Bento-based Treasury-funded AMM with a single-sided or any flexible ratio of LP pair


  1. TWAP oracles are in place and ready to use.
  2. Bento-based AMM in place (optional but ideal related to 7/20)

Description of the Proposal:
The basic idea is to have a new LP structure by having Treasury-funded SUSHI in the place of ETH in LPs. For example, with the new AMM architecure in Bento, for LINK-USDC trading needs, I propose that we will have LINK-SUSHI and SUSHI-USDC pairs instead of LINK-ETH and ETH-USDC pairs. The LINK-SUSHI LP pool participants can provide either a single-sided LP such as $1000 worth of LINK only or any preferred ratio of the LP pair, such as $1000 worth of LINK and $200 worth of SUSHI.

In case of the single-sided LP where the LP provider provides $1000 worth of LINK only for the LINK-SUSHI pair, the other side of the pair for $1000 worth of SUSHI will be provided by the SUSHI treasury-funded Tier1 SUSHI bento pool. The LP providers have 0.25% fees for the LP. The above LINK LP provider will end up earning 50% of the trading fees proportionate to their share. The remaining 50% of the fees will go to the SUSHI-funding pool.
Here in this case, the pool now has the 50% of the fees in the form of LINK-SUSHI LP token. These tokens can now be converted to each of the underlying tokens, LINK and SUSHI respectively. This results in the treasury-funded pool having organic growth in other quality tokens and stable coins without loosing their underlying SUSHI holdings and any negative SELL pressure in the market. This new architecture will contribute to a bigger TVL in total and more use cases of their treasury fund.

In case of the 2nd flexible LP ratio scenario where the LP provider provides $1000 worth of LINK and $200 worth of SUSHI, the remaining $800 worth of SUSHI will be funded by its respective tier1 SUSHI pool, the portion of the fees earned will be slightly higher than the previous scenario where they can get 60% of the fees earned whereas the treasury-funded SUSHI pool will get the remaining 40%, which will contribute to organic growth of our treasury diversification.

I’d propose that we provide more benefits to the LP providers who are providing SUSHI portion as well. Up to a certain portion of the earned fees of the treasury-funded SUSHI pool for each LP pair, we can re-direct incentives to those providers.

With capital efficiencies in place in Bento, we will likely earn more yield and can possibly incentivize those LP providers.

Possible Benefits

  • This potentially results in more buy pressure of SUSHI from LP providers who seek more fees and yields.
  • It can also lead to organic growth of other quality tokens in our Treasury by participating in the new LP markets.
  • This new architecture will make the idle/unused treasury fund actively utilized to increase the overall TVL for trading and the TVL of BentoBox itself.
  • SushiSwap team might not need to be in a rush to sell a lot of SUSHI in the treasury to VCs at a big discount for us to undergo upcoming possible bear market.
  • It can open a door to single-sided LP providers

I am just throwing this idea as a tiny way to improve our current situation and the future of SushiSwap. There can be several downsides to this idea. I hope all aspects of risks, feasibilities , and technical difficulties can be discussed.
Anybody with good intent can steal my idea and I hope it can be implemented eventually.


Interesting! Reminds me of the Bancor design, which uses BNT as the reserve asset in all pairs and offers single-side LP.

It seems like a great design for LPs but not necessarily traders as most folks are trading pairs that don’t include $SUSHI so it would involve routing through multiple pools.

So in such a setup, the SUSHI-ETH pool might be given a lower fee profile, say .1% rather than .3% fee, to balance it out, similar to how BNT-ETH pool has lower fees on Bancor.