CVE-2026-44563 Open WebUI: Ollama Model Access Control Bypass via /api/generate, /api/embed, /api/embeddings, and /api/show

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints accept any model name from the user and forward the request to the Ollama backend without checking whether the...

CVE-2026-44563 Open WebUI: Ollama Model Access Control Bypass via /api/generate, /api/embed, /api/embeddings, and /api/show

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints accept any model name from the user and forward the request to the Ollama backend without checking whether the...

EUVD-2026-30614

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints accept any model name from the user and forward the request to the Ollama backend without checking whether the...

CVE-2026-44563

Open WebUI/Open WebUI’s Ollama integration vulnerability (CVE-2026-44563) affects the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints. These endpoints forward a user-supplied model name to the Ollama backend without enforcing AccessGrants.has_access(), effectively bypassing mo...

CVE-2026-44563

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /api/generate, /api/embed, /api/embeddings, and /api/show endpoints accept any model name from the user and forward the request to the Ollama backend without checking whether the...

MINI-M99V-5F5H-8483

Bulletin has no description...

WS-2024-0010

In some settings, an adversary can perform a training data extraction attack to recover individual training examples by querying the language model. It finds that larger models are more vulnerable than smaller models...

WS-2024-0007

In some settings, an adversary can perform a training data extraction attack to recover individual training examples by querying the language model. It finds that larger models are more vulnerable than smaller models...

MINI-QHM3-FR99-HFMW

Bulletin has no description...

CVE-2026-42559

A flaw was found in rmcp, the official Rust SDK for the Model Context Protocol. The Streamable HTTP server transport in rmcp failed to validate the incoming Host header, enabling a malicious public website to exploit this through a DNS rebinding attack. This allows the attacker to send...

autonomize-model-sdk (>=1.0.6, <=1.1.21), experiment-goodies (>=0.1.1, <=0.1.2) +1 more potentially affected by CVE-2026-2652 via mlflow (2.20.1)

mlflow PYPI version 2.20.1 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - autonomize-model-sdk =1.0.6, =0.1.1, =0.1.2 - terratorch 0.99.8 Source cves: CVE-2026-2652 Source advisory: OSV:GHSA-75CM-X2W3-8MGF...

mlflow-cratedb (2.13.0), mlflowcli (0.0.2) +1 more potentially affected by CVE-2026-2652 via mlflow (2.13.0)

mlflow PYPI version 2.13.0 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - mlflow-cratedb 2.13.0 - mlflowcli 0.0.2 - model-controller 1.0.0 Source cves: CVE-2026-2652 Source advisory: OSV:GHSA-75CM-X2W3-8MGF...

autonomize-model-sdk (1.0.4), databricks-genai (>=1.0.8, <=1.0.13) +1 more potentially affected by CVE-2026-2652 via mlflow (2.14.2)

mlflow PYPI version 2.14.2 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - autonomize-model-sdk 1.0.4 - databricks-genai =1.0.8, =1.0.33, =1.0.37 Source cves: CVE-2026-2652 Source advisory: OSV:GHSA-75CM-X2W3-8MGF...

autonomize-model-sdk (>=1.0.6, <=1.1.21), experiment-goodies (>=0.1.1, <=0.1.2) +1 more potentially affected by CVE-2026-2652 via mlflow (2.20.1)

mlflow PYPI version 2.20.1 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - autonomize-model-sdk =1.0.6, =0.1.1, =0.1.2 - terratorch 0.99.8 Source cves: CVE-2026-2652 Source advisory: OSV:PYSEC-2026-2221...

mlflow-cratedb (2.13.0), mlflowcli (0.0.2) +1 more potentially affected by CVE-2026-2652 via mlflow (2.13.0)

mlflow PYPI version 2.13.0 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - mlflow-cratedb 2.13.0 - mlflowcli 0.0.2 - model-controller 1.0.0 Source cves: CVE-2026-2652 Source advisory: OSV:PYSEC-2026-2221...

emb-model (>=0.1.0, <=0.2.9), mlflow-cratedb (2.11.3) +2 more potentially affected by CVE-2026-2652 via mlflow (2.11.3)

mlflow PYPI version 2.11.3 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - emb-model =0.1.0, =0.0.3.dev0, =0.0.15 - smartdeploymiguel 0.1.1 Source cves: CVE-2026-2652 Source advisory: OSV:PYSEC-2026-2221...

autonomize-model-sdk (1.0.4), databricks-genai (>=1.0.8, <=1.0.13) +1 more potentially affected by CVE-2026-2652 via mlflow (2.14.2)

mlflow PYPI version 2.14.2 is affected by a known vulnerability. The following packages have a transitive dependency on mlflow and may be impacted: - autonomize-model-sdk 1.0.4 - databricks-genai =1.0.8, =1.0.33, =1.0.37 Source cves: CVE-2026-2652 Source advisory: OSV:PYSEC-2026-2221...

CVE-2026-45387

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.5, when setting model permissions so that a group has read access to it, intending for other users to use it, those users also can read the model's system prompt. However users may...

CVE-2026-31232

The CosyVoice project thru commit 6e01309e01bc93bbeb83bdd996b1182a81aaf11e 2025-30-21 contains an insecure deserialization vulnerability CWE-502 in its model loading process. When loading model files .pt from a user-specified directory via the --modeldir argument, the code uses torch.load without...

CVE-2026-31238

The Ludwig framework thru 0.10.4 is vulnerable to insecure deserialization CWE-502 in its model serving component. When starting a model server with the ludwig serve command, the framework loads model weight files using torch.load without enabling the security-restrictive weightsonly=True...