CVE-2021-41228

Code injection in `saved_model_cli`

7.5 HighFix available

TensorFlow is an open source platform for machine learning. In affected versions TensorFlow's saved_model_cli tool is vulnerable to a code injection as it calls eval on user supplied strings. This can be used by attackers to run arbitrary code on the plaform where the CLI tool runs. However, given that the tool is always run manually, the impact of this is not severe. We have patched this by adding a safe flag which defaults to True and an explicit warning for users. The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, as these are also affected and still in supported range.

Attack profile

How easy it is to exploit

Attack vector: Local
Complexity: High
Privileges needed: High
User interaction: None

What an attacker can affect

Confidentiality: High
Integrity: High
Availability: High

Exploit likelihood

0.22%Down 0.4 points since Nov 6, 2021

What each source reports

CVE List Score used7.5

Code injection in `saved_model_cli`

Record: CVE-2021-41228
CVSS 3.1: 7.5 HighCVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H
Weakness: CWE-78
Updated: 2026-11-02

EUVDEUVD-2021-04507.5

Record: EUVD-2021-0450
CVSS 3.1: 7.5 HighCVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H
Updated: 2026-11-03

GitHub AdvisoriesGHSA-3rcw-9p9x-582v5.0

Code injection in `saved_model_cli`

Impact

TensorFlow's saved_model_cli tool is vulnerable to a code injection as it calls eval on user supplied strings

def preprocess_input_exprs_arg_string(input_exprs_str):
  ... 
  for input_raw in filter(bool, input_exprs_str.split(';')):
    ...
    input_key, expr = input_raw.split('=', 1)
    input_dict[input_key] = eval(expr)
  ...

This can be used by attackers to run arbitrary code on the plaform where the CLI tool runs. However, given that the tool is always run manually, the impact of this is not severe. We have patched this by adding a safe flag which defaults to True and an explicit warning for users.

Patches

We have patched the issue in GitHub commit 8b202f08d52e8206af2bdb2112a62fafbc546ec7.

The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, as these are also affected and still in supported range.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by Omer Kaspi from Vdoo.

Record: GHSA-3rcw-9p9x-582v
CVSS 4.0: 5.0 MediumCVSS:4.0/AV:L/AC:L/AT:P/PR:H/UI:N/VC:N/VI:N/VA:N/SC:H/SI:H/SA:H
CVSS 3.1: 7.5 HighCVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H
Weakness: CWE-78, CWE-94
Updated: 2026-11-02

NVD7.5

Record: CVE-2021-41228
CVSS 3.1: 7.5 HighCVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H
CVSS 3.1: 7.8 HighCVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
CVSS 2.0: 4.6 MediumAV:L/AC:L/Au:N/C:P/I:P/A:P
Weakness: CWE-78, CWE-94
Updated: 2026-11-02

OSV7.8

Record: CVE-2021-41228
CVSS 3.1: 7.8 HighCVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Updated: 2026-11-02

OSVGHSA-3rcw-9p9x-582v5.0

Code injection in `saved_model_cli`

Impact

TensorFlow's saved_model_cli tool is vulnerable to a code injection as it calls eval on user supplied strings

def preprocess_input_exprs_arg_string(input_exprs_str):
  ... 
  for input_raw in filter(bool, input_exprs_str.split(';')):
    ...
    input_key, expr = input_raw.split('=', 1)
    input_dict[input_key] = eval(expr)
  ...

Patches

We have patched the issue in GitHub commit 8b202f08d52e8206af2bdb2112a62fafbc546ec7.

The fix will be included in TensorFlow 2.7.0. We will also cherrypick this commit on TensorFlow 2.6.1, TensorFlow 2.5.2, and TensorFlow 2.4.4, as these are also affected and still in supported range.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by Omer Kaspi from Vdoo.

Record: GHSA-3rcw-9p9x-582v
CVSS 4.0: 5.0 MediumCVSS:4.0/AV:L/AC:L/AT:P/PR:H/UI:N/VC:N/VI:N/VA:N/SC:H/SI:H/SA:H
CVSS 3.1: 7.5 HighCVSS:3.1/AV:L/AC:H/PR:H/UI:N/S:C/C:H/I:H/A:H
Weakness: CWE-78, CWE-94
Updated: 2026-11-02

Show 5 moreShow less

OSVBIT-tensorflow-2021-41228No score

Code injection in `saved_model_cli`

Record: BIT-tensorflow-2021-41228
Updated: 2026-11-02

OSVopenSUSE-SU-2024:12116-1No score

tensorflow-lite-2.9.1-1.1 on GA media

These are all security issues fixed in the tensorflow-lite-2.9.1-1.1 package on the GA media of openSUSE Tumbleweed.

