Unverified50% confidenceReleaseExact time
The core change in v0.31.1rc0 is the addition of the capability to expose cached prompt tokens by cache tier in the metrics system, corresponding to PR #56318.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/6/2026
First Seen
Valid until: 1/4/2027
Sources
Related Entities
Related Claims
UnverifiedVersion v2.1.274 adds a memory usage critical alert that displays a warning when memory reaches a dangerous level and provides steps for freeing memory or safely restarting.69% similarUnverifiedv2.1.286 fixes multiple key redaction issues, including no longer exposing credential values when Bearer or Basic appears before the key name in error messages, and the issue of percent-encoded Bearer tokens being only partially masked.68% similarUnverifiedThe vulnerability fixed in version 0.65.5 is: when the requested table name contains a trailing newline character at the end, it is possible to bypass table-level permission checks and expose private row data that should be protected.62% similarUnverifiedAs of vLLM version 0.11, Automatic Prefix Caching uses SHA-256 by default to reduce collision and information leakage risks, but this cannot replace tenant isolation.62% similarUnverifiedDependabot's Security Updates feature is specifically designed to create urgent PRs for known CVE-numbered security issues.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/982066API
curl https://kongchang.com/api/v1/knowledge/claims/982066MCP
get_claim(id=982066)