Datasheet · A stack of parts
Calculator tool + A penalty threat
calculator + You will be penalized if you get this wrong.
Helps on DeFi maths, Haiku 5.5 with thinking off (+40.8 points, replicated).
DeFi maths, Haiku 5.5 with thinking off
Pooled across 2 notebooks: +40.8 points, 95% interval +36.9 to +44.8; replicated.
- EXCITESNotebook #076+40.7 points (+35.3 to +46.2), 432 pairs, cost ×1.46signed verdict 478
- EXCITESNotebook #078+41.7 points (+25.0 to +58.3), 48 pairs, cost ×1.46, stopped early, provisionalsigned verdict 486
Where it acts
It helps on 11 of the 12 families it was read on, most on Two-hop swaps (+96.3), Bonding-curve buys (+95.0), Swaps (+86.3) and 8 more.
- Two-hop swaps+96.3+90.0 to +100.040 tasks
- Bonding-curve buys+95.0+87.5 to +100.040 tasks
- Swaps+86.3+75.0 to +96.340 tasks
- Average entry price+52.5+35.0 to +70.040 tasks
- Profit after fees+46.3+30.0 to +63.840 tasks
- APR to APY+32.5+15.0 to +51.340 tasks
- Liquidation prices+28.8+12.5 to +48.840 tasks
- Token decimals+17.5+3.8 to +35.040 tasks
- Impermanent loss+13.8+3.8 to +26.340 tasks
- Fee splits+11.3+2.5 to +21.340 tasks
- Position size from risk+10.0+1.3 to +23.840 tasks
The other 1 family, none moved clearly
- Funding payments+0.0+0.0 to +0.040 tasks
Read from 480 paired tasks in 2 notebooks, each task against its own notebook’s two controls, with every interval widened for the 12 families read. No notebook preregisters a family: this says where to look, and a notebook says whether it holds. The runs behind it are on each notebook’s page.
Put this datasheet in a README
The badge shows what the signed record says about this part, in the setting measured most, and links here. It follows the record as new notebooks publish. Add ?suite= and a suite’s id to show another setting.
[](https://opsinlab.io/parts/d6a1a0299760a8f7)Measure it in a setting it has not been measured in: propose it. As JSON for an agent: /v1/parts/d6a1a0299760a8f7. Every part: the atlas.