Record: openSUSE-SU-2024:12116-1
Updated: 2026-11-02

OSVPYSEC-2021-420No score

Record: PYSEC-2021-420
Updated: 2026-11-02

OSVPYSEC-2021-637No score

Record: PYSEC-2021-637
Updated: 2026-11-02

OSVPYSEC-2021-835No score

Record: PYSEC-2021-835
Updated: 2026-11-02

Affected products

Affected packages

Affected packages
Package	Affected	Fixed
`tensorflow` PyPI	2.5.0 up to 2.5.2	2.5.2
`tensorflow` PyPI	before 2.4.4	2.4.4
`tensorflow` PyPI	2.6.0 up to 2.6.1	2.6.1
`tensorflow-cpu` PyPI	2.5.0 up to 2.5.2	2.5.2
`tensorflow-cpu` PyPI	before 2.4.4	2.4.4
`tensorflow-cpu` PyPI	2.6.0 up to 2.6.1	2.6.1
`tensorflow-gpu` PyPI	2.5.0 up to 2.5.2	2.5.2
`tensorflow-gpu` PyPI	before 2.4.4	2.4.4
`tensorflow-gpu` PyPI	2.6.0 up to 2.6.1	2.6.1
`tensorflow` Bitnami	2.4.0 up to 2.4.4, 2.5.0 up to 2.5.2, 2.6.0 up to 2.6.1	2.4.42.5.22.6.1

Show 9 moreShow less

Affected packages
Package	Affected	Fixed
`tensorflow` PyPI	2.5.0 up to 2.5.2, plus listed versions 2.5.1	2.5.2
`tensorflow` PyPI	before 2.4.4, plus listed versions 0.12.00.12.11.0.01.0.11.1.0 Show 45 moreShow less 1.10.01.10.11.11.01.12.01.12.21.12.31.13.11.13.21.14.01.15.01.15.21.15.31.15.41.15.51.2.01.2.11.3.01.4.01.4.11.5.01.5.11.6.01.7.01.7.11.8.01.9.02.0.02.0.12.0.22.0.32.0.42.1.02.1.12.1.22.1.32.1.42.2.02.2.12.2.22.2.32.3.02.3.12.3.22.3.32.3.4	2.4.4
`tensorflow-cpu` PyPI	2.5.0 up to 2.5.2, plus listed versions 2.5.1	2.5.2
`tensorflow-cpu` PyPI	before 2.4.4, plus listed versions 1.15.02.1.02.1.12.1.22.1.3 Show 14 moreShow less 2.1.42.2.02.2.12.2.22.2.32.3.02.3.12.3.22.3.32.3.42.4.02.4.12.4.22.4.3	2.4.4
`tensorflow-gpu` PyPI	2.5.0 up to 2.5.2, plus listed versions 2.5.1	2.5.2
`tensorflow-gpu` PyPI	before 2.4.4, plus listed versions 0.12.00.12.11.0.01.0.11.1.0 Show 45 moreShow less 1.10.01.10.11.11.01.12.01.12.21.12.31.13.11.13.21.14.01.15.01.15.21.15.31.15.41.15.51.2.01.2.11.3.01.4.01.4.11.5.01.5.11.6.01.7.01.7.11.8.01.9.02.0.02.0.12.0.22.0.32.0.42.1.02.1.12.1.22.1.32.1.42.2.02.2.12.2.22.2.32.3.02.3.12.3.22.3.32.3.4	2.4.4
`tensorflow` PyPI	before 2.4.4, 2.5.0 up to 2.5.2, 2.6.0 up to 2.6.1, 2.7.0rc0 up to 2.7.0, plus listed versions 0.12.00.12.11.0.01.0.11.1.0 Show 45 moreShow less 1.10.01.10.11.11.01.12.01.12.21.12.31.13.11.13.21.14.01.15.01.15.21.15.31.15.41.15.51.2.01.2.11.3.01.4.01.4.11.5.01.5.11.6.01.7.01.7.11.8.01.9.02.0.02.0.12.0.22.0.32.0.42.1.02.1.12.1.22.1.32.1.42.2.02.2.0rc02.2.0rc12.2.0rc22.2.0rc32.2.0rc42.2.12.2.22.2.3	2.4.42.5.22.6.12.7.0
`tensorflow-cpu` PyPI	before 2.4.4, 2.5.0 up to 2.5.2, 2.6.0 up to 2.6.1, 2.7.0rc0 up to 2.7.0, plus listed versions 1.15.02.1.02.1.12.1.22.1.3 Show 16 moreShow less 2.1.42.2.02.2.12.2.22.2.32.3.02.3.12.3.22.3.32.3.42.4.02.4.12.4.22.4.32.5.12.7.0rc1	2.4.42.5.22.6.12.7.0
`tensorflow-gpu` PyPI	before 2.4.4, 2.5.0 up to 2.5.2, 2.6.0 up to 2.6.1, 2.7.0rc0 up to 2.7.0, plus listed versions 0.12.00.12.11.0.01.0.11.1.0 Show 45 moreShow less 1.10.01.10.11.11.01.12.01.12.21.12.31.13.11.13.21.14.01.15.01.15.21.15.31.15.41.15.51.2.01.2.11.3.01.4.01.4.11.5.01.5.11.6.01.7.01.7.11.8.01.9.02.0.02.0.12.0.22.0.32.0.42.1.02.1.12.1.22.1.32.1.42.2.02.2.12.2.22.2.32.3.02.3.12.3.22.3.32.3.4	2.4.42.5.22.6.12.7.0

History

History
Recorded	Changed	Sources
Nov 6, 2026	facts
Nov 3, 2026	aliases	EUVD
Nov 2, 2026	first indexed	CVE List, GitHub Advisories, NVD, OSV

Dates are when Every CVE recorded each change. Upstream dates may differ.

In short

CVE-2021-41228 is a high vulnerability (CVSS 7.5) in Google Tensorflow, published November 5, 2021. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.22%, the 12th percentile. Fixed in tensorflow 2.5.2.

References

Show 165 moreShow less

This page doesn't list everything

Left out here: 33 versions. The JSON record has them.