{"id":"CVE-2026-73560","aliases":["EUVD-2026-60377","GHSA-4hhp-h66f-j5j7","PYSEC-2026-3934","openSUSE-SU-2026:12185-1"],"status":"published","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to the vLLM process. This issue is fixed in version 0.26.0.","severity":{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-11","score":0.00407,"percentile":0.35486,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-918"],"published":"2026-08-17T20:17:25Z","modified":"2026-11-04T02:45:46Z","fixes":[{"ecosystem":"PyPI","package":"vllm","fixed_in":["0.26.0"],"sources":["ghsa","osv"]},{"ecosystem":"openSUSE:Tumbleweed","package":"python-vllm","fixed_in":["0.32.0-1.1"],"sources":["osv"]}],"affected":[{"source":"ghsa","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.26.0"}]}]},{"source":"osv","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.26.0"}]}],"versions":["0.0.1","0.1.0","0.1.1","0.1.2","0.1.3","0.1.4","0.1.5","0.1.6","0.1.7","0.10.0","0.10.1","0.10.1.1","0.10.2","0.11.0","0.11.1","0.11.2","0.12.0","0.13.0","0.14.0","0.14.1","0.15.0","0.15.1","0.16.0","0.17.0","0.17.1","0.18.0","0.18.1","0.19.0","0.19.1","0.2.0","0.2.1","0.2.1.post1","0.2.2","0.2.3","0.2.4","0.2.5","0.2.6","0.2.7","0.20.0","0.20.1","0.20.2","0.21.0","0.22.0","0.22.1","0.23.0","0.24.0","0.25.0","0.25.1","0.3.0","0.3.1","0.3.2","0.3.3","0.4.0","0.4.0.post1","0.4.1","0.4.2","0.4.3","0.5.0","0.5.0.post1","0.5.1","0.5.2","0.5.3","0.5.3.post1","0.5.4","0.5.5","0.6.0","0.6.1","0.6.1.post1","0.6.1.post2","0.6.2","0.6.3","0.6.3.post1","0.6.4","0.6.4.post1","0.6.5","0.6.6","0.6.6.post1","0.7.0","0.7.1","0.7.2","0.7.3","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.5.post1","0.9.0","0.9.0.1","0.9.1","0.9.2"]},{"source":"osv","ecosystem":"PyPI","package":"vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.26.0"}]}],"versions":["0.0.1","0.1.0","0.1.1","0.1.2","0.1.3","0.1.4","0.1.5","0.1.6","0.1.7","0.10.0","0.10.1","0.10.1.1","0.10.2","0.11.0","0.11.1","0.11.2","0.12.0","0.13.0","0.14.0","0.14.1","0.15.0","0.15.1","0.16.0","0.17.0","0.17.1","0.18.0","0.18.1","0.19.0","0.19.1","0.2.0","0.2.1","0.2.1.post1","0.2.2","0.2.3","0.2.4","0.2.5","0.2.6","0.2.7","0.20.0","0.20.1","0.20.2","0.21.0","0.22.0","0.22.1","0.23.0","0.24.0","0.25.0","0.25.1","0.3.0","0.3.1","0.3.2","0.3.3","0.4.0","0.4.0.post1","0.4.1","0.4.2","0.4.3","0.5.0","0.5.0.post1","0.5.1","0.5.2","0.5.3","0.5.3.post1","0.5.4","0.5.5","0.6.0","0.6.1","0.6.1.post1","0.6.1.post2","0.6.2","0.6.3","0.6.3.post1","0.6.4","0.6.4.post1","0.6.5","0.6.6","0.6.6.post1","0.7.0","0.7.1","0.7.2","0.7.3","0.8.0","0.8.1","0.8.2","0.8.3","0.8.4","0.8.5","0.8.5.post1","0.9.0","0.9.0.1","0.9.1","0.9.2"]},{"source":"osv","ecosystem":"openSUSE:Tumbleweed","package":"python-vllm","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"0.32.0-1.1"}]}]}],"products":[{"vendor":"vllm","product":"vllm"},{"vendor":"vllm-project","product":"vllm"}],"references":[{"url":"https://github.com/vllm-project/vllm/security/advisories/GHSA-4hhp-h66f-j5j7","tags":["x_refsource_CONFIRM","exploit","Mitigation","Vendor Advisory","advisory","web"]},{"url":"https://github.com/vllm-project/vllm/pull/43117","tags":["x_refsource_MISC","Issue Tracking","Patch","fix","web"]},{"url":"https://github.com/vllm-project/vllm/commit/54503ecec0f3ac31e5ecfc5f28652e4cc42307b5","tags":["x_refsource_MISC","Patch","fix","web"]},{"url":"https://github.com/vllm-project/vllm/releases/tag/v0.26.0","tags":["x_refsource_MISC","Release Notes","web"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-73560","tags":["advisory"]},{"url":"https://github.com/advisories/GHSA-4hhp-h66f-j5j7","tags":["advisory"]},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/73xxx/CVE-2026-73560.json","tags":["advisory"]},{"url":"https://github.com/vllm-project/vllm","tags":["package"]},{"url":"https://pypi.org/project/vllm","tags":["package"]},{"url":"","tags":["advisory"]},{"url":"https://www.suse.com/security/cve/CVE-2026-73560","tags":["web"]}],"sources":{"cve":[{"id":"CVE-2026-73560","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to the vLLM