CVE Intelligence
Low Info disclosure

CVE-2026-105752 — vLLM: Harmony tool continuations drop `cache_salt` — restoring a cross-tenant prefix-cache membership oracle

vLLM is an inference and serving engine for large language models. Prior to 0.30.0, Harmony tool continuations submitted through "POST /v1/responses" requests rebuild the next-turn engine input without preserving the cache_salt value, placing the continuation prefix in the global unsalted cache namespace even when the caller enabled salting. On deployments with prefix caching enabled, which is the default, an authenticated tenant who can reconstruct a victim's low-entropy post-tool history can submit the same continuation and use the cached_tokens_per_turn count to determine whether the prefix was previously processed, defeating the intended tenant isolation of salted prefix caching. This issue is fixed in version 0.30.0.

Published Oct 5, 2026 · Updated Oct 13, 2026 · Sources: NIST National Vulnerability Database, CVE Program (cve.org) e CISA ADP · CNA: GitHub_M

Who exploits it and how urgent it is

A 0–100 score combining confirmed exploitation, ransomware and APT use, public exploits, EPSS and severity.

No exploitation signal in the sources consulted so far.
How the score was calculated
  • EPSS 0.2% (30-day exploitation probability)+2

The score orders the patch queue; it does not replace your asset context (internet exposure, criticality, compensating controls).

Triage

Is it exploited, how likely is exploitation, what does it touch, and how severe do the sources call it.

Exploitation
Not confirmed
Not in CISA KEV — an absence of record, not proof it is unexploited.
EPSS FIRST
0.22%
ahead of 11% of scored CVEs · Oct 14, 2026
Affects
Vllm
Vllm and 1 more
CVSS base
3.1
Low · 2 source(s)

CISA SSVC assessment CISA

Three decision points CISA publishes for the CVEs it assesses (SSVC 2.0.3). A stakeholder decision, not a severity score.

Exploitation
None
none · proof-of-concept · active
Automatable
No
can an attacker script the kill-chain steps
Technical impact
Partial
partial · total control of the vulnerable component

Remediation

Fixes and mitigations documented by the sources — linked to the original advisory.

  • Documented patch — github.com (NIST National Vulnerability Database)
  • Documented patch — github.com (NIST National Vulnerability Database)
  • Documented patch — github.com (NIST National Vulnerability Database)
  • Documented mitigation — github.com (NIST National Vulnerability Database)

First 24 hours

Ordered from the record’s own fields — exposure first, because you cannot patch what you have not found. A rule-generated starting point.

  1. Identify exposed assets running the affected vendor/product/version combinations (Vllm Vllm, Vllm-project Vllm-project/vllm).
  2. Apply the fixed version or the patch documented under Remediation.
  3. Search available logs for exploit probes matching the vulnerability class (info disclosure), errors and authentication anomalies.

Affected scope

Vendor, product and versions as the sources word them.

VendorProductVersionsStatusSource
Vllm Vllm < 0.30.0 Vulnerable NIST National Vulnerability Database
Vllm-project Vllm-project/vllm version < 0.30.0 Vulnerable CVE Program (cve.org) e CISA ADP

Attack characteristics

The CVSS vector, decoded. It describes the attack, not your exposure to it.

Attack vector
Network
Attack complexity
High
Privileges required
Low
User interaction
None
Scope
Unchanged
Confidentiality
None
Integrity
Low
Availability
None

Every base score collected

Sources score independently and may disagree; each row says who scored it and under which version.

ScoreVersionSeverityExpl.ImpactScored by
3.1 CVSS 3.1 Low 1.6 1.4 security-advisories@github.com via NIST National Vulnerability Database
3.1 CVSS 3.1 Low — — GitHub_M via CVE Program (cve.org) e CISA ADP
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N (score shown at the top: security-advisories@github.com)

Weakness MITRE

Weakness class according to MITRE’s CWE catalog.

CWE-200 · Exposure of Sensitive Information

Other CVEs with this weakness →

EPSS over time FIRST

Oct 13, 2026 → Oct 14, 2026 · 2 reading(s) · max 0.22%

Detection

Read off the CVSS vector and the weakness class. Starting points, not tested rules.

  • Prioritize edge telemetry for the network-reachable component (proxy, WAF, IDS/IPS).
  • Look for anomalous data read/export volumes and access to sensitive records.
web/proxy/WAF logsnetwork IDS/IPSapplication logsauthentication logs

Timeline

What happened to this CVE, newest first.

  • 2026
  • Affected products added · vllm-project/vllm
    Oct 13, 2026 · CVE Program (cve.org) e CISA ADP
  • SSVC assessment updated
    Oct 13, 2026 · CVE Program (cve.org) e CISA ADP
  • Record published
    Oct 13, 2026
  • Fix documented
    Oct 13, 2026 · NIST National Vulnerability Database
  • CVSS changed · 3.1
    Oct 13, 2026 · NIST National Vulnerability Database
  • Source status changed · → Analyzed
    Oct 13, 2026 · NIST National Vulnerability Database
  • Record created in this database
    Oct 13, 2026 · NIST National Vulnerability Database
  • CVE published
    Oct 5, 2026 · GitHub_M

Source activity

Repeated readings are counted rather than listed.

  • 1× Official description updated · Oct 13, 2026 – Oct 13, 2026

References

5 on the record

Elsewhere on this site

Vllm
every CVE for this vendor
Info disclosure
same class
CWE-200
other pages naming this weakness

Related CVEs (same products)

  • CVE-2026-109311 vLLM before 0.29.0 Heap Overflow DoS via IAMF Audio Transcription Upload 3d
  • CVE-2026-105922 vllm-project vLLM Penalty utils.py get_token_bin_counts_and_mask denial of service 9d
  • CVE-2026-105775 vllm-project vLLM Completions Request mamba_mixer2.py conv_ssm_forward out-of-bounds 9d
  • CVE-2026-105756 vLLM: Loose `cache_salt` validation lets a single request kill EngineCore on LMCache-MP deployments — uncaught downstream `ValueError` denial of service 9d
  • CVE-2026-105758 vLLM: Qwen2-VL / Qwen3-VL video samplers bound on request-controlled max_frames, which the num_frames ceiling does not reach 9d
  • CVE-2026-105755 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` 9d

Not in any source we poll

Listed rather than left blank: an empty field and an unmeasured one look identical on screen, and only one is a reason to look elsewhere.

  • No confirmed IOCs, IP addresses, domains, file hashes, or malware artifacts supplied.
  • No organization-specific asset inventory, compensating-control status, or patch deployment evidence supplied.
  • No exploit packet captures, log samples, or incident case IDs supplied.

Answered from this record

Text assembled only from the fields above — no additional claims.

What should defenders know first?

CVE-2026-105752 is "vLLM: Harmony tool continuations drop `cache_salt` — restoring a cross-tenant prefix-cache membership oracle", rated Low. It affects Vllm (Vllm); versions are in the scope table. It is not in CISA KEV, which does not rule out exploitation. CVSS 3.1 describes technical severity. EPSS 0.22% estimates the probability of exploitation in 30 days; neither replaces your asset context. First verify whether exposed or business-critical assets run the affected versions, apply vendor fixes or mitigations, and preserve logs for review.

Comments

Moderated technical discussion. Read the community guidelines.

No comments yet.

Sign in or create an account to comment and suggest corrections.