A speed claim a reader cannot reproduce is a screenshot. This harness
installs manifoldbt from PyPI like any user would, generates its own data,
and gates every timing behind a parity check: a workload where the two
engines disagree publishes nothing and fails the run.
It lives here rather than in the engine repository because it benchmarks the
published wheel, not the source. Anyone can fork this repository and press
"Run workflow" to get the same table on their own runner.
The workflow runs on demand, weekly, and on every published release, so a
version that gets slower says so in public.
Add the community Discord to the header links, and switch the Quick Start from
mbt.DataStore(data_root="data") (which assumes pre-ingested data) to
mbt.ingest(provider="binance", ...) so it runs as-is. Verified on 0.8.0.
Source-available rather than fully OSI open-source. Use, modify and
self-host remain free. Reselling the software or running it as a paid
hosted service is now explicitly disallowed by the Commons Clause.
- Cross-exchange backtesting (Pro)
- Dict universe format (provider-based symbol resolution)
- Exogenous data support (register_exo + exo() expressions)
- Provider-based data layout (binance/1h/TICKER.arrow)
- Preload fix for provider layout
- Exo column resampling for multi-resolution
- Pro gate for cross-exchange (clean exit)
- ATR/ADX rolling SMA fix
- Precise mode hybrid fills
- Add ingest() function for downloading bar data from providers
- Add manifoldbt CLI entry point (manifoldbt ingest ...)
- Supported providers: binance, hyperliquid (free), databento, massive (Pro)
- Update type stubs for py_ingest
github-actions[bot]