← Packages
python-vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
- Current
- 0.27.1
- Candidate
- 0.33.0
- License
- Apache-2.0
RuntimeDeps
| Dependency | Observed source | Current requirement | Candidate requirement |
|---|---|---|---|
3.14.5 | |||
0.105.2 | |||
0.1.10 | |||
1.0.8 | |||
7.1.1 | |||
6.1.2 | |||
2026.10.19 | Not observed | ||
3.1.2 | |||
0.15.0.1 | |||
0.20.0 | |||
0.8.2 | |||
0.136.1 | |||
fastsafetensorsUnmapped | — | ||
3.29.0 | |||
flashinfer-pythonUnmapped | — | ||
1.16.0 | Not observed | ||
humming-kernels[cu13]Unmapped | — | ||
3.5.0 | |||
instanttensorUnmapped | — | Not observed | |
4.26.0 | |||
1.3.1 | |||
1.7.5 | |||
0.11.3 | |||
2.0.0 | |||
1.11.3 | |||
0.1.14 | |||
0.19.0 | |||
7.2.1 | Not observed | ||
1.13.2 | |||
0.66.0 | |||
2.4.6 | |||
nvidia-cudnn-frontendUnmapped | — | ||
nvidia-cutlass-dsl[cu13]Unmapped | — | ||
nvtxUnmapped | — | ||
2.40.0 | |||
0.0.8 | Not observed | ||
opencv-python-headlessUnmapped | — | ||
1.44.0 | |||
1.42.1 | |||
1.42.1 | |||
0.5.1 | |||
3.11.7 | Not observed | ||
oss-harmonyUnmapped | — | Not observed | |
0.2.14 | |||
0.2.1.1~post7 | |||
13.0.0 | |||
0.25.0 | |||
7.1.0 | |||
7.35.1 | Not observed | ||
7.35.1 | Not observed | ||
7.35.1 | Not observed | ||
7.2.2 | |||
9.0.0 | |||
1.4.3 | |||
2.12.5 | |||
PyNvVideoCodecUnmapped | — | ||
4.1.0 | |||
6.0.3 | |||
27.1.0 | |||
quack-kernelsUnmapped | — | ||
2026.1.15 | |||
2.33.1 | |||
0.8.0 | |||
0.2.1 | |||
1.1.10 | |||
83.0.0 | |||
1.17.0 | |||
1.6.0 | |||
0.12.0 | |||
tilelangUnmapped | — | ||
0.22.2 | |||
tokenspeed-mlaUnmapped | — | ||
2.13.0 | |||
2.11.0 | |||
torchcodecUnmapped | — | ||
0.28.0 | |||
4.67.3 | |||
5.12.1 | |||
4.16.0 | |||
1.2.0 | |||
0.2.1 | |||
3.14.9 |
Optional RuntimeDeps
| Dependency | Observed source | Current requirement | Candidate requirement |
|---|---|---|---|
17.0.1 | |||
b12xUnmapped | — | Not observed | |
5.0.1 | |||
fastsafetensorsUnmapped | — | ||
helionUnmapped | — | ||
instanttensorUnmapped | — | ||
3.11.0 | |||
1.11.3 | |||
nvidia-deepstream-videodecode-cu13Unmapped | — | ||
1.44.0 | |||
1.42.1 | |||
1.42.1 | |||
0.5.1 | |||
3.0.1 | |||
plotlyUnmapped | — | ||
runai-model-streamer[azure, gcs, s3]Unmapped | — | ||
1.17.1 | |||
seabornUnmapped | — | ||
smg-grpc-servicer[vllm]Unmapped | — | ||
0.13.1 | |||
1.0.0 | |||
2.12.1 | |||
vllm-gguf-pluginUnmapped | — | ||
zentorchUnmapped | — |
Dependency conditions
- Applies when (any)
- platform_machine == "x86_64" or platform_machine == "arm64" or platform_machine == "aarch64" or platform_machine == "ppc64le" or platform_machine == "s390x"
- Applies when (any)
- platform_machine != "s390x"
- Applies when (any)
- platform_machine == "s390x"
- Applies when (any)
- python_version > "3.11"
- Applies when (any)
- python_version > "3.11"
- Applies when (any)
- platform_system == "Linux"
- Applies when (any)
- platform_machine == "x86_64" or platform_machine == "aarch64" or platform_machine == "arm64" or platform_machine == "s390x" or platform_machine == "ppc64le"
- Applies when (any)
- extra == "audio"
- Applies when (any)
- extra == "b12x"
- Applies when (any)
- extra == "bench"
- Applies when (any)
- extra == "fastsafetensors"
- Applies when (any)
- extra == "helion"
- Applies when (any)
- extra == "instanttensor"
- Applies when (any)
- extra == "bench"
- Applies when (any)
- extra == "audio"
- Applies when (any)
- extra == "deepstream"
- Applies when (any)
- extra == "otel"
- Applies when (any)
- extra == "otel"
- Applies when (any)
- extra == "otel"
- Applies when (any)
- extra == "otel"
- Applies when (any)
- extra == "bench"
- Applies when (any)
- extra == "bench"
- Applies when (any)
- extra == "runai"
- Applies when (any)
- extra == "audio"extra == "bench"
- Applies when (any)
- extra == "bench"
- Applies when (any)
- extra == "grpc"
- Applies when (any)
- extra == "audio"
- Applies when (any)
- extra == "audio"
- Applies when (any)
- extra == "tensorizer"
- Applies when (any)
- extra == "extra-quant"
- Applies when (any)
- extra == "zen"
Advisory · 38
Provider matches do not establish RPM applicability. Local patches, bundled dependencies and binary artifacts are not evaluated.
