bs4.FeatureNotFound: Couldn't find a tree builder with the features you requested: lxml. Do you need to install a parser library? (Python) — how to fix
Traceback (most recent call last):
File "scrape.py", line 6, in <module>
soup = BeautifulSoup(html, "lxml")
File ".../site-packages/bs4/__init__.py", line 250, in __init__
raise FeatureNotFound(
bs4.FeatureNotFound: Couldn't find a tree builder with the features you requested: lxml. Do you need to install a parser library?
Quick fix: Install the parser into the same interpreter: python -m pip install lxml (or html5lib). Or switch to the built-in parser, which needs nothing extra: BeautifulSoup(html, "html.parser"). For XML use "xml" / "lxml-xml", which also requires lxml.
What it means
The second argument to BeautifulSoup(markup, features) selects a parser:
| features | Library needed | Notes |
|---|---|---|
"html.parser" |
none (standard library) | decent speed, always available |
"lxml" |
lxml |
fastest, lenient |
"xml" / "lxml-xml" |
lxml |
the only XML parser bs4 supports |
"html5lib" |
html5lib |
parses like a browser, slow |
FeatureNotFound means bs4 looked for a builder matching your feature string and none was registered, because the library behind it isn't importable.
Fix 1: install the parser in the right environment
python -m pip install lxml
# or
python -m pip install html5lib
Use the same python that runs your script (activate your venv first). In Jupyter, run %pip install lxml in a cell and restart the kernel — bs4 checks available builders when it's imported, so an already-running kernel won't see the new parser.
Check:
python -c "import lxml, bs4; print(lxml.__version__, bs4.__version__)"
With conda: conda install lxml.
Fix 2: use the built-in parser
If you don't need lxml specifically:
from bs4 import BeautifulSoup
soup = BeautifulSoup(html, "html.parser")
This removes the dependency entirely, which is handy for serverless functions and minimal containers. Results can differ slightly on broken HTML, so pick one parser and use it consistently.
Fix 3: parsing XML
BeautifulSoup(data, "xml") needs lxml — html.parser can't do XML properly (it lowercases tags and mishandles namespaces). Install lxml, or use the standard library for simple XML:
import xml.etree.ElementTree as ET
root = ET.fromstring(data)
Fix 4: lxml fails to install
On new Python versions or unusual platforms (Alpine, ARM), pip may try to build lxml from source and fail with missing libxml2 / libxslt headers. Upgrade pip first so it finds a prebuilt wheel:
python -m pip install --upgrade pip
python -m pip install lxml
If you must build: Debian/Ubuntu sudo apt install libxml2-dev libxslt1-dev python3-dev; Alpine apk add libxml2-dev libxslt-dev gcc musl-dev. See failed building wheel.
Fix 5: typo in the feature string
"lxml ", "html", "html5" or "parser" don't match any builder and raise the same error. Valid values are exactly those in the table above.
pandas.read_html variant
pandas.read_html raises ImportError: lxml not found, please install it (or asks for html5lib + bs4). Same fix: python -m pip install lxml, or pass flavor="bs4" with bs4 and html5lib installed.
Last updated 2026-11-13.