Rezultati pretraživanja za 'Proizvođač: vllm-project'
| ID | CVSS | Sažetak | Zadnje ažurirano | Objavljeno |
|---|---|---|---|---|
| CVE-2026-94626 | 7.5 | vLLM through 0.29.0 fails to validate the tp_size parameter in kv_transfer_params on OpenAI-compatible completion endpoints, allowing attackers to allocate unbounded memory. Attackers can supply arbitrary tp_size values in prefill/decode disaggregate | 2026-09-29 12:46:53 | 2026-09-21 22:17:01 |
| CVE-2026-94627 | 7.5 | vLLM Mooncake connector through 0.29.0 fails to properly manage GPU KV cache block ownership when concurrent child requests share a single transfer ID in prefill/decode disaggregated deployments. Attackers can trigger GPU memory exhaustion by submitt | 2026-09-29 12:46:29 | 2026-09-21 22:17:01 |
| CVE-2026-73556 | 5.3 | vLLM is an inference and serving engine for large language models. Prior to 0.26.0, the structured_outputs.regex parameter in vllm/v1/structured_output/backend_lm_format_enforcer.py is passed to lmformatenforcer.RegexParser without compile_regex_with | 2026-09-28 18:42:04 | 2026-08-13 15:20:17 |
| CVE-2026-73558 | 5.3 | vLLM is an inference and serving engine for large language models. Prior to 0.27.0, an integer overflow in blockIdx.x * 2 * d in activation_kernels.cu can cause act_and_mul_kernel to consume another batched user's input, allowing a request processed | 2026-09-28 18:41:49 | 2026-08-13 15:20:18 |
| CVE-2026-73559 | 6.5 | vLLM is an inference and serving engine for large language models. From 0.19.0 until 0.26.0, the /v1/completions CompletionRequest.prompt field in vllm/entrypoints/openai/completion/protocol.py accepts an unbounded list[str] or list[list[int]], promp | 2026-09-28 18:41:23 | 2026-08-13 16:19:05 |
| CVE-2026-73560 | 6.5 | 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, reque | 2026-09-28 18:41:05 | 2026-08-17 21:16:48 |
| CVE-2026-93989 | 3.1 | vLLM through 0.29.0 fails to properly validate bad_words token indices against the model's generation output width in SamplingParams.update_from_tokenizer(). Attackers can supply out-of-bounds token indices that corrupt logits memory of concurrent re | 2026-09-28 18:38:31 | 2026-09-19 23:17:10 |
| CVE-2026-93436 | 7.5 | vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory without bound unti | 2026-09-28 18:37:42 | 2026-09-17 23:18:54 |
| CVE-2026-93592 | 7.5 | vLLM versions before 0.28.0 fail to validate the lower bound of token IDs in the /v1/embeddings and /pooling endpoints, allowing unauthenticated attackers to crash the engine by submitting negative token IDs. A single request with a negative token ID | 2026-09-28 18:37:06 | 2026-09-18 14:19:10 |
| CVE-2026-93840 | 3.7 | vLLM before 0.29.0 validates allowed_token_ids against tokenizer length instead of model output logits width in SamplingParams._validate_allowed_token_ids(). Attackers can supply token IDs above the output vocabulary that pass validation, causing Log | 2026-09-28 18:36:00 | 2026-09-18 20:17:34 |
| ID | CVSS | Sažetak | Zadnje ažurirano | Objavljeno |
|---|---|---|---|---|
| CVE-2018-4451 | 9.3 | This issue is fixed in macOS Mojave 10.14. A memory corruption issue was addressed with improved input validation. | 2020-10-30 02:01:00 | 2020-10-27 20:15:00 |
| CVE-2019-8712 | 10.0 | A memory corruption issue was addressed with improved memory handling. This issue is fixed in watchOS 6, iOS 13, tvOS 13. An application may be able to execute arbitrary code with system privileges. | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2019-8709 | 9.3 | A memory corruption issue was addressed with improved state management. This issue is fixed in macOS Catalina 10.15, tvOS 13, macOS Catalina 10.15.1, Security Update 2019-001, and Security Update 2019-006, watchOS 6, iOS 13. An application may be abl | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2019-8824 | 9.3 | A memory corruption issue was addressed with improved state management. This issue is fixed in macOS Catalina 10.15.1, Security Update 2019-001, and Security Update 2019-006. An application may be able to execute arbitrary code with kernel privileges | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2018-4452 | 9.3 | A memory consumption issue was addressed with improved memory handling. This issue is fixed in macOS Mojave 10.14.3, Security Update 2019-001 High Sierra, Security Update 2019-001 Sierra, macOS Mojave 10.14.2, Security Update 2018-003 High Sierra, Se | 2020-10-30 02:00:00 | 2020-10-27 20:15:00 |
| CVE-2019-8716 | 10.0 | A memory corruption issue was addressed with improved memory handling. This issue is fixed in macOS Catalina 10.15.1, Security Update 2019-001, and Security Update 2019-006. An application may be able to execute arbitrary code with system privileges. | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2019-8836 | 9.3 | A memory corruption issue was addressed with improved memory handling. This issue is fixed in watchOS 6.1.2, iOS 13.3.1 and iPadOS 13.3.1, tvOS 13.3.1. An application may be able to execute arbitrary code with kernel privileges. | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2019-8740 | 9.3 | A memory corruption vulnerability was addressed with improved locking. This issue is fixed in iOS 13.1 and iPadOS 13.1, watchOS 6, tvOS 13. An application may be able to execute arbitrary code with kernel privileges. | 2021-07-21 11:39:00 | 2020-10-27 20:15:00 |
| CVE-2020-27976 | 10.0 | osCommerce Phoenix CE before 1.0.5.4 allows OS command injection remotely. Within admin/mail.php, a from POST parameter can be passed to the application. This affects the PHP mail function, and the sendmail -f option. | 2020-10-29 23:35:00 | 2020-10-28 15:15:00 |
| CVE-2019-8539 | 9.3 | A memory initialization issue was addressed with improved memory handling. This issue is fixed in macOS Mojave 10.14.6, Security Update 2019-004 High Sierra, Security Update 2019-004 Sierra. A malicious application may be able to execute arbitrary co | 2020-10-30 15:13:00 | 2020-10-27 20:15:00 |
Stranica
/7
Zapisa po stranici: