Gas fees have already been destroyed on Starknet; it's now time to improve our transaction finality and TPS capacity! 🐺 After collecting numerous feedbacks from our community, here is the updated roadmap of Starknet, fully focused on UX improvement 💅 🗓️ v0.13.2 August 💠 Parallel execution 💠 Block Packing The main goals of these two significant upgrades are to decrease the confirmation time to 2 seconds and the block time without increasing L1 costs. Additionally, we aim to enhance the 'execute' limit to ≥10M steps, thereby increasing computation capacity by 2.1 times, and to boost our TPS capacity. WTF does this mean? Very simple: faster transactions for users, without any additional cost, a better developer experience, and an increased congestion threshold 🕺 💅 🗓️ v0.13.3 October/November 💠 Cairo Native This upgrade intends to optimize our Sequencer by incorporating the state-of-the-art Cairo-native by @LambdaClass. The goal is to enable the sequencer to bypass the VM entirely and execute native CPU opcodes. Even more WTF than v0.13.2, what does this mean? Three main things 🤓 🐺 Improved TPS capacity, enabling Starknet to further increase its congestion threshold 🐺 Improved finality, resulting in faster transactions for users 🐺 Enhanced developer experience 💅💅 🗓️ v0.14 January/February The implications of v0.14 are yet to be decided. We will adapt based on network performance and community demand. Here are the possible candidates: 💠 Volition: to further destroy gas fees 💠 zkThreads: a novel framework for execution sharding that provides scalability and customizability without fragmentation (ELI5: L3/appchain benefits without the fragmentation issue) 💠 Mempool: to improve the organization and management of transactions (resulting in fewer rejected transactions), enhancing both user and developer experience 💠 Protocol-level paymaster: easy integration of paymaster for dApps, allowing developers to greatly improve the UX of their dApps 💅💅💅 For further details, or if you have any questions or feature requests 👇