Ethereum blob usage has reached a new all-time high this week. The three-day moving average is now 5.9 blobs per block, and the daily average has climbed to 6.7. Since blobs are the low-cost data slots that rollups use to post transaction batches, higher usage points to stronger Layer 2 demand for Ethereum’s data availability. It also shows that rollups are pushing more data through the network’s dedicated data layer than they have at any point since blobs went live.
The record matters, but it does not mean the network is close to saturation. Current demand sits at roughly 40% to 50% of Ethereum’s 14-blob target. Blocks with the maximum 21 blobs are still rare. The system is active without being under persistent strain. Usage had dropped during the spring, and this rebound now exceeds the previous peaks from late 2025. Rollups still have considerable headroom under today’s configuration.
Blob limits have expanded over time
Ethereum has not kept the same blob target since Dencun introduced the mechanism. Capacity started at a target of 3 blobs and a maximum of 6. Pectra raised those numbers to 6 and 9. BPO1 moved them to 10 and 15. BPO2, which went live in January, set the current target at 14 and maximum at 21.
That stepping pattern is important. Ethereum has chosen to expand data availability before rollups fully use what is already there. This has helped keep Layer 2 fees lower and created more room for transaction batches as rollup adoption grows.
Another increase is under discussion
Developers are already debating another change. The conversation is about moving the target to 21 blobs and the maximum to 32. The math is simple: a larger limit could make Layer 2 fees cheaper, but it would increase bandwidth requirements for node operators. Ethereum is also preparing for a gas-limit increase in Glamsterdam, and that adds to the operational concerns. The next decision is really a balance between keeping rollup execution cheap and making sure the network can carry the extra data without hurting decentralization.
Funding remains a separate bottleneck
There is also a funding problem that gets less attention. Protocol Guild has argued that support for Ethereum client teams is thin for a network of this size. That seems odd for a chain with a market cap around $300 billion, but the mismatch is real. More blob capacity means more complex client work, and that work needs ongoing money. The record usage is real progress, but Ethereum still has room to grow. That growth depends on more than block limits or rollup demand. It also depends on decisions about bandwidth, fee tradeoffs, and basic funding for the teams responsible for keeping the network running.
