NinjaMoves@feddit.nu to LocalLLaMA@sh.itjust.worksEnglish · 15 days agoMistral small 3.1 releasedmistral.aiexternal-linkmessage-square6fedilinkarrow-up130arrow-down11
arrow-up129arrow-down1external-linkMistral small 3.1 releasedmistral.aiNinjaMoves@feddit.nu to LocalLLaMA@sh.itjust.worksEnglish · 15 days agomessage-square6fedilink
minus-squarebrucethemoose@lemmy.worldlinkfedilinkEnglisharrow-up8·15 days agoAnyone tested it at high context yet? I find all Mistral models peter out after like 16K-24K tokes no matter what they advertise the context length as.
Anyone tested it at high context yet? I find all Mistral models peter out after like 16K-24K tokes no matter what they advertise the context length as.