Ollama
Monthly
Server-Side Request Forgery in Ollama versions 0.30.0 through 0.33.2 allows unauthenticated remote attackers to redirect the Ollama server's blob-download requests to arbitrary internal hosts, including cloud metadata endpoints such as the AWS instance metadata service. The flaw exists in the tensor-layer model pulling code path (x/transfer/download.go), where HTTP redirect destinations are not validated against an allowlist or the originating registry host. With no public exploit identified at time of analysis, the primary real-world impact is credential theft from cloud metadata services when Ollama runs on cloud-hosted infrastructure.
Remote denial of service in Ollama (HEAD f0078ae) allows unauthenticated attackers to crash the entire server by uploading a sub-1KB crafted GGUF file whose attacker-controlled string lengths, tensor dimension counts, and metadata array counts are trusted as allocation sizes without bounds checking against the remaining file size. Reachable through the blob upload plus model create/pull API endpoints, the flaw triggers unrecoverable Go runtime out-of-memory fatal errors or makeslice panics that bypass the server's recovery middleware, taking down the whole process rather than a single request. No public exploit is identified at time of analysis and the CVE is not in CISA KEV, but a vendor patch is available and VulnCheck has published an advisory.
Denial-of-service in Ollama's downloadBlob function lets remote, unauthenticated attackers crash affected installations by supplying malformed data that triggers an out-of-bounds array access. The flaw (ZDI-CAN-27277 / ZDI-26-403, CWE-129) carries a CVSS 7.5 with an availability-only impact and requires no authentication or user interaction. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Information disclosure in Ollama's model quantization engine (confirmed affected in v0.13.5) allows remote, unauthenticated attackers to read and exfiltrate the server's heap memory, exposing sensitive in-process data such as model contents, API keys, or other secrets resident in memory. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) indicates trivial network exploitation against an exposed Ollama instance with no authentication. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Heap out-of-bounds read in Ollama's GGUF model loader (<0.17.1) leaks sensitive memory contents including API keys, environment variables, and concurrent user data when processing maliciously crafted model files. Attackers can trigger the vulnerability by uploading a GGUF file with tensor offsets exceeding file bounds via the unauthenticated /api/create endpoint, then exfiltrate leaked memory through /api/push to attacker-controlled registries. While default deployments bind to localhost only, the widely-adopted OLLAMA_HOST=0.0.0.0 configuration exposes instances to network exploitation. Vendor-released patch available in version 0.17.1 with GitHub commit 88d57d0483cca907e0b23a968c83627a20b21047 adding bounds validation to fs/ggml/gguf.go and server/quantization.go.
Remote code execution in Ollama for Windows (versions 0.12.10 through 0.17.5) lets an attacker who can influence HTTP update responses write arbitrary executables outside the update staging directory. The flaw stems from unvalidated attacker-controlled HTTP headers being passed to filepath.Join, permitting path traversal into sensitive locations such as the Windows Startup folder. Chained with CVE-2026-42248 (missing update signature verification), the silent auto-updater executes the planted binary, yielding automatic and persistent code execution with no user interaction. There is no public exploit identified at time of analysis and EPSS is very low (0.03%).
Silent auto-update code execution in Ollama for Windows (versions 0.12.10 through 0.17.5) stems from an update-verification routine that unconditionally returns success, so no signature or trust check is performed before staging and executing downloaded update payloads (CWE-494). Because Ollama on Windows applies updates silently and automatically, an attacker who can substitute the update executable can have arbitrary code installed and run without user awareness. No public exploit identified at time of analysis; EPSS is very low (0.01%) and CISA SSVC records exploitation as none, though technical impact is rated total.
Wallos, an open-source self-hostable subscription tracker, contains a Server-Side Request Forgery (SSRF) vulnerability in versions prior to 4.7.0 that allows authenticated users to access internal network services, cloud metadata endpoints, and localhost-bound services. The vulnerability exists in three unprotected attack surfaces: the AI Ollama host parameter, the AI recommendations endpoint, and the notification cron job-areas that were missed when SSRF protections were partially implemented in an earlier patch (CVE-2026-30840). An attacker with valid credentials can leverage these endpoints to reach sensitive internal resources including AWS IMDSv1, GCP, and Azure metadata services.
Ollama before 0.1.34 does not validate the format of the digest (sha256 with 64 hex digits) when getting the model path, and thus mishandles the TestGetBlobsPath test cases such as fewer than 64 hex. Rated high severity (CVSS 8.8), this vulnerability is remotely exploitable, low attack complexity. Public exploit code available and no vendor patch available.
