CVE-2026-71281

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-71281 peft Unsafe Deserialization via torch.load() Without weights_only in LoRA-GA and CorDA Modules

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

EUVD-2026-53366

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-71281 peft Unsafe Deserialization via torch.load() Without weights_only in LoRA-GA and CorDA Modules

Hugging Face peft's LoRA-GA and CorDA initialization modules src/peft/tuners/lora/corda.py lines 102 and 163, and src/peft/tuners/lora/loraga.py line 101 call torch.load on config-specified cache/covariance files without weightsonly=True, bypassing peft's own safe-loading wrapper used elsewhere i...

CVE-2026-47621

NVIDIA Dynamo for Linux contains a race condition in the LoRA manager singleton initialization (CVE-2026-47621). The issue affects the LoRA manager component and can lead to denial of service and data tampering. The provided materials specify a CVSS 3.1 base score of 6.5 (Medium) with network att...

CVE-2026-47621

NVIDIA Dynamo for Linux contains a vulnerability where an attacker could cause a race condition in the LoRA manager singleton initialization. A successful exploit of this vulnerability might lead to denial of service and data tampering...

CVE-2026-47620

CVE-2026-47620 affects NVIDIA Dynamo for Linux. The issue is a race condition in the LoRA manager singleton initialization, which could allow data tampering and denial of service. The CVSSv3.1 base score is 6.5 (Medium) with Network attack vector, high attack complexity, no privileges required, n...

EUVD-2026-52815

NVIDIA Dynamo for Linux contains a vulnerability where an attacker could cause a race condition in the LoRA manager singleton initialization. A successful exploit of this vulnerability might lead to data tampering and denial of service...

CVE-2026-15969

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

EUVD-2026-51267

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

CVE-2026-15969 CVE-2026-15969

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

CVE-2026-15969

SGLang (CVE-2026-15969) exposes an unauthenticated RCE via /load_lora_adapter_from_tensors caused by bypassing SafeUnpickler’s incomplete denylist, enabling arbitrary command execution through crafted base64-encoded pickle payloads. Multiple sources confirm the vulnerability and its impact on SGL...

PT-2026-66558

Name of the Vulnerable Software and Affected Versions SGLang affected versions not specified Description An unauthenticated remote code execution issue exists in the '/load lora adapter from tensors' endpoint. This occurs due to a bypass of the SafeUnpickler's incomplete denylist, which allows an...

PYSEC-2026-3064 SGLang: Reachable Assertion via  lora_path  in LoRAManager enables remote Denial of Dervice

A security vulnerability has been detected in SGLang 0.5.10.post1. Impacted is an unknown function of the file python/sglang/srt/lora/loramanager.py of the component Inference HTTP Endpoint. Such manipulation of the argument lorapath leads to reachable assertion. The attack can be launched...

Astra Linux – Vulnerability in Linux 6.1

In the Linux kernel, the following vulnerabilities have been resolved: ipv6: rpl: Reserve enough headroom for the MAC header when recompressing the SRH. The function ipv6rplsrhrcv decompresses the RFC 6554 Source Routing Header, swaps the next segment into ipv6hdr-daddr, recompresses it, and then...

CVE-2025-71379

A flaw was found in vLLM. Multiple regular expression denial of service ReDoS vulnerabilities exist in versions greater than or equal to 0.6.3 and less than 0.9.0. An attacker can exploit this by submitting crafted input with nested or repeated structures to specific regex patterns within vLLM,...

CVE-2025-71379

vLLM versions = 0.6.3 and 0.9.0 contain multiple regular expression denial of service ReDoS vulnerabilities. Several regex patterns — in vllm/lora/utils.py, the phi4mini tool parser, and the OpenAI-compatible serving chat endpoint — are susceptible to catastrophic backtracking. An attacker...

CVE-2025-71379

vLLM versions = 0.6.3 and 0.9.0 contain multiple regular expression denial of service ReDoS vulnerabilities. Several regex patterns — in vllm/lora/utils.py, the phi4mini tool parser, and the OpenAI-compatible serving chat endpoint — are susceptible to catastrophic backtracking. An attacker...

CVE-2025-71379

CVE-2025-71379 affects vLLM versions 0.6.3 through 0.8.x (before 0.9.0). The vulnerability is a set of regular expression denial of service (ReDoS) flaws in multiple components: (1) regex patterns in vllm/lora/utils.py, (2) the phi4mini tool parser, and (3) the OpenAI-compatible serving chat endp...

CVE-2025-71379 vllm - Regular Expression Denial of Service in Multiple Components

vLLM versions = 0.6.3 and 0.9.0 contain multiple regular expression denial of service ReDoS vulnerabilities. Several regex patterns — in vllm/lora/utils.py, the phi4mini tool parser, and the OpenAI-compatible serving chat endpoint — are susceptible to catastrophic backtracking. An attacker...