a scoring rule. Multiple mechanisms mean some weight and incentive values should be read in the context of a particular mechanism, not as one blended score across every task the subnet supports.
References: Multiple Incentive Mechanisms Within Subnets, Glossary: Multiple Incentive Mechanisms
Relationship to Netuid
A metagraph and a netuid are related but different parts of Bittensor subnet vocabulary. A netuid selects which subnet context is in scope, while a metagraph names the structured subnet state view for that selected subnet. The Glossary: Netuid places netuid at the subnet level, and the Glossary: Metagraph describes comprehensive information about a subnet’s neurons.
For readers, netuid answers which subnet market is being discussed, while a metagraph answers what subnet-level neuron state looks like inside that selection. Metagraph language should stay paired with the netuid that selects the subnet context.
References: Glossary: Netuid, Glossary: Metagraph
Relationship to UID Slot
A metagraph and a UID slot are related but different parts of subnet participant vocabulary. A UID slot describes a participant position inside a subnet, while a metagraph carries subnet-level state that includes values associated with those positions. The Glossary: UID Slot describes a participant position, and the Subnet Metagraph documentation describes neuron state fields for participants in a selected subnet.
For readers, a UID slot names the position itself, while a metagraph names the broader subnet state view in which that position appears alongside stake, activity, and incentive-related values.
References: Glossary: UID Slot, The Subnet Metagraph
Relationship to Active UID
A metagraph and an active UID are related but different parts of subnet participation vocabulary. An active UID describes activity status for a participant slot, while a metagraph exposes subnet state fields such as activity-related values for neurons in a selected subnet. The Glossary: Active UID describes a slot whose participant is considered active, and the Subnet Metagraph documentation names activity state among the fields in a subnet snapshot.
For readers, an active UID names the activity concept for one slot, while a metagraph names the subnet-level state view that can include activity information for many slots at once.
References: Glossary: Active UID, The Subnet Metagraph
Relationship to Block
A metagraph and a block are related through subnet state timing. Official metagraph documentation describes the metagraph as a snapshot of subnet state at a particular block height. The Glossary: Block describes a block as a unit of chain data on Subtensor, and the Glossary: Metagraph describes comprehensive subnet neuron information used around incentive calculations.
For readers, a block names one point in chain progression, while a metagraph names subnet neuron state as represented at that kind of chain-height context for a selected netuid.
References: Glossary: Block, Glossary: Metagraph, The Subnet Metagraph
Relationship to Neuron
A metagraph and a neuron are related but different parts of Bittensor subnet state vocabulary. A neuron names a registered participant in a subnet, while a metagraph exposes comprehensive subnet neuron information as a state snapshot for a selected netuid. The Glossary: Neuron describes a registered subnet participant, and the Glossary: Metagraph describes comprehensive subnet neuron information used around incentive calculations.
For readers, a neuron names one registered participant, while a metagraph names the subnet-level view that can include many neurons at once.
References: Glossary: Neuron, Glossary: Metagraph
Relationship to Subnet Miner
A metagraph and a subnet miner are related but different parts of Bittensor subnet vocabulary. A subnet miner names the task-performing role inside a subnet, while a metagraph exposes subnet state fields that can include information about participants serving in miner roles for a selected netuid. The Glossary: Subnet Miner describes the work-producing role, and the Subnet Metagraph documentation describes the subnet state snapshot that includes neuron fields.
For readers, a subnet miner names one work role, while a metagraph names the subnet-level state view that can represent many participants and their roles together.
References: Glossary: Subnet Miner, The Subnet Metagraph
Relationship to Subnet Validator
A metagraph and a subnet validator are related but different parts of Bittensor subnet vocabulary. A subnet validator names the evaluation role inside a subnet, while a metagraph exposes subnet state fields that can include information about participants serving in validator roles for a selected netuid. The Glossary: Subnet Validator describes the evaluation role, and the Subnet Metagraph documentation describes the subnet state snapshot that includes neuron fields.
For readers, a subnet validator names one evaluation role, while a metagraph names the subnet-level state view that can represent many participants and their roles together.
References: Glossary: Subnet Validator, The Subnet Metagraph
Relationship to Register
A metagraph and register are related but different parts of the subnet participation lifecycle. Register describes the subnet-entry process that grants a UID slot, while a metagraph exposes subnet state after participants have entered and appear in the subnet snapshot for a selected netuid. The Glossary: Register describes subnet entry, and the Glossary: Metagraph describes comprehensive subnet neuron information used around incentive calculations.
For readers, register names how a participant enters a subnet, while a metagraph names the subnet state view that can include participants after entry.
References: Glossary: Register, Glossary: Metagraph
Relationship to Yuma Consensus
Metagraph 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, metagraph 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
This page defines the term at a high level. It does not report current subnet membership, current stake, current emissions, wallet ownership, or live identity fields. Those values change with chain state and should be checked for the relevant netuid when current data is needed.
Hotkey and Coldkey Fields Identify Each Neuron Row
The Subnet Metagraph documentation lists
hotkeys and coldkeys among metagraph fields for neurons on a selected subnet. On a subnet such
as netuid 1, each row ties a UID to
the operational hotkey and owning
coldkey addresses recorded at the
snapshot block.
Those identity fields are chain-derived state, not wallet guidance. A metagraph reading shows which keys were associated with each neuron position at that block height rather than ownership after a later hotkey swap or coldkey swap.
Axon Info Records Miner Endpoint Material
The metagraph reference describes axons and per-neuron axon_info as network connection details
for axon servers. An axon represents a
server run by a registered miner that can answer validator requests.
On netuid 1, axon fields sit beside incentive numbers inside the same snapshot. They name how a miner endpoint was registered at the recorded block, separate from rank or emission values.
Validator Permit Appears as a Metagraph Flag
The subnet metagraph table includes validator_permit as a boolean array indicating whether each
neuron has validator permits to set weights and participate in consensus. That flag connects
metagraph vocabulary to
validator permit rules
without replacing the permit article itself.
Reading permit status from a metagraph snapshot requires the netuid and block context. The boolean field describes chain-recorded permit state at one height, not a live guarantee for later blocks.