Integer overflow vulnerabilities in Ollama's GGUF model decoder (versions 0.31.0-0.31.1) allow a remote attacker to crash the decoder by supplying a crafted GGUF model file containing malformed string lengths, tensor dimension counts, or array sizes. The overflow triggers in readGGUFV1String and related parsing functions in fs/ggml/gguf.go, causing Ollama to abort model loading and become temporarily unavailable. A public proof-of-concept exists (GitHub issue #17033) and a patch is available in v0.31.2-rc1; exploitation requires the target to load the malicious model file.
Path traversal vulnerability in Ollama up to version 0.20.2 affects the digestToPath function in the Tensor Model Transfer Handler, allowing remote attackers with high complexity to manipulate digest arguments and traverse the filesystem. Public exploit code exists, though the vendor has not acknowledged disclosure attempts. CVSS 6.3 reflects low confidentiality, integrity, and availability impact limited by high attack complexity and no scope change.
Server-side request forgery in Ollama's Model Pull API (via server/download.go) allows authenticated remote attackers to manipulate file processing and trigger SSRF attacks, affecting Ollama versions up to 18.1. The vulnerability carries a CVSS score of 6.3 with moderate impact on confidentiality, integrity, and availability. No public exploit code or active exploitation has been confirmed, and the vendor has not responded to early disclosure attempts.
Ollama MCP Server has a command injection vulnerability in execAsync (EPSS 1.0%) allowing remote attackers to execute arbitrary commands on systems running the Ollama AI integration.
An issue in ollama v.0.12.10 allows a remote attacker to cause a denial of service via the fs/ggml/gguf.go, function readGGUFV1String reads a string length from untrusted GGUF metadata [CVSS 7.5 HIGH]
An issue in ollama v.0.12.10 allows a remote attacker to cause a denial of service via the GGUF decoder [CVSS 7.5 HIGH]
Denial of service in Ollama's multi-modal image processing (versions 0.11.5-rc0 through 0.13.5) lets remote attackers crash the model runner by POSTing crafted base64 image data to the /api/chat endpoint. The decoded data is passed to mtmd_helper_bitmap_init_from_buf without validating that it is valid media; the function returns NULL on malformed input and the unchecked pointer is dereferenced, triggering a segmentation fault that takes the model offline for all users until restart. Rated CVSS 4.0 8.7 (availability-only impact); publicly available exploit code exists via a Huntr bounty, but EPSS is low (0.09%, 25th percentile) and it is not on CISA KEV.
An issue in Ollama v0.1.33 allows attackers to delete arbitrary files via sending a crafted packet to the endpoint /api/pull. Rated medium severity (CVSS 6.6), this vulnerability is no authentication required, low attack complexity. No vendor patch available.
A vulnerability in the Ollama server version 0.5.11 allows a malicious user to cause a Denial of Service (DoS) attack by customizing the manifest content and spoofing a service. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
open-webui v0.5.16 is vulnerable to SSRF in routers/ollama.py in function verify_connection. Rated low severity (CVSS 3.3), this vulnerability is low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama versions <=0.3.14 allows a malicious user to upload and create a customized GGUF model file on the Ollama server. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama <=0.3.14 allows a malicious user to create a customized GGUF model file, upload it to the Ollama server, and create it. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama versions <=0.3.14 allows a malicious user to create a customized GGUF model file that, when uploaded and created on the Ollama server, can cause a crash due to an. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A divide by zero vulnerability exists in ollama/ollama version v0.3.3. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An Out-Of-Memory (OOM) vulnerability exists in the `ollama` server version 0.3.14. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. No vendor patch available.
A vulnerability in Ollama versions <=0.3.14 allows a malicious user to create a customized gguf model file that can be uploaded to the public Ollama server. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.46. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.34. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.46. Rated high severity (CVSS 8.2), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama through 0.3.14. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
extractFromZipFile in model.go in Ollama before 0.1.47 can extract members of a ZIP archive outside of the parent directory. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. This Path Traversal vulnerability could allow attackers to access files and directories outside the intended path.
Ollama before 0.1.29 has a DNS rebinding vulnerability that can inadvertently allow remote access to the full API, thereby letting an unauthorized user chat with a large language model, delete a. Rated medium severity (CVSS 6.6), this vulnerability is low attack complexity. No vendor patch available.
