SGLang

SGLang contains a RCE vulnerability when attempting to load model weights from a HuggingFace repository, specifically within the /update_weights_from_disk, where torch.load(..., weights_only=False) fallback enables pickl…

SGLang contains an RCE vulnerability when the optional dumper subsystem is enabled, allowing for a sandbox escape when DUMPER_SERVER_PORT is set, enabling code execution on inference requests.

SGLang's reranking endpoint (/v1/rerank) achieves Remote Code Execution (RCE) when a model file containing a malcious tokenizer.chat_template is loaded, as the Jinja2 chat templates are rendered using an unsandboxed jinj…

SGLang' encoder parallel disaggregation system is vulnerable to unauthenticated remote code execution through the disaggregation module, which deserializes untrusted data using pickle.loads() without authentication.

SGLang's multimodal generation module is vulnerable to unauthenticated remote code execution through the ZMQ broker, which deserializes untrusted data using pickle.loads() without authentication.

SGLang contains a model weight exfiltration vulnerability when no API keys are configured, as SGLang will expose two endpoints that allow a remote attacker to trigger distributed weight broadcasting using NCCL and then t…

SGLang contains a credential leakage vulnerability in the /server_info endpoint, which will return API keys and SSL keyfile information when only the --admin-api-key is configured.

SGLangs `replay_request_dump.py` contains an insecure pickle.load() without validation and proper deserialization.

SGLang contains an SSRF and local file read in the multimodal generation endpoint /v1/chat/completions due to unsanitized image_url, allowing access to internal metadata, secrets, and services.

SGLang contains an unauthenticated RCE in /load_lora_adapter_from_tensors via bypass of SafeUnpickler’s incomplete denylist, allowing arbitrary command execution through crafted base64-encoded pickle payloads.

SGLangs multimodal generation runtime is vulnerable to unauthenticated remote code execution when the --enable-custom-logit-processor option is enabled, as Python objects loaded via dill.loads() will be deserialized with…

SGLang uses an expert-parallel backup subsystem that exposes a ZeroMQ PULL socket on a routable network interface that does not contain authentication or deserialization safeguards, allowing an attacker to provide a mali…

SGLangs multimodal generation runtime scheduler's ROUTER socket binds to 0.0.0.0 by default and contains a sink that calls pickle.loads() on incoming messages, enabling RCE when exposed to the internet.

SGLangs multimodal generation runtime is vulnerable to an unauthenticated path traversal vulnerability, allowing an attacker to write arbitrary files anywhere the server process has write access, by including ../ sequenc…

A vulnerability was determined in sgl-project SGLang up to 0.5.11.