Although Ethereum (ETH) prices have been rising recently, the network’s daily trading numbers have shifted to high gear over the past two weeks. In fact, since the ledger was launched on July 30, 2015, it has given over five of its busiest day in August alone.
August turns into a record month of Ethereum transfers
This month, Ethereum broke half of the top 10 records and handled a recent impressive wave of transfers. The climbing began in early July to mid-July, with the seven-day average increasing from 1.38 million (July 9-14) to 1.73 million per day from August 7th, with the 30-day average exceeding 1.5 million on July 31st.
Since July 29th, the network has routinely handled 170,000 to 1.9 million transactions per day. It was the top of 1.7 million cases for the eight days of the 10 days until August 7th, with a high of 1,878,031 and 1,833,756 on August 5th, and 1,833,756 on August 6th, and the weekend was 1,515,955 since August 3rd. A rare historic summit of Ethereum.
Prominent highs in the past include May 8-11, 2021 (transfers for 1st, 1st, 1st), December 9, 2022 (1,932,711), January 14, 2024 (1,961,144), and June 25, 2025 (1,751,819). January 14th, 2024 and December 9th, 2022, ranked in the top two, and August 5th, 2025, ranked 3rd. Despite this intense activity, on-chain fees remain low, hovering between 0.252 and 0.261 GWEI.
At the end of the first week of August, Ethereum block usage was 49.53% and 50% shy, with an average block size of 171 and a pending queue of 134,668 transactions. Common actions remain inexpensive. Decentralized Exchange (DEX) swaps averaged $0.38, NFT sales were $0.65, filling $0.12, borrowing $0.33.
From August 1st to 8th, the average gas price was lighter and landed between $0 and $4 for most of the time. The sharpest spikes appeared from August 5th between 14:00 and 16:00 UTC. This was the third busiest day.
The more modest bumps on the on-chain fees reappeared on August 8th, both early morning and mid-afternoon deals during the same time on August 6th. Overall, Ethereum drives transaction counts close to the record, tame the costs supported by high block usage and throughput headroom.
