{"items":[{"id":"NVI-00009196","kind":"vulnerability","title":"vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)","revision":4,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-q97c-p368-75c4","CVE-2026-115916"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-2dcf6dd61c2d412ac2d415110c52f926"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-28T21:15:09.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-ef0c78e15322e8eb18b5559c566be136"},{"kind":"CVSS","as_of":"2026-11-02T18:18:39.033Z","value":"6.5 MEDIUM (3.1)","meaning":"security-advisories@github.com severity; no combined Null risk score.","source_id":"sr-0f8c568821c91650dd45e50a8647bb4f"},{"kind":"EPSS","as_of":"2026-11-01T12:00:16.000Z","value":"0.312%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.24098. Missing scores are unknown.","source_id":"sr-2dcf6dd61c2d412ac2d415110c52f926"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-ef0c78e15322e8eb18b5559c566be136","url":"https://github.com/advisories/GHSA-q97c-p368-75c4","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-q97c-p368-75c4","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T21:56:08.420Z"},{"id":"sr-0f8c568821c91650dd45e50a8647bb4f","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-115916","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-115916","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:59:00.529Z"},{"id":"sr-2dcf6dd61c2d412ac2d415110c52f926","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-115916","publisher":"FIRST EPSS","observed_at":"2026-11-01T14:28:32.267Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-ef0c78e15322e8eb18b5559c566be136"],"technology":"vllm","known_exclusions":[],"conditions_summary":"For a model whose vocabulary size is an exact multiple of the embedding padding size (64) — e.g. any Llama-2-family model, vocab 32000 — a single request carrying a pre-tokenized prompt with token id equal to the vocab size (32000) passes vLLM's input-id validation but is one past the last embedding row. At tensor-parallel size 1 the embedding lookup does not mask indices, so the id reaches `F.embedding` out of range and raises a CUDA device-side assert, which poisons the engine's CUDA context and takes the engine down. Any client that can send pre-tokenized input (`/v1/completions` with `prompt` as token ids, `/tokenize`-fed flows, or the `kv_transfer_params[\"prompt_token_ids\"]` reuse path) can trigger it.\n\n## Root cause\n`vllm/v1/engine/input_processor.py` accepts a token id when `id <= max(tokenizer.max_token_id, model_config.get_vocab_size() - 1)`.\n`vllm/tokenizers/hf.py` computes `max_token_id = max(max(vocab.values()), tokenizer.vocab_size)`. For Llama-2 tokenizers `max(vocab.values()) = 31999` but `tokenizer.vocab_size = 32000`, so `max_token_id = 32000` — one greater than the largest valid id.\nThe engine bound therefore becomes `max(32000, 31999) = 32000`, and id 32000 is ac","public_affected_ranges":["< 0.32.0"]}],"null_published_at":"2026-11-06T15:18:11.480Z","source_updated_at":"2026-10-28T21:15:09.000Z","source_disclosed_at":"2026-10-28T21:15:08.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Dated EPSS estimate updated; this is not a new disclosure or proof of exploitation.","null_first_observed_at":"2026-10-31T21:56:08.420Z"},{"id":"NVI-00009196","kind":"vulnerability","title":"vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)","revision":3,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-q97c-p368-75c4","CVE-2026-115916"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-a933358640dead761fd17020a4a66124"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-28T21:15:09.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-ef0c78e15322e8eb18b5559c566be136"},{"kind":"CVSS","as_of":"2026-11-02T18:18:39.033Z","value":"6.5 MEDIUM (3.1)","meaning":"security-advisories@github.com severity; no combined Null risk score.","source_id":"sr-0f8c568821c91650dd45e50a8647bb4f"},{"kind":"EPSS","as_of":"2026-10-31T12:00:27.000Z","value":"0.312%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.24103. Missing scores are unknown.","source_id":"sr-a933358640dead761fd17020a4a66124"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-ef0c78e15322e8eb18b5559c566be136","url":"https://github.com/advisories/GHSA-q97c-p368-75c4","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-q97c-p368-75c4","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T21:56:08.420Z"},{"id":"sr-0f8c568821c91650dd45e50a8647bb4f","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-115916","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-115916","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:59:00.529Z"},{"id":"sr-a933358640dead761fd17020a4a66124","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-115916","publisher":"FIRST EPSS","observed_at":"2026-10-31T22:06:15.374Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-ef0c78e15322e8eb18b5559c566be136"],"technology":"vllm","known_exclusions":[],"conditions_summary":"For a model whose vocabulary size is an exact multiple of the embedding padding size (64) — e.g. any Llama-2-family model, vocab 32000 — a single request carrying a pre-tokenized prompt with token id equal to the vocab size (32000) passes vLLM's input-id validation but is one past the last embedding row. At tensor-parallel size 1 the embedding lookup does not mask indices, so the id reaches `F.embedding` out of range and raises a CUDA device-side assert, which poisons the engine's CUDA context and takes the engine down. Any client that can send pre-tokenized input (`/v1/completions` with `prompt` as token ids, `/tokenize`-fed flows, or the `kv_transfer_params[\"prompt_token_ids\"]` reuse path) can trigger it.\n\n## Root cause\n`vllm/v1/engine/input_processor.py` accepts a token id when `id <= max(tokenizer.max_token_id, model_config.get_vocab_size() - 1)`.\n`vllm/tokenizers/hf.py` computes `max_token_id = max(max(vocab.values()), tokenizer.vocab_size)`. For Llama-2 tokenizers `max(vocab.values()) = 31999` but `tokenizer.vocab_size = 32000`, so `max_token_id = 32000` — one greater than the largest valid id.