Server-Side Request Forgery in Ollama versions 0.30.0 through 0.33.2 allows unauthenticated remote attackers to redirect the Ollama server's blob-download requests to arbitrary internal hosts, including cloud metadata endpoints such as the AWS instance metadata service. The flaw exists in the tensor-layer model pulling code path (x/transfer/download.go), where HTTP redirect destinations are not validated against an allowlist or the originating registry host. With no public exploit identified at time of analysis, the primary real-world impact is credential theft from cloud metadata services when Ollama runs on cloud-hosted infrastructure.
Remote denial of service in Ollama (HEAD f0078ae) allows unauthenticated attackers to crash the entire server by uploading a sub-1KB crafted GGUF file whose attacker-controlled string lengths, tensor dimension counts, and metadata array counts are trusted as allocation sizes without bounds checking against the remaining file size. Reachable through the blob upload plus model create/pull API endpoints, the flaw triggers unrecoverable Go runtime out-of-memory fatal errors or makeslice panics that bypass the server's recovery middleware, taking down the whole process rather than a single request. No public exploit is identified at time of analysis and the CVE is not in CISA KEV, but a vendor patch is available and VulnCheck has published an advisory.
Denial-of-service in Ollama's downloadBlob function lets remote, unauthenticated attackers crash affected installations by supplying malformed data that triggers an out-of-bounds array access. The flaw (ZDI-CAN-27277 / ZDI-26-403, CWE-129) carries a CVSS 7.5 with an availability-only impact and requires no authentication or user interaction. There is no public exploit identified at time of analysis, and it is not listed in CISA KEV.
Information disclosure in Ollama's model quantization engine (confirmed affected in v0.13.5) allows remote, unauthenticated attackers to read and exfiltrate the server's heap memory, exposing sensitive in-process data such as model contents, API keys, or other secrets resident in memory. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N) indicates trivial network exploitation against an exposed Ollama instance with no authentication. There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Heap out-of-bounds read in Ollama's GGUF model loader (<0.17.1) leaks sensitive memory contents including API keys, environment variables, and concurrent user data when processing maliciously crafted model files. Attackers can trigger the vulnerability by uploading a GGUF file with tensor offsets exceeding file bounds via the unauthenticated /api/create endpoint, then exfiltrate leaked memory through /api/push to attacker-controlled registries. While default deployments bind to localhost only, the widely-adopted OLLAMA_HOST=0.0.0.0 configuration exposes instances to network exploitation. Vendor-released patch available in version 0.17.1 with GitHub commit 88d57d0483cca907e0b23a968c83627a20b21047 adding bounds validation to fs/ggml/gguf.go and server/quantization.go.
Remote code execution in Ollama for Windows (versions 0.12.10 through 0.17.5) lets an attacker who can influence HTTP update responses write arbitrary executables outside the update staging directory. The flaw stems from unvalidated attacker-controlled HTTP headers being passed to filepath.Join, permitting path traversal into sensitive locations such as the Windows Startup folder. Chained with CVE-2026-42248 (missing update signature verification), the silent auto-updater executes the planted binary, yielding automatic and persistent code execution with no user interaction. There is no public exploit identified at time of analysis and EPSS is very low (0.03%).
Silent auto-update code execution in Ollama for Windows (versions 0.12.10 through 0.17.5) stems from an update-verification routine that unconditionally returns success, so no signature or trust check is performed before staging and executing downloaded update payloads (CWE-494). Because Ollama on Windows applies updates silently and automatically, an attacker who can substitute the update executable can have arbitrary code installed and run without user awareness. No public exploit identified at time of analysis; EPSS is very low (0.01%) and CISA SSVC records exploitation as none, though technical impact is rated total.
Wallos, an open-source self-hostable subscription tracker, contains a Server-Side Request Forgery (SSRF) vulnerability in versions prior to 4.7.0 that allows authenticated users to access internal network services, cloud metadata endpoints, and localhost-bound services. The vulnerability exists in three unprotected attack surfaces: the AI Ollama host parameter, the AI recommendations endpoint, and the notification cron job-areas that were missed when SSRF protections were partially implemented in an earlier patch (CVE-2026-30840). An attacker with valid credentials can leverage these endpoints to reach sensitive internal resources including AWS IMDSv1, GCP, and Azure metadata services.
Ollama before 0.1.34 does not validate the format of the digest (sha256 with 64 hex digits) when getting the model path, and thus mishandles the TestGetBlobsPath test cases such as fewer than 64 hex. Rated high severity (CVSS 8.8), this vulnerability is remotely exploitable, low attack complexity. Public exploit code available and no vendor patch available.
