Some notes from the @Lighter_xyz fireside chat with @VitalikButerin, @ethereumJoseph, @vnovakovski and @n_lighterxyz
• Lighter does ~5k TPS on a regular day, peaked at ~20k yesterday during volatile market. Uses <1% of Ethereum blockspace, posts 2-3 txs/min on L1
• EVM sidecar rollup currently in DevNet, first apps expected in the coming months
• Prover cluster scaled ~100x since launch. Proving is no longer the main bottleneck -- coordination of systems and data movement is
• Stage 1 is close, stage 2 is the goal. Concern raised: 30day upgrade delay makes shipping new primitives (onchain options) harder
• User mix: maker side is ~90% pro (MMs, HFTs). Taker side roughly 50/50 retail vs institutional.
• Institutions on Lighter mostly hedge and arbitrage. Retail mostly speculates. Vitalik wants to see more "positive-sum" use cases independent of crypto native demand
• Single sequencer with FIFO ordering. TEE and MPC encrypted mempools being researched. Vitalik: the goal isn't decentralizing the sequencer per se -- it's giving users mathematically strong guarantees against adversarial MEV
• Hybrid DA model: balances and open positions are recoverable from Ethereum (escape hatch works), order book state is not posted. If they posted everything it would exceed 100% of blob capacity
• Vitalik's main take: oracles are the skeleton in the closet of the entire industry. No point hardening ZK proofs and pushing Stage 2 while oracles often run on a single server. "Security is only as strong as the weakest link"
• Current oracle setup: combination of third party providers for major markets, AMM based pricing for pre-launch/exotic markets, futures based pricing for commodities like crude oil. Oracle inputs controlled by the sequencer
• Vitalik's longterm vision: universal futures markets enable personalized CPIs -- hedging rent, AI chip costs, individual consumption baskets. Speculators arrive first, real financial substitute comes later
gLighter🕯️