llamaCVEVulnerability Intelligence
/
Advisory · GitHub · PyPI · vllm

GHSA-33CG-GXV8-3P8G

vLLM denial of service via prompt embeds on M-RoPE models

Parche disponible
Descargar PNG JSON
CVSS
—
high
Fuente: —
ExplotaciónSin evidenciaNo figura en CISA KEV
Exploits públicos0Ninguno conocido
Parche1 versiónPrimer parche a -20 d de la divulgación
Exposición122 dDesde la divulgación, 20 jul 2026, 2:13 p. m.
Detección119 dDesde la divulgación hasta su ingreso al índice
Timeline

Ciclo de vida de la vulnerabilidad

Hoy
Divulgación
Publicada
Modificada
Explotación
Remediación
Parche 0.24.0
Monitoreo
Detectada
30 jun 26
20 jul 26
10 sept 26
17 nov 26
20 nov 26
30 jun 26
5 ago 26
9 sept 26
15 oct 26
20 nov 26
FechaFaseEventoDetalle
Remediación Parche 0.24.0 vllm · -20 d desde divulgación
Divulgación Publicada GHSA-33CG-GXV8-3P8G divulgada
Divulgación Modificada Última actualización del registro
Monitoreo Detectada Ingresó al índice
Resumen
Descripción del advisory
Texto original · GitHub · en inglés

### Summary
_Short summary of the problem. Make the impact and severity as clear as possible. For example: An unsafe deserialization vulnerability allows any unauthenticated user to execute arbitrary code on the server._

Sending a pure prompt embeds payload in a `/v1/completions` request with a model using M-RoPE causes the EngineCore to fail an assertion and fatally crash, shutting down the entire server application.

Any remote user who is authorized to make a `/v1/completions` endpoint can trivially make such a request and induce a crash.

### Details
_Give all details on the vulnerability. Pointing to the incriminated source code is very helpful for the maintainer._

In commit 56669c1 https://github.com/vllm-project/vllm/commit/56669c1f293d5c53b6a19ddf2f78802fa9fff2c2, a simple assert intended to be a type-narrowing assert was added to the `_init_mrope_positions` method in `GPUModelRunner` (the offending line on main at the time of writing: https://github.com/vllm-project/vllm/blob/2d481f8a946ee0521872af0f098674a8ee01ce4a/vllm/v1/worker/gpu_model_runner.py#L1588-L1607).

assert req_state.prompt_token_ids is not None, (
            "M-RoPE requires prompt_token_ids to be available."
        )

This type narrowing assert is to prevent mypy errors later in the function because `None` is not a valid type for `mrope_model.get_mrope_input_positions`. Unfortunately, this assertion is not always true. `/v1/completions` requests that specify `prompt=None` and `prompt_embeds=<not none>` will indeed create a CachedRequestState where `prompt_token_ids` is `None`. This triggers the assertion, which in turn crashes the EngineCore and the Server application.

(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 3997, in execute_model
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]     deferred_state_corrections_fn = self._update_states(scheduler_output)
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]                                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 1239, in _update_states
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]     self._init_mrope_positions(req_state)
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/worker/gpu_model_runner.py", line 1582, in _init_mrope_positions
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]     assert req_state.prompt_token_ids is not None, (
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore pid=351) ERROR 06-11 00:48:03 [core.py:1167] AssertionError: M-RoPE requires prompt_token_ids to be available.
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704] AsyncLLM output_handler failed.
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704] Traceback (most recent call last):
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/async_llm.py", line 660, in output_handler
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704]     outputs = await engine_core.get_output_async()
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704]               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704]   File "/usr/local/lib/python3.12/dist-packages/vllm/v1/engine/core_client.py", line 1030, in get_output_async
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704]     raise self._format_exception(outputs) from None
(APIServer pid=1) ERROR 06-11 00:48:03 [async_llm.py:704] vllm.v1.engine.exceptions.EngineDeadError: EngineCore encountered an issue. See stack trace (above) for the root cause.

All requests using the `/v1/chat/completions` endpoints will have text/`prompt_token_ids` parts (corresponding to the chat template), and `prompt_embeds` parts are handled as mm_features. This method (rightly) filters out those `prompt_embeds` content parts as they are treated as text positions.

