CVE-2026-92971
InternLM LMDeploy through 0.17.0 contains a reachable assertion vulnerability in the DistServe decode migration loop that allows unauthenticated attackers to terminate the inference engine. Attackers can submit a migration_request with an empty remote_block_ids list to trigger an AssertionError that crashes the engine loop and causes subsequent inference requests to fail.
Weakness
InternLM LMDeploy through 0.17.0 contains a reachable assertion vulnerability in the DistServe decode migration loop that allows unauthenticated attackers to terminate the inference engine. Attackers can submit a migration_request with an empty remote_block_ids list to trigger an AssertionError that crashes the engine loop and causes subsequent inference requests to fail.
References
- https://github.com/InternLM/lmdeploy
- https://github.com/InternLM/lmdeploy/blob/v0.17.0/lmdeploy/pytorch/engine/engine_loop.py#L560-L564
- https://github.com/InternLM/lmdeploy/blob/v0.17.0/lmdeploy/pytorch/utils.py#L187-L214
- https://github.com/InternLM/lmdeploy/blob/v0.17.0/lmdeploy/serve/openai/endpoints/completions.py#L130-L136
- https://github.com/InternLM/lmdeploy/issues/4965
- https://www.vulncheck.com/advisories/internlm-lmdeploy-through-0.17.0-assertion-denial-of-service
Find the bug before an attacker does.
Sign in with GitHub and run your first audit in under a minute. The free plan needs no credit card.