\nThe engine bound therefore becomes `max(32000, 31999) = 32000`, and id 32000 is ac","public_affected_ranges":["< 0.32.0"]}],"null_published_at":"2026-11-05T03:52:54.754Z","source_updated_at":"2026-10-28T21:15:09.000Z","source_disclosed_at":"2026-10-28T21:15:08.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Dated EPSS estimate updated; this is not a new disclosure or proof of exploitation.","null_first_observed_at":"2026-10-31T21:56:08.420Z"},{"id":"NVI-00009196","kind":"vulnerability","title":"vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)","revision":2,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-q97c-p368-75c4","CVE-2026-115916"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-0f8c568821c91650dd45e50a8647bb4f"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-28T21:15:09.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-ef0c78e15322e8eb18b5559c566be136"},{"kind":"CVSS","as_of":"2026-11-02T18:18:39.033Z","value":"6.5 MEDIUM (3.1)","meaning":"security-advisories@github.com severity; no combined Null risk score.","source_id":"sr-0f8c568821c91650dd45e50a8647bb4f"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-ef0c78e15322e8eb18b5559c566be136","url":"https://github.com/advisories/GHSA-q97c-p368-75c4","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-q97c-p368-75c4","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T21:56:08.420Z"},{"id":"sr-0f8c568821c91650dd45e50a8647bb4f","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-115916","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-115916","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:59:00.529Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-ef0c78e15322e8eb18b5559c566be136"],"technology":"vllm","known_exclusions":[],"conditions_summary":"For a model whose vocabulary size is an exact multiple of the embedding padding size (64) — e.g. any Llama-2-family model, vocab 32000 — a single request carrying a pre-tokenized prompt with token id equal to the vocab size (32000) passes vLLM's input-id validation but is one past the last embedding row. At tensor-parallel size 1 the embedding lookup does not mask indices, so the id reaches `F.embedding` out of range and raises a CUDA device-side assert, which poisons the engine's CUDA context and takes the engine down. Any client that can send pre-tokenized input (`/v1/completions` with `prompt` as token ids, `/tokenize`-fed flows, or the `kv_transfer_params[\"prompt_token_ids\"]` reuse path) can trigger it.\n\n## Root cause\n`vllm/v1/engine/input_processor.py` accepts a token id when `id <= max(tokenizer.max_token_id, model_config.get_vocab_size() - 1)`.\n`vllm/tokenizers/hf.py` computes `max_token_id = max(max(vocab.values()), tokenizer.vocab_size)`. For Llama-2 tokenizers `max(vocab.values()) = 31999` but `tokenizer.vocab_size = 32000`, so `max_token_id = 32000` — one greater than the largest valid id.\nThe engine bound therefore becomes `max(32000, 31999) = 32000`, and id 32000 is ac","public_affected_ranges":["< 0.32.0"]}],"null_published_at":"2026-11-03T10:03:31.369Z","source_updated_at":"2026-10-28T21:15:09.000Z","source_disclosed_at":"2026-10-28T21:15:08.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Source facts changed; compare the preserved prior revision.","null_first_observed_at":"2026-10-31T21:56:08.420Z"},{"id":"NVI-00009196","kind":"vulnerability","title":"vLLM: Out-of-vocabulary prompt token id passes input validation and triggers a CUDA device-side assert (engine crash / DoS)","revision":1,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected."],"lifecycle":"active","identifiers":["GHSA-q97c-p368-75c4","CVE-2026-115916"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-ef0c78e15322e8eb18b5559c566be136"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-28T21:15:09.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-ef0c78e15322e8eb18b5559c566be136"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-ef0c78e15322e8eb18b5559c566be136","url":"https://github.com/advisories/GHSA-q97c-p368-75c4","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-q97c-p368-75c4","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T21:56:08.420Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-ef0c78e15322e8eb18b5559c566be136"],"technology":"vllm","known_exclusions":[],"conditions_summary":"For a model whose vocabulary size is an exact multiple of the embedding padding size (64) — e.g. any Llama-2-family model, vocab 32000 — a single request carrying a pre-tokenized prompt with token id equal to the vocab size (32000) passes vLLM's input-id validation but is one past the last embedding row. At tensor-parallel size 1 the embedding lookup does not mask indices, so the id reaches `F.embedding` out of range and raises a CUDA device-side assert, which poisons the engine's CUDA context and takes the engine down. Any client that can send pre-tokenized input (`/v1/completions` with `prompt` as token ids, `/tokenize`-fed flows, or the `kv_transfer_params[\"prompt_token_ids\"]` reuse path) can trigger it.\n\n## Root cause\n`vllm/v1/engine/input_processor.py` accepts a token id when `id <= max(tokenizer.max_token_id, model_config.get_vocab_size() - 1)`.\n`vllm/tokenizers/hf.py` computes `max_token_id = max(max(vocab.values()), tokenizer.vocab_size)`. For Llama-2 tokenizers `max(vocab.values()) = 31999` but `tokenizer.vocab_size = 32000`, so `max_token_id = 32000` — one greater than the largest valid id.\nThe engine bound therefore becomes `max(32000, 31999) = 32000`, and id 32000 is ac","public_affected_ranges":["< 0.32.0"]}],"null_published_at":"2026-10-31T22:13:31.289Z","source_updated_at":"2026-10-28T21:15:09.000Z","source_disclosed_at":"2026-10-28T21:15:08.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"First indexed by Null; source disclosure date is shown separately.","null_first_observed_at":"2026-10-31T21:56:08.420Z"}],"next_cursor":null,"snapshot_revision":"42594"}