Sampler Subclass Counter Shadowing Bypasses PyNvVideoCodec Decoder Limits and GPU Memory Accounting
CVE-2026-100649 · GHSA-j682-9xp5-rrf3 · published 2026-09-12 · updated 2026-09-28
- Severity
- Low
- CVSS score
- 3.7
- Chance of exploitation (EPSS)
- Not scored
- Fixed in
- 0.29.0
- Exploitation No known exploitation
- Automatable No
- Technical impact Partial
- CISA SSVC
This is a CWE-770 resource-limit and GPU-memory-accounting bypass. An unauthenticated client can select multiple stock sampler subclasses, exceed the configured process-wide PyNvVideoCodec decoder limit, and retain GPU decoder resources that were not included in the corresponding startup reservation. The amplification occurs independently in each API frontend process. Under memory-constrained deployments, the additional retained decoder resources reduce the GPU safety margin and can degrade availability or contribute to allocation failures.
Upgrade vllm to 0.29.0 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.
| Package | Registry | Affected | Fixed in |
|---|---|---|---|
| vllm | PyPI | <0.29.0 | 0.29.0 |
- HightsParticles: Unbounded HTTP Request Body Buffering in @tsparticles/mcp-server Leads to Remote DoS@tsparticles/mcp-server
- HighMCP TypeScript SDK: HTTP server transports and handlers did not limit request body size or JSON-RPC batch length@modelcontextprotocol/hono, @modelcontextprotocol/node, @modelcontextprotocol/sdk, @modelcontextprotocol/server
- HighMCP TypeScript SDK: One HTTP request within the default size limit could exhaust server memory during validation@modelcontextprotocol/sdk, @modelcontextprotocol/server
- HighHTTP server transports and OAuth endpoints read request bodies with no size limitmcp, mcp
- HighLMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loadinglmdeploy
- HighRMCP: Unauthenticated permanent session-table leak in rmcp Streamable HTTP server transport leads to remote denial-of-servicermcp
- CriticalvLLM: OpenAI auth bypassvllm
- CriticalvLLM has RCE In Video Processingvllm
- CriticalvLLM Allows Remote Code Execution via PyNcclPipe Communication Servicevllm
- CriticalvLLM Vulnerable to Remote Code Execution via Mooncake Integrationvllm
- CriticalCVE-2025-24357 Malicious model remote code execution fix bypass with PyTorch < 2.6.0vllm
- CriticalvLLM Deserialization of Untrusted Data vulnerabilityvllm
The impact, patch and workaround text comes from the advisory, shortened (published by vLLM). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.