Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐍
🐍 PyPI
Not in CISA KEV
LOW severity

GHSA-6g85-3hm8-83f9

LOWFix: tensorflow/tensorflow@20431e9

GHSA-6g85-3hm8-83f9 is a low-severity (CVSS 2.5) CWE-754 vulnerability in tensorflow. 1 public exploit reference exists, so weaponization risk is real. O3 Security confirms whether GHSA-6g85-3hm8-83f9 is actually reachable in your code before you act, and blocks exploitation at runtime until you patch.

CHECK-fail in `QuantizeAndDequantizeV4Grad`

Also known asBIT-tensorflow-2021-29544CVE-2021-29544PYSEC-2021-181PYSEC-2021-472PYSEC-2021-670
Published
May 21, 2021
Updated
Jul 8, 2026
Affected
3 pkgs
Patched
3 / 3
Exploits
1 known
Exploitation data as of Jul 8, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Real-World Exposure

3 pkgs affected
🐍tensorflow🐍tensorflow-cpu🐍tensorflow-gpu

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Impact

An attacker can trigger a denial of service via a CHECK-fail in tf.raw_ops.QuantizeAndDequantizeV4Grad:

import tensorflow as tf

gradient_tensor = tf.constant([0.0], shape=[1])
input_tensor = tf.constant([0.0], shape=[1])
input_min = tf.constant([[0.0]], shape=[1, 1])
input_max = tf.constant([[0.0]], shape=[1, 1])

tf.raw_ops.QuantizeAndDequantizeV4Grad(
  gradients=gradient_tensor, input=input_tensor,
  input_min=input_min, input_max=input_max, axis=0)

This is because the implementation does not validate the rank of the input_* tensors. In turn, this results in the tensors being passes as they are to QuantizeAndDequantizePerChannelGradientImpl:

template <typename Device, typename T>
struct QuantizeAndDequantizePerChannelGradientImpl {
  static void Compute(const Device& d,
                      typename TTypes<T, 3>::ConstTensor gradient,
                      typename TTypes<T, 3>::ConstTensor input,
                      const Tensor* input_min_tensor,
                      const Tensor* input_max_tensor,
                      typename TTypes<T, 3>::Tensor input_backprop,
                      typename TTypes<T>::Flat input_min_backprop,
                      typename TTypes<T>::Flat input_max_backprop) {
    ...
    auto input_min = input_min_tensor->vec<T>();
    auto input_max = input_max_tensor->vec<T>();
    ...
}

However, the vec<T> method, requires the rank to 1 and triggers a CHECK failure otherwise.

Patches

We have patched the issue in GitHub commit 20431e9044cf2ad3c0323c34888b192f3289af6b.

The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2 as this is the only other affected version.

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 Yakun Zhang and Ying Wang of Baidu X-Team.

Affected Packages

3 total 3 fixed
EcosystemPackageVulnerable rangeFix
🐍PyPItensorflow2.4.0&&< 2.4.22.4.2
🐍PyPItensorflow-cpu2.4.0&&< 2.4.22.4.2
🐍PyPItensorflow-gpu2.4.0&&< 2.4.22.4.2
Exploits & PoCs
1

Research use only. For defensive security, authorized penetration testing, and academic research only. Never execute exploit code against systems without explicit written authorization.

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for tensorflow. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.

  2. Fix

    Update tensorflow to 2.4.2 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-6g85-3hm8-83f9 is resolved across your whole dependency graph.

  3. Workarounds

    If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.

  4. How O3 protects you

    O3 pinpoints whether GHSA-6g85-3hm8-83f9 is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.

Tailored to GHSA-6g85-3hm8-83f9. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

### Impact An attacker can trigger a denial of service via a `CHECK`-fail in `tf.raw_ops.QuantizeAndDequantizeV4Grad`: ```python import tensorflow as tf gradient_tensor = tf.constant([0.0], shape=[1]) input_tensor = tf.constant([0.0], shape=[1]) input_min = tf.constant([[0.0]], shape=[1, 1]) input_max = tf.constant([[0.0]], shape=[1, 1]) tf.raw_ops.QuantizeAndDequantizeV4Grad( gradients=gradient_tensor, input=input_tensor, input_min=input_min, input_max=input_max, axis=0) ``` This is because the [implementation](https://github.com/tensorflow
O3 Security · Impact-Aware SCA

Is GHSA-6g85-3hm8-83f9 in your dependencies?

O3 detects GHSA-6g85-3hm8-83f9 across PyPI dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.