process. This issue is fixed in version 0.26.0.","status":"published","published":"2026-08-17T20:17:25.769Z","modified":"2026-08-18T16:00:35.302Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium","assessor":"cna"}],"cwes":["CWE-918"],"updated_at":"2026-11-02T13:43:33.766176Z"}],"euvd":[{"id":"EUVD-2026-60377","description":"vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to the vLLM process. This issue is fixed in version 0.26.0.","status":"published","published":"2026-08-17T20:17:25Z","modified":"2026-08-18T16:00:35Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium"}],"updated_at":"2026-11-02T18:01:46.405062Z"}],"ghsa":[{"id":"GHSA-4hhp-h66f-j5j7","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"### Summary\n\n`vllm/transformers_utils/processors/mimo_v2_omni.py` — the multimodal processor for `MiMoV2OmniForCausalLM` — issues `requests.get(...)` directly on user-supplied image and audio URL strings and `Image.open(...)` on user-supplied local paths, **without** the SSRF / `allowed_local_media_path` checks that `vllm.multimodal.utils.MediaConnector` was hardened with in **GHSA-qh4c-xf7m-gxfc**, **GHSA-v359-jj2v-j536**, and **GHSA-pf3h-qjgv-vcpr**.\n\nThis is the same bug class as those three published advisories, in a code path the patches missed. When a user passes a URL or local-file string through `multi_modal_data` (e.g. `LLM.generate(multi_modal_data={\"image\": \"http://...\"})`), the processor takes the unsanitized string and dispatches it without any URL-scheme allowlist, network-target allowlist, size cap, or local-path allowlist.\n\n### Details\n\n**File:** `vllm/transformers_utils/processors/mimo_v2_omni.py` (current `main`)\n\n**Sink 1 — image SSRF + local-file read (`_fetch_image`, lines 231–249):**\n\n```python\ndef _fetch_image(src: Any) -\u003e Image.Image:\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        if src.startswith((\"http://\", \"https://\")):\n            r = requests.get(src, timeout=30)              # SSRF: no allowlist, follows redirects\n            r.raise_for_status()\n            return _to_rgb(copy.deepcopy(Image.open(BytesIO(r.content))))\n        if src.startswith(\"file://\"):\n            return _to_rgb(Image.open(src[7:]))            # arbitrary local file read\n        if src.startswith(\"data:image\"):\n            ...\n        return _to_rgb(Image.open(src))                    # fallback also opens local files\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\n**Sink 2 — audio SSRF (around line 471):**\n\n```python\nelif audio.startswith((\"http://\", \"https://\")):\n    r = requests.get(audio, timeout=30)                    # SSRF: same pattern\n    r.raise_for_status()\n    file_obj = io.BytesIO(r.content)\n```\n\n**Reachability.** `_fetch_image` is invoked from `MiMoVLProcessor.process_image`:\n\n```python\ndef process_image(self, image: ImageInput) -\u003e torch.Tensor:\n    kw = self._resolve_img_kw(image)\n    src = image.image\n    if isinstance(src, (str, bytes)):\n        src = _fetch_image(src)\n    ...\n```\n\n`MiMoVLProcessor` is wrapped by `MiMoV2OmniMultiModalProcessor` and registered for the `MiMoV2OmniForCausalLM` model architecture (`vllm/model_executor/models/mimo_v2_omni.py:1169`). Whenever a user passes a string into `multi_modal_data[\"image\"]` (or `[\"audio\"]`) for this model, the unsanitized URL/path reaches the sink.\n\n**Comparison to the recent fixes.** The remediation pattern adopted in the three earlier advisories was to route every external resource fetch through `MediaConnector`, which checks `allowed_local_media_path` and applies SSRF protection before issuing the network request. `chat_utils.py` (lines 838, 902, 924, 963, 1053, 1081) already uses `self._connector.fetch_image / fetch_audio / fetch_video`. The model processor in `mimo_v2_omni.py` was added later and skipped the connector — it calls `requests.get` and `Image.open` directly. Result: the public OpenAI chat-completion path is protected, but library use (`LLM.generate(multi_modal_data=...)