In June 2024, Ethereum core developers and technical teams from the Layer 2 and ZK ecosystems gathered in Berlin for five days of deep collaboration. This event, known as Berlinterop, was more than just an engineering workshop — it showcased three clear technical trajectories Ethereum is pursuing for 2025: accelerated execution layer upgrades, a redefined mechanism for L1–L2 collaboration, and the standardization of the zero-knowledge (ZK) stack.

Berlinterop continues the tradition of “interop hacking week,” but with a sharper focus on Ethereum’s next technical chapter.

One of the central technical focuses at Berlinterop was the Fusaka upgrade. Developers launched two testnets (fusaka-devnet-1 and berlinterop-devnet-2) and conducted intensive experiments around execution performance, block-building optimization, and parallel execution paths.

As a significant execution-layer upgrade, Fusaka is designed not only to improve throughput but also to pave the way for future upgrades such as Pectra (e.g., proposals around parallel EVM and shorter slot times).

The dev week also set a clear timeline: fusaka-devnet-2 is expected to be released for community testing soon, with a rollout to the Sepolia testnet by late summer — preparing the ground for a mainnet upgrade.

This is Ethereum’s most direct response to the “performance” challenge in recent years and sends a strong signal to modular and high-performance Layer 2 protocols.

Berlinterop featured a dedicated L2 Collaboration Day, with representatives from Arbitrum, Base, OP Labs, Polygon, Scroll, Starkware, World Chain, and ZKsync. Together, they discussed the current state and limitations of L1–L2 collaboration.

A new consensus is emerging: Layer 2 is no longer just a user of Ethereum — it is a co-builder and experimental ground for protocol evolution.

During the discussions, L2 teams shared three main requests:

This marks a pivotal shift: Layer 2s are not just “accelerators” — they are becoming part of Ethereum’s core operating system.

We can expect future EIPs, gas models, and protocol upgrades to feature greater L2 involvement and design coordination, laying the foundation for a standardized multi-chain + cross-chain abstraction future.

Another major highlight of Berlinterop was the zk-focused track. Researchers from Scroll, Succinct, Starkware, ZKsync, ZKM, RISC Zero, and others held in-depth discussions around zkEVMs, stateless clients, and ISA standardization.

The Stateless Client track drew particular attention — envisioning a lightweight Ethereum client that verifies blocks using zk proofs without needing local state. This could become the default shape of next-gen Ethereum nodes.

Their roadmap targets a working prototype by the end of 2025, addressing key challenges such as:

These efforts will fundamentally redefine what it means to run an Ethereum node: lightweight, verifiable, modular execution will become the new norm. In other words, we are one step closer to the vision of “compressing Ethereum into a browser plugin.”

Together, these form the three major upgrade challenges and opportunities that Ethereum will face in 2025.

Berlinterop sent a clear message to the entire ecosystem:

Ethereum’s technical iteration is accelerating, and L1–L2 coordination has entered a deeper phase. Execution performance, interoperability, and modular zk infrastructure are emerging as the three most definitive trends for 2025.

As a Layer 2 built on the Superchain, committed to building next-generation infrastructure for on-chain NFTs and Agents, Mint is closely tracking and actively engaging with these future-defining technical shifts.

We believe that only by continuously pushing for core protocol upgrades and ecosystem collaboration can we deliver a more efficient, secure, and innovative blockchain experience to users.

Mint is already on the path.