https://cve.radiocsirt.org Contains only the most 10 recent entries. http://www.rssboard.org/rss-specification python-feedgen en Thu, 08 Oct 2026 17:50:59 +0000 https://cve.radiocsirt.org/vuln/bdu:2025-11321 bdu:2025-11321 https://cve.radiocsirt.org/vuln/bdu:2025-11321 https://cve.radiocsirt.org/vuln/euvd-2026-242243 EUVD-2026-242243 https://cve.radiocsirt.org/vuln/euvd-2026-242243 https://cve.radiocsirt.org/vuln/fkie_cve-2025-48942 <p>vLLM is an inference and serving engine for large language models (LLMs). In versions 0.8.0 up to but excluding 0.9.0, hitting the /v1/completions API with a invalid json_schema as a Guided Param kills the vllm server. This vulnerability is similar GHSA-9hcf-v7m4-6m2j/CVE-2025-48943, but for regex instead of a JSON schema. Version 0.9.0 fixes the issue.</p> <p>vLLM is an inference and serving engine for large language models (LLMs). In versions 0.8.0 up to but excluding 0.9.0, hitting the /v1/completions API with a invalid json_schema as a Guided Param kills the vllm server. This vulnerability is similar GHSA-9hcf-v7m4-6m2j/CVE-2025-48943, but for regex instead of a JSON schema. Version 0.9.0 fixes the issue.</p> https://cve.radiocsirt.org/vuln/fkie_cve-2025-48942 https://cve.radiocsirt.org/vuln/ghsa-6qc9-v4r8-22xg <p><strong>Affected:</strong> PyPI: vllm</p> <p>### Summary Hitting the /v1/completions API with a invalid json_schema as a Guided Param will kill the vllm server</p> <p>### Details The following API call `(venv) [derekh@ip-172-31-15-108 ]$ curl -s http://localhost:8000/v1/completions -H &#34;Content-Type: application/json&#34; -d &#39;{&#34;model&#34;: &#34;meta-llama/Llama-3.2-3B-Instruct&#34;,&#34;prompt&#34;: &#34;Name two great reasons to visit Sligo &#34;, &#34;max_tokens&#34;: 10, &#34;temperature&#34;: 0.5, &#34;guided_json&#34;:&#34;{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}&#34;}&#39; ` will provoke a Uncaught exceptions from xgrammer in `./lib64/python3.11/site-packages/xgrammar/compiler.py `</p> <p>Issue with more information: https://github.com/vllm-project/vllm/issues/17248</p> <p>### PoC Make a call to vllm with invalid json_scema e.g. `{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}`</p> <p>`curl -s http://localhost:8000/v1/completions -H &#34;Content-Type: application/json&#34; -d &#39;{&#34;model&#34;: &#34;meta-llama/Llama-3.2-3B-Instruct&#34;,&#34;prompt&#34;: &#34;Name two great reasons to visit Sligo &#34;, &#34;max_tokens&#34;: 10, &#34;temperature&#34;: 0.5, &#34;guided_json&#34;:&#34;{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}&#34;}&#39; ` ### Impact vllm crashes</p> <p>example traceback ``` ERROR 03-26 17:25:01 [core.py:340] EngineCore hit an exception: Traceback (most recent call last): ERROR 03-26 17:25:01 [core.py:340] File &#34;/home/derekh/workarea/vllm/vllm/v1/engine/core.py&#34;, line 333, in run_engine_core ERROR 03-26 17:25:01 [core.py:340] engine_core.run_busy_loop() ERROR 03-26 17:25:01 [core.py:340] File &#34;/home/derekh/workarea/vllm/vllm…</p> <p><strong>Affected:</strong> PyPI: vllm</p> <p>### Summary Hitting the /v1/completions API with a invalid json_schema as a Guided Param will kill the vllm server</p> <p>### Details The following API call `(venv) [derekh@ip-172-31-15-108 ]$ curl -s http://localhost:8000/v1/completions -H &#34;Content-Type: application/json&#34; -d &#39;{&#34;model&#34;: &#34;meta-llama/Llama-3.2-3B-Instruct&#34;,&#34;prompt&#34;: &#34;Name two great reasons to visit Sligo &#34;, &#34;max_tokens&#34;: 10, &#34;temperature&#34;: 0.5, &#34;guided_json&#34;:&#34;{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}&#34;}&#39; ` will provoke a Uncaught exceptions from xgrammer in `./lib64/python3.11/site-packages/xgrammar/compiler.py `</p> <p>Issue with more information: https://github.com/vllm-project/vllm/issues/17248</p> <p>### PoC Make a call to vllm with invalid json_scema e.g. `{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}`</p> <p>`curl -s http://localhost:8000/v1/completions -H &#34;Content-Type: application/json&#34; -d &#39;{&#34;model&#34;: &#34;meta-llama/Llama-3.2-3B-Instruct&#34;,&#34;prompt&#34;: &#34;Name two great reasons to visit Sligo &#34;, &#34;max_tokens&#34;: 10, &#34;temperature&#34;: 0.5, &#34;guided_json&#34;:&#34;{\&#34;properties\&#34;:{\&#34;reason\&#34;:{\&#34;type\&#34;: \&#34;stsring\&#34;}}}&#34;}&#39; ` ### Impact vllm crashes</p> <p>example traceback ``` ERROR 03-26 17:25:01 [core.py:340] EngineCore hit an exception: Traceback (most recent call last): ERROR 03-26 17:25:01 [core.py:340] File &#34;/home/derekh/workarea/vllm/vllm/v1/engine/core.py&#34;, line 333, in run_engine_core ERROR 03-26 17:25:01 [core.py:340] engine_core.run_busy_loop() ERROR 03-26 17:25:01 [core.py:340] File &#34;/home/derekh/workarea/vllm/vllm…</p> https://cve.radiocsirt.org/vuln/ghsa-6qc9-v4r8-22xg https://cve.radiocsirt.org/vuln/pysec-2025-54 <p><strong>Affected:</strong> PyPI: vllm</p> <p>vLLM is an inference and serving engine for large language models (LLMs). In versions 0.8.0 up to but excluding 0.9.0, hitting the /v1/completions API with a invalid json_schema as a Guided Param kills the vllm server. This vulnerability is similar GHSA-9hcf-v7m4-6m2j/CVE-2025-48943, but for regex instead of a JSON schema. Version 0.9.0 fixes the issue.</p> <p><strong>Affected:</strong> PyPI: vllm</p> <p>vLLM is an inference and serving engine for large language models (LLMs). In versions 0.8.0 up to but excluding 0.9.0, hitting the /v1/completions API with a invalid json_schema as a Guided Param kills the vllm server. This vulnerability is similar GHSA-9hcf-v7m4-6m2j/CVE-2025-48943, but for regex instead of a JSON schema. Version 0.9.0 fixes the issue.</p> https://cve.radiocsirt.org/vuln/pysec-2025-54