Raw HyperEVM block data

Builders running a non-validating node can index the HyperEVM using data written to ~/hl/data/evm_block_and_receipts . This data is written after committed blocks are verified by the node, and therefore has no additional trust assumptions compared to running the EVM RPC directly from the node itself.

Builders that wish to index the HyperEVM without running a node can use the S3 bucket: aws s3 ls s3://hl-mainnet-evm-blocks/ --request-payer requester.

There is a similar bucket s3://hl-testnet-evm-blocks/ for testnet.

Builders interested in robustness can merge the two data sources, relying primarily on local data and falling back to S3 data.

Some potential applications include a JSON-RPC server with custom rate limits, a HyperEVM block explorer, or other indexed services and tooling for builders.

While the data is public for anyone to use, the requester must pay for data transfer costs. The filenames are predictably indexed by EVM block number, e.g. s3://hl-mainnet-evm-blocks/0/6000/6123.rmp.lz4. An indexer can copy block data from S3 on new HyperEVM blocks. The files are stored in MessagePack format and then compressed using LZ4.

Note that testnet starts with directory s3://hl-testnet-evm-blocks/18000000and the earlier testnet RPC blocks were not backfilled.

An example can be found in the Python SDK: https://github.com/hyperliquid-dex/hyperliquid-python-sdk/blob/master/examples/evm_block_indexer.py