SOFR swap NPV and cashflow different from BBG results using Python Quantlib

I'm using python quantlib to price a swap, but the NPV and cashflow result differently from BBG results as all the input controls the same, fixed schedule should be the same but floating cashflow is not. I have no idea abount what's wrong with it, can help figure out where I should modify?

code NPV result: -28451.35 BBG NPV result: -28449.30

code cashflow and BBG cashflows are:enter image description hereenter image description here

input and BBG NPV: enter image description here

curve data: enter image description here

Below is my python code.

import pandas as pd
import QuantLib as ql

calculation_date = ql.Date(22, 7, 2024)

ql.Settings.instance().evaluationDate = calculation_date

yts = ql.RelinkableYieldTermStructureHandle()
index = ql.OvernightIndex("USD Overnight Index", 0, ql.USDCurrency(), ql.UnitedStates(ql.UnitedStates.Settlement), 
                          ql.Actual360(), yts)

swaps = {
    ql.Period("1W"): 5.34063/100,
    ql.Period("2W"): 5.34045/100,
    ql.Period("3W"): 5.34350/100,
    ql.Period("1M"): 5.34750/100,
    ql.Period("2M"): 5.34040/100,
    ql.Period("3M"): 5.28600/100,
    ql.Period("4M"): 5.24215/100,
    ql.Period("5M"): 5.20170/100,
    ql.Period("6M"): 5.14835/100,
    ql.Period("7M"): 5.09080/100,
    ql.Period("8M"): 5.04535/100,
    ql.Period("9M"): 4.98930/100,
    ql.Period("10M"): 4.93130/100,
    ql.Period("11M"): 4.88620/100,
    ql.Period("12M"): 4.83670/100,
    ql.Period("18M"): 4.51801/100,
    ql.Period("2Y"): 4.33160/100,
    ql.Period("3Y"): 4.08150/100,
    ql.Period("4Y"): 3.94900/100,
    ql.Period("5Y"): 3.87846/100,
    ql.Period("6Y"): 3.84260/100,
    ql.Period("7Y"): 3.82400/100,
    ql.Period("8Y"): 3.81550/100,
    ql.Period("9Y"): 3.81350/100,
    ql.Period("10Y"): 3.81640/100,
    ql.Period("12Y"): 3.83020/100,
    ql.Period("15Y"): 3.84900/100,
    ql.Period("20Y"): 3.83529/100,
    ql.Period("25Y"): 3.75940/100,
    ql.Period("30Y"): 3.67355/100,
    ql.Period("40Y"): 3.47606/100,
    ql.Period("50Y"): 3.26680/100
}

tenors = list(swaps.keys())
quotes = [ql.SimpleQuote(swaps[k]) for k in tenors]
handles = [ql.QuoteHandle(quote) for quote in quotes]

rate_helpers = []

for quote, tenor in zip(handles, tenors):
    helper = ql.OISRateHelper(2, tenor, quote, index)
    rate_helpers.append(helper)
    

curve = ql.PiecewiseFlatForward(calculation_date, rate_helpers, ql.Actual360())
yts.linkTo(curve)

swap = ql.MakeOIS(ql.Period('10Y'), 
                  index, 0.03736141, 
                  ql.Period('0D'), 
                  swapType=-1, 
                  nominal=10000000, 
                  effectiveDate=ql.Date(24, 10, 2024), 
                  paymentLag=2)

# Set the pricing engine
engine = ql.DiscountingSwapEngine(yts)
swap.setPricingEngine(engine)

# Calculate the original NPV
npv = swap.NPV()

print(npv)

cashflows = pd.DataFrame({
    'nominal': cf.nominal(),
    'accrualStartDate': cf.accrualStartDate().ISO(),
    'accrualEndDate': cf.accrualEndDate().ISO(),
    'rate': cf.rate(),
    'amount': cf.amount()
    } for cf in map(ql.as_coupon, swap.leg(1)))

print(cashflows)