For an on-chain transfer, people don’t need to know who you are.
They only need your wallet address
to follow your transaction history and see how your balance changes and where your funds go
and even figure out your habits
This is already a problem for ordinary users
and even more so for institutions managing large sums of money
@primus_labs’ recent technical updates have focused on exactly this issue
On November 13, Primus shared technical progress on AutoHoG
The core goal is to improve the computational efficiency of FHE (Fully Homomorphic Encryption)
FHE allows data to be computed on without being decrypted
Under traditional approaches, processing a single 64-bit operation
may require breaking it down into thousands of logic gates
and performing extensive Bootstrapping
Computational cost has long been an obstacle to real-world adoption
AutoHoG’s approach is to automatically design custom gates with multiple inputs and outputs
combining subcircuits that can be merged into a single computation
and reducing redundant Bootstrapping
The results reported by the team:
8 consumer-grade RTX 5090 GPUs
A latency of about 90ms for 64-bit confidential transfers
Throughput of over 210 TPS
Integer operations guarantee byte-for-byte identical results across different nodes
On-chain finance can’t just run a computation on a single machine and call it a day
Multiple nodes must agree on the execution results
If privacy computing affects verification and consensus
it will be difficult to make it part of financial infrastructure
Other parts of Primus’s strategy:
zkTLS verifies off-chain data
FHE enables computation on encrypted data
Confidential Finance brings privacy capabilities to transfers and yield scenarios
What Primus is working toward:
Sensitive data shouldn’t have to be exposed to everyone
while preserving its verifiability and usefulness
The team also put it very directly:
Privacy shouldn’t come after the damage. It should be there from the start.
Privacy shouldn’t be something we try to fix only after damage has been done
These words carry real weight in on-chain finance
especially for institutional funds, private credit, and on-chain asset management
Privacy isn’t decoration
It’s an important condition for bringing more traditional financial players on-chain
The 90ms and 210 TPS figures are technical test results published by the team
More validation is needed before large-scale real-world deployment
Primus is no longer just discussing the concept of privacy
It is tackling the computational efficiency of FHE
and working to bring it into real financial applications
#Primus @primus_labs #FHE
They only need your wallet address
to follow your transaction history and see how your balance changes and where your funds go
and even figure out your habits
This is already a problem for ordinary users
and even more so for institutions managing large sums of money
@primus_labs’ recent technical updates have focused on exactly this issue
On November 13, Primus shared technical progress on AutoHoG
The core goal is to improve the computational efficiency of FHE (Fully Homomorphic Encryption)
FHE allows data to be computed on without being decrypted
Under traditional approaches, processing a single 64-bit operation
may require breaking it down into thousands of logic gates
and performing extensive Bootstrapping
Computational cost has long been an obstacle to real-world adoption
AutoHoG’s approach is to automatically design custom gates with multiple inputs and outputs
combining subcircuits that can be merged into a single computation
and reducing redundant Bootstrapping
The results reported by the team:
8 consumer-grade RTX 5090 GPUs
A latency of about 90ms for 64-bit confidential transfers
Throughput of over 210 TPS
Integer operations guarantee byte-for-byte identical results across different nodes
On-chain finance can’t just run a computation on a single machine and call it a day
Multiple nodes must agree on the execution results
If privacy computing affects verification and consensus
it will be difficult to make it part of financial infrastructure
Other parts of Primus’s strategy:
zkTLS verifies off-chain data
FHE enables computation on encrypted data
Confidential Finance brings privacy capabilities to transfers and yield scenarios
What Primus is working toward:
Sensitive data shouldn’t have to be exposed to everyone
while preserving its verifiability and usefulness
The team also put it very directly:
Privacy shouldn’t come after the damage. It should be there from the start.
Privacy shouldn’t be something we try to fix only after damage has been done
These words carry real weight in on-chain finance
especially for institutional funds, private credit, and on-chain asset management
Privacy isn’t decoration
It’s an important condition for bringing more traditional financial players on-chain
The 90ms and 210 TPS figures are technical test results published by the team
More validation is needed before large-scale real-world deployment
Primus is no longer just discussing the concept of privacy
It is tackling the computational efficiency of FHE
and working to bring it into real financial applications
#Primus @primus_labs #FHE