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.808%FIRST
- Returned advisory
- PYSEC-2026-4184OSV
- Aliases
- GHSA-p6g9-7v3x-m8mv, PYSEC-2026-4184OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.28.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.34%FIRST
- Returned advisory
- PYSEC-2026-4324OSV
- Aliases
- GHSA-hhv2-872h-628q, PYSEC-2026-4324OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Summary
- vLLM: Harmony tool continuations drop `cache_salt` — restoring a cross-tenant prefix-cache membership oracleOSVOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.216%FIRST
- Returned advisory
- GHSA-935w-9g4m-p28pOSV
- Returned advisory
- PYSEC-2026-4308OSV
- Aliases
- GHSA-935w-9g4m-p28p, PYSEC-2026-4308OSV
- KEV
- NoCISA
- Summary
- vLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertion in the engine coreOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.427%FIRST
- Returned advisory
- GHSA-ph3r-5jfg-f84fOSV
- Returned advisory
- PYSEC-2026-4197OSV
- Aliases
- GHSA-ph3r-5jfg-f84f, PYSEC-2026-4197OSV
- KEV
- NoCISA
- Summary
- vLLM: Flash late-interaction scoring caches query embeddings under a caller-controlled request id — cross-request integrity break and induced errors on `/score` and `/rerank`OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.202%FIRST
- Returned advisory
- GHSA-2phq-3phc-84pxOSV
- Returned advisory
- PYSEC-2026-4198OSV
- Aliases
- GHSA-2phq-3phc-84px, PYSEC-2026-4198OSV
- KEV
- NoCISA
- Summary
- vLLM: Loose `cache_salt` validation lets a single request kill EngineCore on LMCache-MP deployments — uncaught downstream `ValueError` denial of serviceOSVOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.332%FIRST
- Returned advisory
- GHSA-2823-qmq8-rwvjOSV
- Returned advisory
- PYSEC-2026-4306OSV
- Aliases
- GHSA-2823-qmq8-rwvj, PYSEC-2026-4306OSV
- KEV
- NoCISA
- Summary
- vLLM: Structured-output request errors escape the request boundary and terminate the shared EngineCore — engine-fatal denial of service (3 sites)OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.314%FIRST
- Returned advisory
- GHSA-85xf-c7hm-whqwOSV
- Returned advisory
- PYSEC-2026-4199OSV
- Aliases
- GHSA-85xf-c7hm-whqw, PYSEC-2026-4199OSV
- KEV
- NoCISA
- Summary
- vLLM: Qwen2-VL / Qwen3-VL video samplers bound on request-controlled max_frames, which the num_frames ceiling does not reachOSVOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.303%FIRST
- Returned advisory
- GHSA-x6mc-67gf-chw4OSV
- Returned advisory
- PYSEC-2026-4310OSV
- Aliases
- GHSA-x6mc-67gf-chw4, PYSEC-2026-4310OSV
- KEV
- NoCISA
- Fixed events
- 0.32.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.311%FIRST
- Returned advisory
- PYSEC-2026-4319OSV
- Aliases
- PYSEC-2026-4319OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:P/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Summary
- vLLM: Forged multimodal UUID + P0/P1 processor-cache eviction drift: unauthenticated remote crash of the whole vLLM engineOSV
- Fixed events
- 0.31.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.207%FIRST
- Returned advisory
- GHSA-p92p-rxj5-7p2xOSV
- Aliases
- GHSA-p92p-rxj5-7p2xOSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:HOSV
- Fixed events
- 0.31.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.506%FIRST
- Returned advisory
- PYSEC-2026-4361OSV
- Aliases
- GHSA-823j-m4cj-hmjf, PYSEC-2026-4361OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Summary
- vLLM: Request-controlled mm_processor_kwargs.code_revision allows remote code executionOSV
- Fixed events
- 0.31.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.462%FIRST
- Returned advisory
- GHSA-h3rc-6mm3-gc2mOSV
- Aliases
- GHSA-h3rc-6mm3-gc2mOSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:HOSV
- Fixed events
- 0.31.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.275%FIRST
- Returned advisory
- PYSEC-2026-4362OSV
- Aliases
- GHSA-rh6f-3x46-j33q, PYSEC-2026-4362OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Summary
- vLLM : DoS via unvalidated mm_processor_kwargs in multimodal image processors (Isaac, InternVL, H2OVL, etc.)OSV
- Fixed events
- 0.31.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.42%FIRST
- Returned advisory
- GHSA-gx7p-2j49-hfq4OSV
- Aliases
