Cream (http://cream.finance/) shows a good APR for USDT and the Idle is already testing with IdleRAI the connector.
I am exploring if we want to enable C.R.E.A.M. on IdleUSDT Best Yield.
All the current integrators could benefit from it (Harvest, Yearn etc).
Feel free to share your opinion or any question, thanks 
I think adding another borrower pool is definitely something to be explored especially given USDT is our 3rd largest pool (currently $20m TVL). Looking at the Analytics Dashboard, currently, the USDT pool is primarily allocated to Aave, which I believe is averaging around 11% APY vs. C.R.E.A.M’s 9.9% APY currently (screenshots below), I think it would be good to add an additional pool for diversification and ensuring Idle provides the consistent best yield.
It’s not clear how often Cream has a better yield than Aave (perhaps you have a view) but nonetheless, it’s a great diversification enhancement for our 3rd largest pool.
Looking forward to the results of your exploration! 
Thanks @emilianobonassi for bringing this up. As you said we are currently testing a IdleRAI implementation in beta which have a Cream integration, so the code is ready.
What’s your take on their security model @emilianobonassi ?
Btw to move this forward we would need to make a snapshot poll to test community sentiment and if successful an on-chain proposal which adds our Cream wrapper to the IdleUSDT.
As you stated APR depends on the time we look into, sometimes Aave offers better rates sometime is C.R.E.A.M. For instance, in this moment they are very close (Aave 10.05% and C.R.E.A.M 9,98%)
To expand my initial statement, C.R.E.A.M shows in general a good APR which aligns to other lending platforms, adding to IdleUSDT has several pros:
Regarding the security model, I would say that the Cream wrapper aligns with the Best Yield risks and probably will match even better the riskiest Tranche when it will be available.
C.R.E.A.M. has also a bounty with ImmuneFi offering 1.5M max in case of high severity issues
Considered that, Idle is doing a great job in terms of allocation strategy based on risk, so could be evaluated to apply some risk based allocation and do not go fully only to maximise the APR. Anyway, Best Yield in this moment represent the best fit for this kind of allocation.
Honestly I would like to see an Idle which is more agile, this choice represents a two-way door. You can always disable in the case it won’t give the desidered results. The success criteria can be increase in the TVL and yield generated for the users.
Thanks for the analysis @emilianobonassi, I see that Cream have a great bug bounty program and generally it follows all the guidelines here, the only ‘concern’ I have is the relative low liquidity for USDT (8.3M) but maybe it’s not a real issue. I would like to hear the opinion of other members too
Here the snapshot!
https://snapshot.org/#/idlefinance.eth/proposal/QmXSJQyL32LK35kVzMqEqAgXevh6Uf6H3SymUmWWhcjcDM
Feel free to express your support
hey @emilianobonassi really exciting to see. I work with the C.R.E.A.M. team on growth - we’re excited to be collaborating here.
I’d love to explore getting Idle integrated with the Iron Bank. As you can see here (https://app.cream.finance/markets) there’s a lot more USDT liquidity on the Iron Bank than on Cream V1 on Ethereum
