{"id":"GHSA-hf3c-wxg2-49q9","aliases":[],"status":"published","title":"vLLM vulnerable to Denial of Service by abusing xgrammar cache","description":"### Impact\n\nThis report is to highlight a vulnerability in XGrammar, a library used by the structured output feature in vLLM. The XGrammar advisory is here: https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3\n\nThe [xgrammar](https://xgrammar.mlc.ai/docs/) library is the default backend used by vLLM to support structured output (a.k.a. guided decoding). Xgrammar provides a required, built-in cache for its compiled grammars stored in RAM. xgrammar is available by default through the OpenAI compatible API server with both the V0 and V1 engines.\n\nA malicious user can send a stream of very short decoding requests with unique schemas, resulting in an addition to the cache for each request. This can result in a Denial of Service by consuming all of the system's RAM.\n\nNote that even if vLLM was configured to use a different backend by default, it is still possible to choose xgrammar on a per-request basis using the `guided_decoding_backend` key of the `extra_body` field of the request with the V0 engine. This per-request choice is not available when using the V1 engine. \n### Patches\n\n* https://github.com/vllm-project/vllm/pull/16283\n\n### Workarounds\n\nThere is no way to workaround this issue in existing versions of vLLM other than preventing untrusted access to the OpenAI compatible API server.\n\n### References\n\n* https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3","severity":{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","level":"medium","from":"ghsa"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-1395","CWE-770"],"published":"2025-04-15T21:21:04Z","modified":"2025-04-15T21:21:06Z","fixes":[{"ecosystem":"PyPI","package":"vllm","fixed_in":["0.8.4"],"sources":["ghsa","osv"]}],"affected":[{"source":"ghsa","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.6.5"},{"fixed":"0.8.4"}]}]},{"source":"osv","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0.6.5"},{"fixed":"0.8.4"}]}],"versions":["0.6.5","0.6.6","0.6.6.post1","0.7.0","0.7.1","0.7.2","0.7.3","0.8.0","0.8.1","0.8.2","0.8.3"]}],"references":[{"url":"https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3","tags":["web"]},{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-hf3c-wxg2-49q9","tags":["web"]},{"url":"https://github.com/vllm-project/vllm/pull/16283","tags":["web"]},{"url":"https://github.com/vllm-project/vllm/commit/cb84e45ac75b42ba6795145923e8eb323bb825ad","tags":["web"]},{"url":"https://github.com/advisories/GHSA-hf3c-wxg2-49q9"},{"url":"https://github.com/vllm-project/vllm","tags":["package"]}],"sources":{"ghsa":[{"id":"GHSA-hf3c-wxg2-49q9","title":"vLLM vulnerable to Denial of Service by abusing xgrammar cache","description":"### Impact\n\nThis report is to highlight a vulnerability in XGrammar, a library used by the structured output feature in vLLM. The XGrammar advisory is here: https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3\n\nThe [xgrammar](https://xgrammar.mlc.ai/docs/) library is the default backend used by vLLM to support structured output (a.k.a. guided decoding). Xgrammar provides a required, built-in cache for its compiled grammars stored in RAM. xgrammar is available by default through the OpenAI compatible API server with both the V0 and V1 engines.\n\nA malicious user can send a stream of very short decoding requests with unique schemas, resulting in an addition to the cache for each request. This can result in a Denial of Service by consuming all of the system's RAM.\n\nNote that even if vLLM was configured to use a different backend by default, it is still possible to choose xgrammar on a per-request basis using the `guided_decoding_backend` key of the `extra_body` field of the request with the V0 engine. This per-request choice is not available when using the V1 engine. \n### Patches\n\n* https://github.com/vllm-project/vllm/pull/16283\n\n### Workarounds\n\nThere is no way to workaround this issue in existing versions of vLLM other than preventing untrusted access to the OpenAI compatible API server.\n\n### References\n\n* https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3","status":"published","published":"2025-04-15T21:21:04Z","modified":"2025-04-15T21:21:06Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","level":"medium"}],"cwes":["CWE-770","CWE-1395"],"updated_at":"2026-11-02T14:38:39.85223Z"}],"osv":[{"id":"GHSA-hf3c-wxg2-49q9","title":"vLLM vulnerable to Denial of Service by abusing xgrammar cache","description":"### Impact\n\nThis report is to highlight a vulnerability in XGrammar, a library used by the structured output feature in vLLM. The XGrammar advisory is here: https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3\n\nThe [xgrammar](https://xgrammar.mlc.ai/docs/) library is the default backend used by vLLM to support structured output (a.k.a. guided decoding). Xgrammar provides a required, built-in cache for its compiled grammars stored in RAM. xgrammar is available by default through the OpenAI compatible API server with both the V0 and V1 engines.\n\nA malicious user can send a stream of very short decoding requests with unique schemas, resulting in an addition to the cache for each request. This can result in a Denial of Service by consuming all of the system's RAM.\n\nNote that even if vLLM was configured to use a different backend by default, it is still possible to choose xgrammar on a per-request basis using the `guided_decoding_backend` key of the `extra_body` field of the request with the V0 engine. This per-request choice is not available when using the V1 engine. \n### Patches\n\n* https://github.com/vllm-project/vllm/pull/16283\n\n### Workarounds\n\nThere is no way to workaround this issue in existing versions of vLLM other than preventing untrusted access to the OpenAI compatible API server.\n\n### References\n\n* https://github.com/mlc-ai/xgrammar/security/advisories/GHSA-389x-67px-mjg3","status":"published","published":"2025-04-15T21:21:04Z","modified":"2025-04-15T21:21:04Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","level":"medium"}],"cwes":["CWE-1395","CWE-770"],"updated_at":"2026-11-02T15:46:48.920784Z"}]}}