vLLM has released version 0.31.0, introducing a new security measure. This update rejects per-request arguments for `mm_processor_kwargs` and `media_io_kwargs` by default. To enable these arguments, users must explicitly set the `--trust-request-mm-kwargs` flag, enhancing the security of the vLLM framework. AI
IMPACT Enhances security for users of the vLLM framework, requiring explicit opt-in for certain request parameters.
RANK_REASON This is a software update for a specific tool, not a frontier release or significant industry event.
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →