NIP-2: Change stake saturation and active set to improve network decentralisation and efficiency

ID title created status kind author(s) champions
NIP-2 Change stake saturation and active set to improve network decentralisation and efficiency 2025-06-17 proposed token economics Claudia Diaz, Nym Chief Scientist claudia@nymtech.net Jaya Klara Brekke, Nym CSO jaya@nymtech.net Serinko, Operator tech lead serinko@nymtech.net Daniel Vasquez, Head of Growth Latam daniel@nymtech.net

Summary

This is a governance proposal to lower the Nym node stake saturation point from 1M to 500k or 250k NYM, and change the mixnet active set composition to increase the number of available gateways.

Motivation

The change is proposed to improve the efficiency and decentralization of the Nym network.

The Nym network currently has a high stake saturation point making it hard for new operators to become competitive. This leads to low competition and a centralising tendency in the network.

Furthermore, NymVPN usage shows a much higher demand for gateways (as they also run WireGuard fastmode). The proposal therefore also includes a change to active set composition to increase the number of gateways and decrease mixnodes in order to better meet demand.

Details

Aims and objectives

The aims and objectives of the proposal are:

  • Prevent centralization of node operators, rewards and stake.
  • Lower the barrier to entry for operators to enable a more competitive system where the best performing nodes can easily rise to the top.
  • Achieve an increase in available 2-hop gateways.
  • Ensure stake and rewards are distributed rationally and fairly to nodes performing work for end-users.

Who is affected?

The adoption and usage of NymVPN has shown that there is more demand for Fast (2-hop wireguard) mode rather than mixnet anonymous mode. This means the current mixnet active set has too many mixnodes, with rewards being distributed to nodes that are not directly serving user demand. This also affects the efficiency of the system: when there are many available routes, but not very much usage, the amount of mixing per node decreases. This means worse anonymity than with fewer mix nodes, as well as many rewards being distributed to nodes that are not contributing a significant amount of bandwidth which can create unnecessary pressure on token price.

Current composition: 50 entry gateways / 120 mixnodes (40 per layer) and 70 exit gateways.

Proposed composition: 80 entry gateways / 60 mixnodes (20 per layer)/ 100 exit gateways.

In this context, we are proposing a change to stake saturation and active set composition to enable new gateway operators to more easily reach saturation and compete for rewards.

Voting options

  1. New stake saturation: 500k NYM
    • Active set composition: 80 entry gateways / 60 mixnodes (20 per layer)/ 100 exit gateways
  2. New stake saturation: 250k NYM
    • Active set composition: 80 entry gateways / 60 mixnodes (20 per layer)/ 100 exit gateways
  3. Stake saturation: 1M NYM (current saturation point)
    • 80 entry gateways / 60 mixnodes (20 per layer)/ 100 exit gateways
  4. No changes.

Background information

What is the economic impact of these proposed changes?
Nym Chief Scientist Claudia Diaz has modeled the economic impact and collated into a report. The report can be downloaded and read in full here (english version).

What is ‘active set’?

Active set describes the nodes that are actively routing user traffic in a given epoch (approximately 1hr). Every epoch, the active nodes change, to weed out the badly performing nodes, so that the active nodes are as performant as possible; to give more nodes opportunities to participate and healthy competition; to prevent attacks where the adversary adapts to fixed node positions to construct fully malicious paths making it very hard to construct a malicious pathway.

What is ‘active set composition’?

Active set composition describes the available roles that a node can fill. The current active set composition has 50 entry gateway roles / 120 mixnode roles across 3 layers and 70 exit gateway roles. The proposal for a new active set composition is 80 entry gateways / 60 mixnodes (20 per layer)/ 100 exit gateway roles.

What is staking in the Nym ecosystem?

Anyone holding native NYM tokens can stake on a Nym node as a signal of good reputation and quality of service. This acts as a form of decentralized quality control in the network. In return, token holders receive a share of the nodes rewards and so are incentivised to stake on well performing nodes. Stake affects the likelihood of a node being selected to do work and therefore affects the rewards it will get.

What is stake saturation point?

To prevent a handful of nodes from getting all the stake, there is a softcap on how much stake a node can receive, before it starts to see diminishing returns. This is called the stake saturation point.

How is the stake saturation point determined?

Stake saturation point is arrived at by first assessing the current staking supply (total number of NYM tokens available for staking). This figure is the total supply (1Bn), minus the mixmining reward pool, giving a total stake supply of 812,772,499.

However, not everyone stakes their tokens. So this total is then adjusted by a staking factor (the fraction of total supply that is estimated will be staked), to arrive at target staking supply number. The target staking supply is then divided across the total number of nodes in the active set (240), to arrive at the stake saturation point, which is currently approximately 1M NYM.

The vote will result in an adjustment to the staking factor, which in turn adjusts the saturation point.

5 Likes

I think a saturation ceiling should also be added, so people can’t delegate beyond 100% saturation.
Many users don’t follow project updates closely — some delegate and then forget for a while. As a result, a node runner currently has no mechanism to prevent going over 100% saturation and losing rewards. Specially, if the saturation point will change.

A saturation ceiling would also help distribute delegations more evenly. If users can’t stake above 250K/500K tokens, they’ll be more likely to delegate to another node with lower saturation or make partial delegations across multiple nodes.

Additionally, during a discussion within our RawBox squad, our member Oleksky suggested that without a saturation ceiling and with low saturation point, a malicious actor could launch a delegation poisoning attack by over-delegating to the nodes with high saturation — reducing their rewards and making them less profitable or even unprofitable.

4 Likes

Let’s test the new Nym Governator!

As you know, Nym Improvement Proposals (NIPs) are being introduced. NIP-1 and NIP-2 above are already live for discussion, and later this week voting will open up for Nym node operators.

You can help us test the new tool so it’s ready for prime time when NIP-1 and NIP-2 voting kicks off, and select my webcam background from the options shared above for next week’s community call!

Here’s how you can participate:

  1. add your Nym wallet address to Keplr (use the address your Nym node is bonded to. Search and select the Nyx chain)
  2. go to the proposal on Nym Governator.
  3. authenticate yourself with Keplr
  4. select your favorite background (options were provided by the community on this thread)

And viola, congratulations - you are among the very first users of Nym Governator! :1st_place_medal:

If you have questions or run into any issues, please let me know!

Ps: this test vote will be open for 3 days so hurry up :running_woman:

Nice one, I like it. So if I understood correctly, the idea is a hard cap at saturation rather than a soft-cap as is now? @claudianym what do you think about this proposal.

And in terms of compounding stake, that would imply that stake would compound until saturated, after which rewards would simply get sent to the address and remain liquid.

1 Like

Nice idea, and I don’t see a problem disallowing new delegations that would oversaturate a node if they went through.

In terms of compounding, I don’t know how complex is to allow compounding until saturation and then sending rewards to the address that remain liquid. If this is complex to implement, which could be, it should not be a big problem to allow compounding to oversaturate, since it would be an oversaturation amount that grows quite slowly.

The oversaturation that can be created with new delegations is however instant and unbounded, so a much bigger issue than compunding creeping up over the saturation level.

The compounding stake issue can be resolved by making only staked/bonded tokens active. This way, rewards would accumulate in the background and wouldn’t play an active role in the node until the moment a node runner or delegator withdraws and re-delegates/bonds them to the node.

Additionally, this could encourage people to stay more engaged on a monthly basis — checking their rewards, re-delegating them, or changing the node they delegate to. Currently, a delegator can stake and forget.

1 Like

After this proposal passes, how soon before the changes go into affect?