- GHSA-gx7p-2j49-hfq4OSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:HOSV
- Summary
- vLLM: Public multimodal SHM handle forgery reaches worker-side `pickle.loads()` with insecure serialization disabledOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.51%FIRST
- Returned advisory
- GHSA-6757-hggm-hp2wOSV
- Returned advisory
- PYSEC-2026-4345OSV
- Aliases
- GHSA-6757-hggm-hp2w, PYSEC-2026-4345OSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:HOSV
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:HOSV
- Summary
- vLLM: Phi-4 multimodal audio encoder materializes and retains full attention matrices: a single max-length audio OOM-crashes the vLLM engineOSV
- Fixed events
- 0.32.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.329%FIRST
- Returned advisory
- GHSA-8397-8qf6-f2gmOSV
- Aliases
- GHSA-8397-8qf6-f2gmOSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:HOSV
- Summary
- vLLM: Untrusted EPS inputs can trigger denial of service or Ghostscript RCEOSV
- Fixed events
- 0.32.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.66%FIRST
- Returned advisory
- GHSA-h3rc-j82p-2c3hOSV
- Aliases
- GHSA-h3rc-j82p-2c3h, PYSEC-2026-4346OSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:HOSV
- Summary
- vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.312%FIRST
- Returned advisory
- GHSA-q97c-p368-75c4OSV
- Returned advisory
- PYSEC-2026-4347OSV
- Aliases
- GHSA-q97c-p368-75c4, PYSEC-2026-4347OSV
- KEV
- NoCISA
- Summary
- vLLM: Permissive default CORS origins allow cross-origin requests to the vLLM HTTP serverOSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.24%FIRST
- Returned advisory
- GHSA-vvpg-rrww-wpmhOSV
- Returned advisory
- PYSEC-2026-4348OSV
- Aliases
- GHSA-vvpg-rrww-wpmh, PYSEC-2026-4348OSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:L/I:N/A:NOSV
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:L/I:N/A:NOSV
- Fixed events
- 0.28.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.308%FIRST
- Returned advisory
- PYSEC-2026-3985OSV
- Aliases
- GHSA-3c86-2m5g-59q7, PYSEC-2026-3985OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.755%FIRST
- Returned advisory
- PYSEC-2026-3996OSV
- Aliases
- PYSEC-2026-3996OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.28.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.544%FIRST
- Returned advisory
- PYSEC-2026-3997OSV
- Aliases
- GHSA-25q3-v2hm-8vpf, PYSEC-2026-3997OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.627%FIRST
- Returned advisory
- PYSEC-2026-4004OSV
- Aliases
- PYSEC-2026-4004OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.627%FIRST
- Returned advisory
- PYSEC-2026-4005OSV
- Aliases
- PYSEC-2026-4005OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Summary
- vllm p2p never connected zmq_exhaustionOSV
- Fixed events
- 0.32.0OSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.434%FIRST
- Returned advisory
- GHSA-frvg-qw88-hww9OSV
- Returned advisory
- PYSEC-2026-4006OSV
- Aliases
- GHSA-frvg-qw88-hww9, PYSEC-2026-4006OSV
- KEV
- NoCISA
- CVSS_V3
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:HOSV
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.521%FIRST
- Returned advisory
- PYSEC-2026-4007OSV
- Aliases
- PYSEC-2026-4007OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:L/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
- Fixed events
- 0.30.0OSV
- EPSS model date
- 2026-11-09FIRST
- EPSS probability
- 0.627%FIRST
- Returned advisory
- PYSEC-2026-4008OSV
- Aliases
- PYSEC-2026-4008OSV
- KEV
- NoCISA
- CVSS_V4
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:XOSV
Description
vLLM is a fast and easy-to-use library for LLM inference and serving, featuring PagedAttention for efficient management of attention key/value memory, continuous batching of incoming requests, and an OpenAI-compatible API server.
Changelog
PR · 3
Checks
| Monitor | Observation | Last checked |
|---|---|---|
Checked | ||
Checked | ||
Checked | ||
Checked | ||
Not configured No reliable monitor identity/configuration. | — | |
Checked Compared same-provider, same-project SPDX metadata for this version pair. | Last attempted | |
Checked | Last attempted | |
Checked Current source component only; local patches, build options, bundled dependencies and binary artifacts are not evaluated. | Last attempted | |
Checked |