LMDeploy has Server-Side Request Forgery (SSRF) via Vision-Language Image Loading
CVE-2026-33626 · GHSA-6w67-hwm5-92mq · published 2026-04-21 · updated 2026-09-23
- Severity
- High
- CVSS score
- 7.5
- Chance of exploitation (EPSS)
- 45.3%
- Fixed in
- No fix yet
- Exploitation Public proof of concept
- Automatable Yes
- Technical impact Partial
- CISA SSVC
Cloud Credential Theft: Access AWS/GCP/Azure metadata APIs Internal Service Access: Reach services not exposed to internet Information Disclosure: Port scan internal networks Lateral Movement: Pivot point for further attacks
No fixed release is listed yet for lmdeploy. Follow the source advisory for updates, and until one ships:
General guidance from ShieldPi, not a vendor fix
- Server-side request forgery. Restrict its outbound network access to the hosts it needs, and keep internal addresses and cloud metadata endpoints out of its reach.
| Package | Registry | Affected | Fixed in |
|---|---|---|---|
| lmdeploy | PyPI | <= 0.12.2 | No fix yet |
- CriticalMLflow: Unauthenticated full-read SSRF in webhook delivery: _validate_webhook_url bypassed via unvalidated HTTP redirects (and DNS rebindingmlflowExploited
- Critical@zereight/mcp-gitlab Vulnerable to Server-Side Request Forgery@zereight/mcp-gitlab
- CriticalSSRF + LLM API key exfiltration via user-controlled basepath / baseUrl in 8 chat-model nodesflowise, flowise-components
- CriticalLocalAI POST /models/apply permits unauthenticated server-side request forgery through gallery URLsgithub.com/mudler/LocalAI
- CriticalFastMCP OpenAPI Provider has an SSRF & Path Traversal Vulnerabilityfastmcp
- CriticalBentoML SSRF Vulnerability in File Upload Processingbentoml
- CriticalLMDeploy has Remote Code Execution by Pickle Deserialization via handle_zmq_recv in lmdeploy/lmdeploy/pytorch/disagg/conn/engine_conn.pylmdeploy
- CriticalLMdeploy has Remote Code Execution by Pickle Deserialization via zmq_rpc.call_and_response() in InterLM/lmdeploylmdeploy
- HighLMDeploy has an SSRF bypasslmdeploy
- HighLMDeploy vulnerable to arbitrary code execution via eval() of untrusted quant_dtype in model config loadinglmdeploy
- Highlmdeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-outlmdeploy
- HighLMDeploy: Arbitrary code execution via hardcoded trust_remote_code=True in lmdeploy model initializationlmdeploy
The impact, patch and workaround text comes from the advisory, shortened (GitHub Advisory Database, CC BY 4.0). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.