When EIP-4844 shipped with the Dencun upgrade in March 2024, blobs were almost embarrassingly cheap. Layer 2 teams were posting data to Ethereum for fractions of a cent per transaction. The design included a target of three blobs per block and a hard cap of six, with a fee mechanism modeled on EIP-1559 - base fee adjusts based on whether you’re above or below the target. For most of 2024 and into early 2025, the blob base fee sat near its floor. The market was effectively uncontested.
That’s changed. As of mid-2026, blob base fees have been spiking intermittently during periods of high L2 activity - particularly when Arbitrum, Base, and Optimism post data simultaneously during volatile market conditions. The blob fee market is behaving the way Ethereum’s execution fee market did in 2020 and 2021: calm until it isn’t, then brutal.
This matters for a few reasons. First, it confirms that blobs are actually being used at scale. The original concern post-Dencun was that L2s would stay on Ethereum for data availability but eventually migrate to cheaper alternatives - EigenDA, Celestia, Avail - once their ecosystems matured. Some have. Base briefly experimented with posting to EigenDA. But Ethereum blobs remain the dominant choice for the largest rollups, partly for security reasons and partly because the cost difference narrowed as blob fees rose elsewhere.
The Pectra Effect
Pectra, which shipped in May 2025, raised the blob target to six and the cap to nine. That relief was real but temporary. More L2 throughput followed, absorbing the additional capacity faster than most analysts expected. The Ethereum core devs are already discussing whether Fusaka - the next major upgrade - should push blob counts higher again or focus on other changes like EOF and PeerDAS.
PeerDAS is the more interesting bet. It restructures how nodes sample blob data, enabling a much larger blob count without proportionally increasing node bandwidth requirements. If it ships cleanly, the blob ceiling goes up significantly without compromising decentralization. The catch: PeerDAS is technically complex, and Ethereum’s history with ambitious networking changes suggests slippage is likely.
What L2 Teams Are Actually Doing
Behind the scenes, rollup teams are hedging. Several have implemented blob-or-calldata fallback mechanisms that automatically switch posting strategies based on relative fees. It’s inelegant but functional - the kind of engineering workaround that signals a real market signal being taken seriously.
The blob fee wars aren’t a sign Ethereum is breaking. They’re a sign it’s being used.