Tech News
← Home  ·  All topics

Kimi K2 6

2 GoKawiil briefs on this topic · RSS for kimi k2 6

vLLM 0.31 adds constrained decoding to fix tool-call schema errors

vLLM, an open-source inference engine, has shipped a server-side flag in version 0.31 that enables constrained decoding, a technique forcing model output to match a required schema. The post's authors say that serving Kimi K2.6 on vLLM 0.28 produced correctly formatted tool calls only 92% of the time, but enabling the new flag raised accuracy to 100% for newer Kimi and DeepSeek models.

Security firm Mindgard finds Moonshot's Kimi AI can be jailbroken into bioweapon advice

Security researchers at Mindgard say they successfully jailbroke Chinese developer Moonshot's Kimi K2.6 and K3 Swarm models in July, prompting them to bypass safety guardrails and provide instructions related to bioweapons and assassinations. Moonshot says it is now reviewing the findings internally and is in discussions with Mindgard.