A sufficient solution to type narrowing here without raising a fatal assertion is to instead replace the assertion with a using dummy token ids:

def _init_mrope_positions(self, req_state: CachedRequestState):
    model = self.get_model()
    assert supports_mrope(model), "M-RoPE support is not implemented."
    mrope_model = cast(SupportsMRoPE, model)

    # Filter out prompt_embeds modality (text-only position info)
    mrope_features = [
        f for f in req_state.mm_features if f.modality != "prompt_embeds"
    ]
    
    # Handle both token_ids and embeddings-only inputs
    if req_state.prompt_token_ids is not None:
        input_tokens = req_state.prompt_token_ids
    elif req_state.prompt_embeds is not None:
        # For text-only embeddings, dummy token IDs are safe since
        # get_mrope_input_positions only uses len(input_tokens) when mm_features is empty
        seq_len = req_state.prompt_embeds.shape[0]
        input_tokens = list(range(seq_len))
        # Verify no mm_features remain (should be true after prompt_embeds filter)
        assert len(mrope_features) == 0, (
            "M-RoPE with prompt_embeds-only input should have no multimodal features"
        )
    else:
        raise ValueError(
            "M-RoPE requires either prompt_token_ids or prompt_embeds."
        )

    req_state.mrope_positions, req_state.mrope_position_delta = (
        mrope_model.get_mrope_input_positions(
            input_tokens,
            mrope_features,
        )
    )

Technically, in isolation, this method still crashes in the case where `req_state.prompt_token_ids is None and req_state.mm_features`, so the solution above still leaves that potential vector open. As far as can be determined, however, such a `req_state` is impossible in the first place in online mode, because it would require a `/v1/completions` request with `prompt_embeds` AND multimodal features, but the `/v1/completions` request schema does not expose multimodal inputs in any discernible way. Today, those are the only two endpoints with `prompt_embeds` support.

When in offline mode, it *is technically* possible to directly create an `EngineCoreRequest` that has `prompt_embeds and not prompt_token_ids and mm_features`, and pass that to `LLM.generate`. That would trigger this same assertion, and no validation would prevent that combination. It is strongly suspected, though, that this combination would be undefined in any model that support M-RoPE, because it would not be possible to determine which token positions correspond to `mm_features`. The proposed solution above would end up not setting `req_state.mrope_positions` and `req_state.mrope_position_delta` in this scenario, which could result in undefined behavior.

`prompt_embeds` is far more familiar here than M-RoPE, and it is understood that each model that supports it is responsible for defining its own `get_mrope_input_positions` which have varying implementations. There is insufficient knowledge to be prescriptive in how the two features should interact in the offline case, other than possibly raising a validation error earlier on preventing that combination (which would emulate the current assertion behavior). Regardless, in offline mode, the chances of a remote user being able to exploit this are slim-to-nil compared to the online case which is incredibly straightforward.

### Impact
_What kind of vulnerability is it? Who is impacted?_

  • Denial of Service caused by an incorrect assertion inside of the `GPUModelRunner` which causes a fatal EngineCore exception
  • Any configuration with `--enable-prompt-embeds` and M-RoPE-supported model is vulnerable
  • The attack is extremely easy from the remote attacker's perspective (copying the official `prompt_embeds` online mode docs examples almost-verbatim, accounting for model-name and connection details, of course, will induce a guaranteed shutdown)

Incluye detalles técnicos publicados por el investigador. Úselos solo para validar y corregir sus propios sistemas.

Versiones

Rangos afectados

Afectada Corregida Última
vllm≥ 0.12.0 < 0.24.0
0.12.0
0.24.0
0.34.0
Remediación

Estado del parche por paquete

Paquete / productoEstadoCorregida enDesde divulgaciónÚltima versión
pypivllm
Corregida 0.24.0 -20 d 0.34.0 no afectada
Versiones corregidas0.24.0
Explotación

Arsenal público

Sin exploits, módulos ni PoC públicos conocidos a la fecha de verificación.

Referencias
Advisoryhttps://nvd.nist.gov/vuln/detail/CVE-2026-55514
Otrahttps://github.com/vllm-project/vllm/security/advisories/GHSA-33cg-gxv8-3p8g
Otrahttps://github.com/vllm-project/vllm/pull/45252
Otrahttps://github.com/vllm-project/vllm/commit/470229c37efaf69c86e8bc97482b0b1ff7551c65
Otrahttps://github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2026-2303.yaml
Otrahttps://github.com/vllm-project/vllm
Otrahttps://github.com/vllm-project/vllm/releases/tag/v0.24.0

Las referencias externas se muestran para copiar: cve.llama.ec no lo redirige a otros sitios.