CVE-2025-46560

A critical performance vulnerability has been identified in the input preprocessing logic of the multimodal tokenizer. The code dynamically replaces placeholder tokens (e.g., <|audio_*|>, <|image_*|>) with repeated tokens based on precomputed lengths. Due to ​​inefficient list concatenation operations​​, the algorithm exhibits ​​quadratic time complexity (O(n²))​​, allowing malicious actors to trigger resource exhaustion via specially crafted inputs.

Details

​​Affected Component​​: input_processor_for_phi4mm function.
https://github.com/vllm-project/vllm/blob/8cac35ba435906fb7eb07e44fe1a8c26e8744f4e/vllm/model_executor/models/phi4mm.py#L1182-L1197

The code modifies the input_ids list in-place using input_ids = input_ids[:i] + tokens + input_ids[i+1:]. Each concatenation operation copies the entire list, leading to O(n) operations per replacement. For k placeholders expanding to m tokens, total time becomes O(kmn), approximating O(n²) in worst-case scenarios.

PoC

Test data demonstrates exponential time growth:

test_cases = [100, 200, 400, 800, 1600, 3200, 6400]
run_times = [0.002, 0.007, 0.028, 0.136, 0.616, 2.707, 11.854]  # seconds

Doubling input size increases runtime by ~4x (consistent with O(n²)).

Remediation Recommendations​

Precompute all placeholder positions and expansion lengths upfront.
Replace dynamic list concatenation with a single preallocated array.

# Pseudocode for O(n) solution
new_input_ids = []
for token in input_ids:
    if token is placeholder:
        new_input_ids.extend([token] * precomputed_length)
    else:
        new_input_ids.append(token)

​​Denial-of-Service (DoS):​​ An attacker could submit inputs with many placeholders (e.g., 10,000 <|audio_1|> tokens), causing CPU/memory exhaustion.
Example: 10,000 placeholders → ~100 million operations.

A regular expression with worst-case exponential or polynomial matching time is applied to untrusted input, causing excessive CPU use. Typical impact: denial of service when input is crafted to trigger backtracking.

CVE-2025-46560 has a CVSS score of 6.5 (Medium). The vector is network-reachable, low privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. A fixed version is available (0.8.5); upgrading removes the vulnerable code path.

Kodem intelligence

1. What is CVE-2025-46560? CVE-2025-46560 is a medium-severity inefficient regular expression (ReDoS) vulnerability in vllm (pip), affecting versions >= 0.8.0, < 0.8.5. It is fixed in 0.8.5. A regular expression with worst-case exponential or polynomial matching time is applied to untrusted input, causing excessive CPU use.
2. How severe is CVE-2025-46560? CVE-2025-46560 has a CVSS score of 6.5 (Medium). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
3. Which versions of vllm are affected by CVE-2025-46560? vllm (pip) versions >= 0.8.0, < 0.8.5 is affected.
4. Is there a fix for CVE-2025-46560? Yes. CVE-2025-46560 is fixed in 0.8.5. Upgrade to this version or later.
5. Is CVE-2025-46560 exploitable, and should I be worried? Whether CVE-2025-46560 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
6. What actually determines whether CVE-2025-46560 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
7. How do I fix CVE-2025-46560? Upgrade vllm to 0.8.5 or later.