Allocation of Resources Without Limits or Throttling Affecting vllm package, versions [0.22.0,0.24.0)
Threat Intelligence
The probability is the direct output of the EPSS model, and conveys an overall sense of the threat of exploitation in the wild. The percentile measures the EPSS probability relative to all known EPSS scores. Note: This data is updated daily, relying on the latest available EPSS model version. Check out the EPSS documentation for more details.
Do your applications use this vulnerable package?
In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.
Test your applications- Snyk IDSNYK-PYTHON-VLLM-21707819
- published25 Oct 2026
- disclosed25 Oct 2026
- creditRex Liu
Introduced: 25 Oct 2026
NewCVE-2026-116172 (opens in a new tab)How to fix?
Upgrade vllm to version 0.24.0 or higher.
Overview
vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs
Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling in the Rust gRPC converter (rust/src/server/src/grpc/convert.rs) via the CandidateTokens.top_n field in ResponseOptions.output_candidates and ResponseOptions.prompt_candidates. The converter casts the protobuf uint32 top_n value directly to a signed i32 without bounds checking and forwards it to EngineCoreSamplingParams.logprobs or EngineCoreSamplingParams.prompt_logprobs without enforcing the configured max_logprobs limit that the Python API path applies. An attacker can send a Generate or GenerateStream request with an oversized top_n (e.g. 1,000,000) or an integer-wrapping value (e.g. 2,147,483,648) to cause torch.topk to fail or trigger an assertion in the V1 sampler, terminating EngineCore.
Note: This is only exploitable when the Rust gRPC endpoint is network-reachable, and the impact is limited to deployments where max_logprobs has not been explicitly set to -1 (no-cap mode).