diff --git a/content/projects/plasma-fold.md b/content/projects/plasma-fold.md
index d81532d..0dd14b0 100644
--- a/content/projects/plasma-fold.md
+++ b/content/projects/plasma-fold.md
@@ -30,13 +30,13 @@ Unlike traditional rollups that post extensive transaction data to L1, Plasma Fo
 
 ## Why Plasma Fold?
 
-Most L2s today struggle to scale without relying on increasingly expensive data availability solutions (e.g., blobs). Plasma Fold offers a low-cost, high-throughput alternative. Inspired by Vitalik Buterin's vision of a Plasma-EVM hybrid, Plasma Fold explores the use of folding schemes to support lightweight client-side proving—ideal for mobile or browser-based clients. In theory, this architecture could enable transaction rates in the range of 90k–260k TPS, matching or exceeding centralized systems like VISA.
+Most L2s today struggle to scale without relying on increasingly expensive data availability solutions (e.g., blobs). Plasma Fold offers a low-cost, RAM and proving-speed efficient, high-throughput alternative. Inspired by Vitalik Buterin's vision of a Plasma-EVM hybrid, Plasma Fold explores the use of folding schemes to support lightweight client-side proving—ideal for mobile or browser-based clients. In theory, this architecture could enable transaction rates in the range of 90k–260k TPS, matching or exceeding centralized systems like VISA.
 
 ## Key Innovations
 
-- **Client-Side Proving with Folding Schemes:** Users generate recursive ZK proofs on their devices to attest to their own balances and transactions, minimizing trust and computation on the L1.
+- **Client-Side Proving with Folding Schemes:** Users generate IVC proofs on their devices to attest to their own balances and transactions, minimizing trust and computation on the L1.
 - **UTXO-Based Design with IVC:** Transactions are modeled as UTXO trees, allowing users to maintain and prove their own balances incrementally across time.
-- **Minimal Onchain Footprint:** Only block roots, nullifier indices, and signer bitmaps are posted on L1—greatly reducing gas costs and data needs.
+- **Minimal Onchain Footprint:** Only block roots, nullifier indices, and signer bitmaps are posted on L1—greatly reducing gas costs and data needs, using Intmax low onchain data model.
 - **Instant Exits & Offline-Friendliness:** Users can exit the system at any time using a locally stored proof. They do not need to stay online between blocks unless actively transacting.
 - **Composable and Future-Proof:** Designed to eventually support multiple assets, order books, and potentially even programmable features.