`), batch processing, and any other path that lets a string reach the processor receive no protection.\n\n### Impact\n\n1. **SSRF — internal-network probing / cloud-metadata theft.** Standard `requests.get` follows redirects and accepts any URL. An attacker who controls a `multi_modal_data` value can:\n   - read AWS / GCP / Azure instance metadata (e.g. `http://169.254.169.254/latest/meta-data/iam/security-credentials/`),\n   - probe internal services on the vLLM host (`http://127.0.0.1:\u003cport\u003e`, `http://10.x.y.z`),\n   - exfiltrate via DNS / HTTP timing oracles even when the body is rejected by `Image.open`.\n2. **Arbitrary local file read** via `file://path` (line 242) and the unguarded fallback `Image.open(src)` (line 248). Any file readable by the vLLM process is reachable through the model pipeline; with suitable formats this exposes `/etc/passwd`, `~/.aws/credentials`, etc.\n3. **Server-side traffic generation / amplification** by hammering arbitrary URLs from the vLLM host, with a 30-second timeout per request.\n\n### Suggested remediation\n\nReplace direct `requests.get` and bare `Image.open` paths with `MediaConnector.fetch_image` / `fetch_audio_async` (or pass the inputs through `MediaConnector` before they reach the processor):\n\n```python\n# vllm/transformers_utils/processors/mimo_v2_omni.py\nfrom vllm.multimodal.utils import MediaConnector\n\n_connector = MediaConnector()\n\ndef _fetch_image(src):\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        return _to_rgb(_connector.fetch_image(src))   # delegates to the hardened path\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\nSame change for the audio loader at line 471. This re-uses the SSRF allowlist, `allowed_local_media_path` policy, and size caps that the previous patches added.\n\nAlternative: forbid `str` `src` from reaching the processor and require all multi-modal pre-processing to go through `chat_utils.py` / `MediaConnector` before hitting the model. Larger surface change, but completes the architectural fix.\n\n### Discovery\n\nStatic review on `vllm@main` (HEAD as of 2026-04-30) — found by triaging the file list against the three recent SSRF advisories: the `mimo_v2_omni.py` processor, added after those fixes, reintroduced the same bypass class.\n\n### Reporter\n\nIevgen Bondarenko — `sactransport2000@gmail.com` — GitHub `@ibondarenko1`","status":"published","published":"2026-09-08T20:42:00Z","modified":"2026-09-08T20:42:01Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium"}],"cwes":["CWE-918"],"updated_at":"2026-11-02T14:41:14.764963Z"}],"nvd":[{"id":"CVE-2026-73560","description":"vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to the vLLM process. This issue is fixed in version 0.26.0.","status":"published","published":"2026-08-17T21:16:48.96Z","modified":"2026-10-23T16:10:00.313Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium","assessor":"secondary"}],"cwes":["CWE-918"],"updated_at":"2026-11-02T14:12:38.647684Z"}],"osv":[{"id":"CVE-2026-73560","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the MiMoV2OmniMultiModalProcessor in vllm/transformers_utils/processors/mimo_v2_omni.py passes attacker-controlled image and audio strings through _fetch_image, requests.get, and Image.open instead of MediaConnector, bypassing allowed_media_domains and allowed_local_media_path protections and allowing server-side requests and reads of arbitrary files accessible to the vLLM process. This issue is fixed in version 0.26.0.","status":"published","published":"2026-08-17T20:17:25.769Z","modified":"2026-11-10T02:45:15.756524807Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium"}],"cwes":["CWE-918"],"updated_at":"2026-11-10T03:16:41.950556Z"},{"id":"GHSA-4hhp-h66f-j5j7","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"### Summary\n\n`vllm/transformers_utils/processors/mimo_v2_omni.py` — the multimodal processor for `MiMoV2OmniForCausalLM` — issues `requests.get(...)` directly on user-supplied image and audio URL strings and `Image.open(...)