Integer overflow vulnerabilities in Ollama's GGUF model decoder (versions 0.31.0-0.31.1) allow a remote attacker to crash the decoder by supplying a crafted GGUF model file containing malformed string lengths, tensor dimension counts, or array sizes. The overflow triggers in readGGUFV1String and related parsing functions in fs/ggml/gguf.go, causing Ollama to abort model loading and become temporarily unavailable. A public proof-of-concept exists (GitHub issue #17033) and a patch is available in v0.31.2-rc1; exploitation requires the target to load the malicious model file.
Path traversal vulnerability in Ollama up to version 0.20.2 affects the digestToPath function in the Tensor Model Transfer Handler, allowing remote attackers with high complexity to manipulate digest arguments and traverse the filesystem. Public exploit code exists, though the vendor has not acknowledged disclosure attempts. CVSS 6.3 reflects low confidentiality, integrity, and availability impact limited by high attack complexity and no scope change.
Server-side request forgery in Ollama's Model Pull API (via server/download.go) allows authenticated remote attackers to manipulate file processing and trigger SSRF attacks, affecting Ollama versions up to 18.1. The vulnerability carries a CVSS score of 6.3 with moderate impact on confidentiality, integrity, and availability. No public exploit code or active exploitation has been confirmed, and the vendor has not responded to early disclosure attempts.
Ollama MCP Server has a command injection vulnerability in execAsync (EPSS 1.0%) allowing remote attackers to execute arbitrary commands on systems running the Ollama AI integration.
An issue in ollama v.0.12.10 allows a remote attacker to cause a denial of service via the fs/ggml/gguf.go, function readGGUFV1String reads a string length from untrusted GGUF metadata [CVSS 7.5 HIGH]
An issue in ollama v.0.12.10 allows a remote attacker to cause a denial of service via the GGUF decoder [CVSS 7.5 HIGH]
Denial of service in Ollama's multi-modal image processing (versions 0.11.5-rc0 through 0.13.5) lets remote attackers crash the model runner by POSTing crafted base64 image data to the /api/chat endpoint. The decoded data is passed to mtmd_helper_bitmap_init_from_buf without validating that it is valid media; the function returns NULL on malformed input and the unchecked pointer is dereferenced, triggering a segmentation fault that takes the model offline for all users until restart. Rated CVSS 4.0 8.7 (availability-only impact); publicly available exploit code exists via a Huntr bounty, but EPSS is low (0.09%, 25th percentile) and it is not on CISA KEV.
An issue in Ollama v0.1.33 allows attackers to delete arbitrary files via sending a crafted packet to the endpoint /api/pull. Rated medium severity (CVSS 6.6), this vulnerability is no authentication required, low attack complexity. No vendor patch available.
A vulnerability in the Ollama server version 0.5.11 allows a malicious user to cause a Denial of Service (DoS) attack by customizing the manifest content and spoofing a service. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
open-webui v0.5.16 is vulnerable to SSRF in routers/ollama.py in function verify_connection. Rated low severity (CVSS 3.3), this vulnerability is low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama versions <=0.3.14 allows a malicious user to upload and create a customized GGUF model file on the Ollama server. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama <=0.3.14 allows a malicious user to create a customized GGUF model file, upload it to the Ollama server, and create it. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A vulnerability in ollama/ollama versions <=0.3.14 allows a malicious user to create a customized GGUF model file that, when uploaded and created on the Ollama server, can cause a crash due to an. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
A divide by zero vulnerability exists in ollama/ollama version v0.3.3. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An Out-Of-Memory (OOM) vulnerability exists in the `ollama` server version 0.3.14. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. No vendor patch available.
A vulnerability in Ollama versions <=0.3.14 allows a malicious user to create a customized gguf model file that can be uploaded to the public Ollama server. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.46. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.34. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama before 0.1.46. Rated high severity (CVSS 8.2), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
An issue was discovered in Ollama through 0.3.14. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available and no vendor patch available.
extractFromZipFile in model.go in Ollama before 0.1.47 can extract members of a ZIP archive outside of the parent directory. Rated high severity (CVSS 7.5), this vulnerability is remotely exploitable, no authentication required, low attack complexity. This Path Traversal vulnerability could allow attackers to access files and directories outside the intended path.
Ollama before 0.1.29 has a DNS rebinding vulnerability that can inadvertently allow remote access to the full API, thereby letting an unauthorized user chat with a large language model, delete a. Rated medium severity (CVSS 6.6), this vulnerability is low attack complexity. No vendor patch available.