Don't Scrape Zillow.
Use the API.

import asyncio
from playwright.async_api import async_playwright
from bs4 import BeautifulSoup
import random, time

PROXIES = [
    "http://user:pass@45.12.55.10:10001",
    "http://user:pass@45.12.55.11:10001",
    # … hundreds more residential IPs
]
UA_POOL = [
    "Mozilla/5.0 (Windows NT 10.0; Win64) AppleWebKit/537.36 …",
    # … rotate per request
]

async def scrape(zpid):
    proxy = random.choice(PROXIES)
    async with async_playwright() as p:
        browser = await p.chromium.launch(
            proxy={"server": proxy},
            args=["--disable-blink-features=AutomationControlled"],
        )
        ctx = await browser.new_context(user_agent=random.choice(UA_POOL))
        page = await ctx.new_page()

        await page.goto(f"https://www.zillow.com/homedetails/{zpid}_zpid/")
        # check for press-and-hold captcha
        if await page.query_selector("#px-captcha"):
            await solve_captcha(page)  # ← external service, $$/solve
            await page.reload()

        await page.wait_for_selector("[data-testid='price']", timeout=15000)
        html = await page.content()
        await browser.close()

    soup = BeautifulSoup(html, "html.parser")
    # selectors break ~every 6 weeks
    price  = soup.select_one("[data-testid='price']").text
    beds   = soup.select_one("[data-testid='bed-bath-item']").text
    # … 40 more selectors
    return {"price": price, "beds": beds, ...}

# this works for ~3 weeks before bans escalate, then rebuild

curl -X POST https://api.realestateinvestingapi.com/v1/zillow \
  -H "Authorization: Bearer reia_live_••••••••" \
  -d '{"action":"propertyDetails","params":{"zpid":"29453621"}}'