` on user-supplied local paths, **without** the SSRF / `allowed_local_media_path` checks that `vllm.multimodal.utils.MediaConnector` was hardened with in **GHSA-qh4c-xf7m-gxfc**, **GHSA-v359-jj2v-j536**, and **GHSA-pf3h-qjgv-vcpr**.\n\nThis is the same bug class as those three published advisories, in a code path the patches missed. When a user passes a URL or local-file string through `multi_modal_data` (e.g. `LLM.generate(multi_modal_data={\"image\": \"http://...\"})`), the processor takes the unsanitized string and dispatches it without any URL-scheme allowlist, network-target allowlist, size cap, or local-path allowlist.\n\n### Details\n\n**File:** `vllm/transformers_utils/processors/mimo_v2_omni.py` (current `main`)\n\n**Sink 1 — image SSRF + local-file read (`_fetch_image`, lines 231–249):**\n\n```python\ndef _fetch_image(src: Any) -\u003e Image.Image:\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        if src.startswith((\"http://\", \"https://\")):\n            r = requests.get(src, timeout=30)              # SSRF: no allowlist, follows redirects\n            r.raise_for_status()\n            return _to_rgb(copy.deepcopy(Image.open(BytesIO(r.content))))\n        if src.startswith(\"file://\"):\n            return _to_rgb(Image.open(src[7:]))            # arbitrary local file read\n        if src.startswith(\"data:image\"):\n            ...\n        return _to_rgb(Image.open(src))                    # fallback also opens local files\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\n**Sink 2 — audio SSRF (around line 471):**\n\n```python\nelif audio.startswith((\"http://\", \"https://\")):\n    r = requests.get(audio, timeout=30)                    # SSRF: same pattern\n    r.raise_for_status()\n    file_obj = io.BytesIO(r.content)\n```\n\n**Reachability.** `_fetch_image` is invoked from `MiMoVLProcessor.process_image`:\n\n```python\ndef process_image(self, image: ImageInput) -\u003e torch.Tensor:\n    kw = self._resolve_img_kw(image)\n    src = image.image\n    if isinstance(src, (str, bytes)):\n        src = _fetch_image(src)\n    ...\n```\n\n`MiMoVLProcessor` is wrapped by `MiMoV2OmniMultiModalProcessor` and registered for the `MiMoV2OmniForCausalLM` model architecture (`vllm/model_executor/models/mimo_v2_omni.py:1169`). Whenever a user passes a string into `multi_modal_data[\"image\"]` (or `[\"audio\"]`) for this model, the unsanitized URL/path reaches the sink.\n\n**Comparison to the recent fixes.** The remediation pattern adopted in the three earlier advisories was to route every external resource fetch through `MediaConnector`, which checks `allowed_local_media_path` and applies SSRF protection before issuing the network request. `chat_utils.py` (lines 838, 902, 924, 963, 1053, 1081) already uses `self._connector.fetch_image / fetch_audio / fetch_video`. The model processor in `mimo_v2_omni.py` was added later and skipped the connector — it calls `requests.get` and `Image.open` directly. Result: the public OpenAI chat-completion path is protected, but library use (`LLM.generate(multi_modal_data=...)`), batch processing, and any other path that lets a string reach the processor receive no protection.\n\n### Impact\n\n1. **SSRF — internal-network probing / cloud-metadata theft.** Standard `requests.get` follows redirects and accepts any URL. An attacker who controls a `multi_modal_data` value can:\n   - read AWS / GCP / Azure instance metadata (e.g. `http://169.254.169.254/latest/meta-data/iam/security-credentials/`),\n   - probe internal services on the vLLM host (`http://127.0.0.1:\u003cport\u003e`, `http://10.x.y.z`),\n   - exfiltrate via DNS / HTTP timing oracles even when the body is rejected by `Image.open`.\n2. **Arbitrary local file read** via `file://path` (line 242) and the unguarded fallback `Image.open(src)` (line 248). Any file readable by the vLLM process is reachable through the model pipeline; with suitable formats this exposes `/etc/passwd`, `~/.aws/credentials`, etc.\n3. **Server-side traffic generation / amplification** by hammering arbitrary URLs from the vLLM host, with a 30-second timeout per request.