← Back
CVE-2025-67729
Unsafe Deserialization in lmdeploy (CVE-2025-67729)
Summary
vulnerability in lmdeploy (CVE-2025-67729). Successful exploitation can lead to full system takeover. Exploitable via ``create_malicious_checkpoint.py``. Mitigation: upgrade to `0.11.1` or later.
AI summary snake-internal / snake-material-v2
A vulnerability tracked as **CVE-2025-67729** has been found in lmdeploy.
Attackers can target a specific entry point like ``create_malicious_checkpoint.py`` over the network to misuse the product.
Successful exploitation can lead to full system takeover. CVSS score: ?/10.
What to do: upgrade lmdeploy to **0.11.1** or later.
If unsure, ask your IT team or search "lmdeploy CVE-2025-67729" on the vendor's site.
CVE-2025-67729 (lmdeploy) — CWE-502 /
Attack vector: remote (network-reachable) / unauthenticated
Attack surface: `create_malicious_checkpoint.py` / `exploit.py` / `pytorch_model.bin`
Patched: `0.11.1` — apply immediately
Plan: 1) Audit SBOM/dependencies, 2) Stage→prod upgrade, 3) Add WAF/proxy monitoring on affected endpoints, 4) Hunt IOCs in logs.
Refs: see the GHSA / vendor advisory / patched release linked on this page.
❓ What is the problem
**unsafe deserialization** (CWE-502) exists in lmdeploy. Attackers reach the vulnerable code path via ``create_malicious_checkpoint.py`` without authentication.
📍 Affected scope
lmdeploy — . Attack surface: `create_malicious_checkpoint.py` / `exploit.py` / `pytorch_model.bin`.
🔥 Severity
Severity: ?. Successful exploitation can lead to full system takeover
🔧 How to fix
Update to **0.11.1**.
🛡️ Workaround
Until the patch is applied: disable the affected feature, apply WAF rules, or restrict access via network ACLs.
🔍 Detection
Search webserver/proxy logs for unusual request patterns matching this CVE's known IOCs. Run `grep -r 'lmdeploy' .` against your dependency files (package-lock.json, requirements.txt, go.sum) to find affected services.
Response Actions (7 steps)
Concrete steps and command examples for SOC/SRE teams to execute in order
-
1Identify exposure identify
grep -r 'lmdeploy' . | grep -v node_modulesリポジトリと本番環境の依存ファイル (package-lock.json / requirements.txt / go.sum / Gemfile.lock 等) で `lmdeploy` を grep し、稼働しているサービス・バージョンを把握する。
-
6Apply patch patch
Upgrade lmdeploy to 0.11.1ステージング環境で 0.11.1 に上げて回帰テスト → 本番反映。回帰テストはアプリの主要ハッピーパスと、Step 3 で見つけた異常検知の続報チェックを含めること。
-
7Post-deployment verification verify
Confirm patched version is live in productionパッチ適用後、ステージングで PoC または同等の悪用パターンを再現して脆弱性が閉じたことを確認。本番では Step 3 と同じログクエリでアラート再発が無いか継続監視。
Affected packages
PyPI
lmdeploy
[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.11.1"}]}]
References
- advisory https://nvd.nist.gov/vuln/detail/CVE-2025-67729
- advisory https://github.com/advisories/GHSA-9pf3-7rrr-x5jh
- package https://github.com/InternLM/lmdeploy
- package https://pypi.org/project/lmdeploy
- web https://github.com/InternLM/lmdeploy/security/advisories/GHSA-9pf3-7rrr-x5jh
- web https://github.com/InternLM/lmdeploy/commit/eb04b4281c5784a5cff5ea639c8f96b33b3ae5ee