Welcome to the next crypto level
← Nexorena

Altcoins

TrajectoryRL enables a miner reward burn mode by default

TrajectoryRL validator 0.7.8, tagged November 8 on GitHub, directs weights by default to the owner identifier for Bittensor subnet 11. The effect on distribution depends on validator agreement.

Nexorena Editorial Team

TrajectoryRL enables a miner reward burn mode by default

The validator changes where weights go

TrajectoryRL tagged version 0.7.8 of its validator software on November 8, 2026. The code published for Bittensor subnet 11 enables a setting called BURN_ALL by default. When this mode is in effect, the validator assigns 100% of the weight it submits to owner identifier 74. In this branch of the code, the software does not consult the seated winner to choose where the weights go. The change applies to the validator’s weight-setting routine.

TrajectoryRL’s documentation describes normal operation as a competition in which miners submit policy bundles and validators evaluate their performance. The system identifies a winner who keeps the seat until a challenger takes it. In normal mode, validators direct the miner share to the seated winner. Directing weights to the owner was already used as a fallback when there was no eligible winner. The new setting uses that destination even when a winner is seated.

Burning depends on the weights actually submitted

In the change request, the maintainers explain that the subnet is configured to burn the miner share when the weights follow this route. The owner identifier is the destination of the weights. That destination should therefore not be read as a payment of the miner share to the owner. The mechanism concerns the distribution of that share under the subnet settings described by the maintainers.

The project says the effect depends on weights submitted by multiple validators. In the Yuma consensus mechanism, a validator’s weight for its own identifier is ignored, according to the explanation accompanying the change. The owner therefore cannot produce the effect with its own row of weights alone. Making the setting the default is intended to align validators using the version. The version tag confirms the code is available, while the collective effect requires checking weights and distribution on the network.

An option preserves the winner’s weights

The BURN_ALL setting can be set to zero to restore the normal branch, in which the weights go to the winner when one is seated. The tests added to the repository cover the default, the destination of weights when the mode is enabled, and the return to winner weights when the option is disabled. They verify software behavior in defined scenarios. They do not measure how many validators are running the version.

The November 8 change thus separates the outcome of the competition from the assignment of rewards. The competition software can consider a policy the winner while validators using the new setting send weights elsewhere. Determining whether a given share was burned, and to what extent, still requires observing subnet data after operators put the version into service.

Written with love byRyan MitchellMeet this editorial pen name ↗

Explorer Nexorena