\n\n### Suggested remediation\n\nReplace direct `requests.get` and bare `Image.open` paths with `MediaConnector.fetch_image` / `fetch_audio_async` (or pass the inputs through `MediaConnector` before they reach the processor):\n\n```python\n# vllm/transformers_utils/processors/mimo_v2_omni.py\nfrom vllm.multimodal.utils import MediaConnector\n\n_connector = MediaConnector()\n\ndef _fetch_image(src):\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        return _to_rgb(_connector.fetch_image(src))   # delegates to the hardened path\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\nSame change for the audio loader at line 471. This re-uses the SSRF allowlist, `allowed_local_media_path` policy, and size caps that the previous patches added.\n\nAlternative: forbid `str` `src` from reaching the processor and require all multi-modal pre-processing to go through `chat_utils.py` / `MediaConnector` before hitting the model. Larger surface change, but completes the architectural fix.\n\n### Discovery\n\nStatic review on `vllm@main` (HEAD as of 2026-04-30) — found by triaging the file list against the three recent SSRF advisories: the `mimo_v2_omni.py` processor, added after those fixes, reintroduced the same bypass class.\n\n### Reporter\n\nIevgen Bondarenko — `sactransport2000@gmail.com` — GitHub `@ibondarenko1`","status":"published","published":"2026-09-08T20:42:00Z","modified":"2026-09-10T12:25:25.165237414Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium"}],"cwes":["CWE-918"],"updated_at":"2026-11-02T15:46:37.148947Z"},{"id":"PYSEC-2026-3934","title":"vLLM: SSRF + arbitrary local file read in MiMoV2OmniMultiModalProcessor `_fetch_image` and audio loader bypass MediaConnector protections","description":"### Summary\n\n`vllm/transformers_utils/processors/mimo_v2_omni.py` — the multimodal processor for `MiMoV2OmniForCausalLM` — issues `requests.get(...)` directly on user-supplied image and audio URL strings and `Image.open(...)` on user-supplied local paths, **without** the SSRF / `allowed_local_media_path` checks that `vllm.multimodal.utils.MediaConnector` was hardened with in **GHSA-qh4c-xf7m-gxfc**, **GHSA-v359-jj2v-j536**, and **GHSA-pf3h-qjgv-vcpr**.\n\nThis is the same bug class as those three published advisories, in a code path the patches missed. When a user passes a URL or local-file string through `multi_modal_data` (e.g. `LLM.generate(multi_modal_data={\"image\": \"http://...\"})`), the processor takes the unsanitized string and dispatches it without any URL-scheme allowlist, network-target allowlist, size cap, or local-path allowlist.\n\n### Details\n\n**File:** `vllm/transformers_utils/processors/mimo_v2_omni.py` (current `main`)\n\n**Sink 1 — image SSRF + local-file read (`_fetch_image`, lines 231–249):**\n\n```python\ndef _fetch_image(src: Any) -\u003e Image.Image:\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        if src.startswith((\"http://\", \"https://\")):\n            r = requests.get(src, timeout=30)              # SSRF: no allowlist, follows redirects\n            r.raise_for_status()\n            return _to_rgb(copy.deepcopy(Image.open(BytesIO(r.content))))\n        if src.startswith(\"file://\"):\n            return _to_rgb(Image.open(src[7:]))            # arbitrary local file read\n        if src.startswith(\"data:image\"):\n            ...\n        return _to_rgb(Image.open(src))                    # fallback also opens local files\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\n**Sink 2 — audio SSRF (around line 471):**\n\n```python\nelif audio.startswith((\"http://\", \"https://\")):\n    r = requests.get(audio, timeout=30)                    # SSRF: same pattern\n    r.raise_for_status()\n    file_obj = io.BytesIO(r.content)\n```\n\n**Reachability.** `_fetch_image` is invoked from `MiMoVLProcessor.process_image`:\n\n```python\ndef process_image(self, image: ImageInput) -\u003e torch.Tensor:\n    kw = self._resolve_img_kw(image)\n    src = image.image\n    if isinstance(src, (str, bytes)):\n        src = _fetch_image(src)\n    ...