CVE-2025-66019
pypdf manipulated LZWDecode streams can exhaust RAM
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
Attack profile
How easy it is to exploit
- Attack vector
- Network
- Complexity
- Low
- Privileges needed
- None
- User interaction
- None
What an attacker can affect
- Confidentiality
- None
- Integrity
- None
- Availability
- High
Exploit likelihood
What each source reports
CVE List Score used
6.6
pypdf manipulated LZWDecode streams can exhaust RAM
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
- Record
CVE-2025-66019- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Weakness
- CWE-400, CWE-409
- Updated
- 2026-11-02
EUVDEUVD-2025-199017
6.6
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
- Record
EUVD-2025-199017- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Updated
- 2026-11-02
GitHub AdvisoriesGHSA-m449-cwjh-6pw7
6.6
pypdf's LZWDecode streams be manipulated to exhaust RAM
Impact
An attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter.
This is a follow up to GHSA-jfx9-29x2-rv3j to align the default limit with the one for zlib.
Patches
This has been fixed in pypdf==6.4.0.
Workarounds
If users cannot upgrade yet, use the line below to overwrite the default in their code:
pypdf.filters.LZW_MAX_OUTPUT_LENGTH = 75_000_000
- Record
GHSA-m449-cwjh-6pw7- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Weakness
- CWE-400, CWE-409
- Updated
- 2026-11-02
NVD
6.6
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
- Record
CVE-2025-66019- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X - Weakness
- CWE-400, CWE-409
- Updated
- 2026-11-02
OSV
6.6
pypdf manipulated LZWDecode streams can exhaust RAM
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
- Record
CVE-2025-66019- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Weakness
- CWE-400, CWE-409
- Updated
- 2026-11-02
OSVGHSA-m449-cwjh-6pw7
6.6
pypdf's LZWDecode streams be manipulated to exhaust RAM
Impact
An attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter.
This is a follow up to GHSA-jfx9-29x2-rv3j to align the default limit with the one for zlib.
Patches
This has been fixed in pypdf==6.4.0.
Workarounds
If users cannot upgrade yet, use the line below to overwrite the default in their code:
pypdf.filters.LZW_MAX_OUTPUT_LENGTH = 75_000_000
- Record
GHSA-m449-cwjh-6pw7- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Weakness
- CWE-400, CWE-409
- Updated
- 2026-11-02
Show 4 moreShow less
OSVPYSEC-2026-1832
6.6
pypdf's LZWDecode streams be manipulated to exhaust RAM
Impact
An attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter.
This is a follow up to GHSA-jfx9-29x2-rv3j to align the default limit with the one for zlib.
Patches
This has been fixed in pypdf==6.4.0.
Workarounds
If users cannot upgrade yet, use the line below to overwrite the default in their code:
pypdf.filters.LZW_MAX_OUTPUT_LENGTH = 75_000_000
- Record
PYSEC-2026-1832- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Updated
- 2026-11-02
OSVUBUNTU-CVE-2025-66019
6.6
pypdf is a free and open-source pure-python PDF library. Prior to version 6.4.0, an attacker who uses this vulnerability can craft a PDF which leads to a memory usage of up to 1 GB per stream. This requires parsing the content stream of a page using the LZWDecode filter. This issue has been patched in version 6.4.0.
- Record
UBUNTU-CVE-2025-66019- CVSS 4.0
- 6.6 Medium
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:U - Updated
- 2026-11-02
OSVCLSA-2026-1787843313 No score
TuxCare security update for pypdf (10 CVEs)
TuxCare rebuilt pypdf to address 10 CVEs. Fixed in 5.9.0.post16+tuxcare.
- Record
CLSA-2026-1787843313- Updated
- 2026-11-02
OSVopenSUSE-SU-2025:15776-1 No score
python311-pypdf-6.4.0-1.1 on GA media
These are all security issues fixed in the python311-pypdf-6.4.0-1.1 package on the GA media of openSUSE Tumbleweed.
- Record
openSUSE-SU-2025:15776-1- Updated
- 2026-11-02
Affected products
Affected packages
| Package | Affected | Fixed |
|---|---|---|
pypdfPyPI |
before 6.4.0 | 6.4.0 |
pypdfTuxCare:PyPI From an advisory covering 10 CVEs |
5.9.0 up to 5.9.0.post16+tuxcare, plus listed versions 5.9.0.post1+tuxcare5.9.0.post10+tuxcare5.9.0.post11+tuxcare5.9.0.post12+tuxcare5.9.0.post13+tuxcare Show 10 moreShow less5.9.0.post14+tuxcare5.9.0.post15+tuxcare5.9.0.post2+tuxcare5.9.0.post3+tuxcare5.9.0.post4+tuxcare5.9.0.post5+tuxcare5.9.0.post6+tuxcare5.9.0.post7+tuxcare5.9.0.post8+tuxcare5.9.0.post9+tuxcare |
5.9.0.post16+tuxcare |
pypdfPyPI |
before 6.4.0, plus listed versions 1.01.11.101.111.12 Show 45 moreShow less1.131.21.31.41.51.61.71.81.93.1.03.10.03.11.03.11.13.12.03.12.13.12.23.13.03.14.03.15.03.15.13.15.23.15.33.15.43.15.53.16.03.16.13.16.23.16.33.16.43.17.03.17.13.17.23.17.33.17.43.2.03.2.13.3.03.4.03.4.13.5.03.5.13.5.23.6.03.7.03.7.1 |
6.4.0 |
pypdfUbuntu:24.04:LTS |
All versions, plus listed versions 3.4.1-13.17.4-14.0.0-14.0.1-14.0.2-1 |
None known |
pypdfUbuntu:25.10 |
All versions, plus listed versions 4.3.1-15.4.0-1 |
None known |
pypdfUbuntu:26.10 |
All versions, plus listed versions 6.9.2-16.9.2-2 |
None known |
pypdfUbuntu:26.04:LTS |
All versions, plus listed versions 5.4.0-16.9.2-1 |
None known |
python-pypdfopenSUSE:Tumbleweed |
before 6.4.0-1.1 | 6.4.0-1.1 |
History
| Recorded | Changed | Sources |
|---|---|---|
| affected, facts | ||
| first indexed | CVE List, EUVD, GitHub Advisories, NVD, OSV |
Dates are when Every CVE recorded each change. Upstream dates may differ.
In short
CVE-2025-66019 is a medium vulnerability (CVSS 6.6) in Py-pdf Pypdf, published November 24, 2025. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.39%, the 33rd percentile. Fixed in pypdf 6.4.0.
References
github.com/py-pdf/pypdf/security/advisories/GHSA-m449-cwjh-6pw7
github.com/py-pdf/pypdf/commit/96186725e5e6f237129a58a97cd19204a9ce40b2
github.com/py-pdf/pypdf/releases/tag/6.4.0
aydinnyunus.github.io/2025/12/20/cve-2025-66019-pypdf-lzw-dos
github.com/py-pdf/pypdf/security/advisories/GHSA-jfx9-29x2-rv3j
nvd.nist.gov/vuln/detail/CVE-2025-66019
Left out here: 31 versions. The JSON record has them.