\n```\n\n`MiMoVLProcessor` is wrapped by `MiMoV2OmniMultiModalProcessor` and registered for the `MiMoV2OmniForCausalLM` model architecture (`vllm/model_executor/models/mimo_v2_omni.py:1169`). Whenever a user passes a string into `multi_modal_data[\"image\"]` (or `[\"audio\"]`) for this model, the unsanitized URL/path reaches the sink.\n\n**Comparison to the recent fixes.** The remediation pattern adopted in the three earlier advisories was to route every external resource fetch through `MediaConnector`, which checks `allowed_local_media_path` and applies SSRF protection before issuing the network request. `chat_utils.py` (lines 838, 902, 924, 963, 1053, 1081) already uses `self._connector.fetch_image / fetch_audio / fetch_video`. The model processor in `mimo_v2_omni.py` was added later and skipped the connector — it calls `requests.get` and `Image.open` directly. Result: the public OpenAI chat-completion path is protected, but library use (`LLM.generate(multi_modal_data=...)`), batch processing, and any other path that lets a string reach the processor receive no protection.\n\n### Impact\n\n1. **SSRF — internal-network probing / cloud-metadata theft.** Standard `requests.get` follows redirects and accepts any URL. An attacker who controls a `multi_modal_data` value can:\n   - read AWS / GCP / Azure instance metadata (e.g. `http://169.254.169.254/latest/meta-data/iam/security-credentials/`),\n   - probe internal services on the vLLM host (`http://127.0.0.1:\u003cport\u003e`, `http://10.x.y.z`),\n   - exfiltrate via DNS / HTTP timing oracles even when the body is rejected by `Image.open`.\n2. **Arbitrary local file read** via `file://path` (line 242) and the unguarded fallback `Image.open(src)` (line 248). Any file readable by the vLLM process is reachable through the model pipeline; with suitable formats this exposes `/etc/passwd`, `~/.aws/credentials`, etc.\n3. **Server-side traffic generation / amplification** by hammering arbitrary URLs from the vLLM host, with a 30-second timeout per request.\n\n### Suggested remediation\n\nReplace direct `requests.get` and bare `Image.open` paths with `MediaConnector.fetch_image` / `fetch_audio_async` (or pass the inputs through `MediaConnector` before they reach the processor):\n\n```python\n# vllm/transformers_utils/processors/mimo_v2_omni.py\nfrom vllm.multimodal.utils import MediaConnector\n\n_connector = MediaConnector()\n\ndef _fetch_image(src):\n    if isinstance(src, Image.Image):\n        return _to_rgb(src)\n    if isinstance(src, bytes):\n        return _to_rgb(copy.deepcopy(Image.open(BytesIO(src))))\n    if isinstance(src, str):\n        return _to_rgb(_connector.fetch_image(src))   # delegates to the hardened path\n    raise ValueError(f\"Unrecognized image source: {type(src)}\")\n```\n\nSame change for the audio loader at line 471. This re-uses the SSRF allowlist, `allowed_local_media_path` policy, and size caps that the previous patches added.\n\nAlternative: forbid `str` `src` from reaching the processor and require all multi-modal pre-processing to go through `chat_utils.py` / `MediaConnector` before hitting the model. Larger surface change, but completes the architectural fix.\n\n### Discovery\n\nStatic review on `vllm@main` (HEAD as of 2026-04-30) — found by triaging the file list against the three recent SSRF advisories: the `mimo_v2_omni.py` processor, added after those fixes, reintroduced the same bypass class.\n\n### Reporter\n\nIevgen Bondarenko — `sactransport2000@gmail.com` — GitHub `@ibondarenko1`","status":"published","published":"2026-09-10T09:45:00.831333Z","modified":"2026-09-10T12:15:15.566777999Z","cvss":[{"version":"3.1","score":6.5,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N","level":"medium"}],"updated_at":"2026-11-02T15:52:57.255153Z"},{"id":"openSUSE-SU-2026:12185-1","title":"python313-vllm-0.32.0-1.1 on GA media","description":"These are all security issues fixed in the python313-vllm-0.32.0-1.1 package on the GA media of openSUSE Tumbleweed.","status":"published","published":"2026-10-18T00:00:00Z","modified":"2026-10-19T09:30:05.163668446Z","updated_at":"2026-11-02T16:58:40.245664Z"}]}}