{"contentHash":"464e3889c79e9071275c0f3dc0c509058ba96895b2477edf5da0afbef465a6f8","protocol":"bitcoin","authorPubkey":"9b5ea23967a72d09674e4a7f330f8d228b85b01ae547affc25ebdf2866f2ceb0","text":"Why is llamacpp taking 18gb of RAM if all my models and their KV caches are 100% on VRAM?","topicHash":"913c62c2ee31793c7e8517600dbc8f0c337afae133700a128a7d8ffb8a21c93e","topicName":"technology","parentHash":"96a7aacf62c3282f2c5d05a419aa2203b3eab23cab16823a342e71b1fa6b1080","costSunk":"0","biasSupport":0,"biasOppose":0,"biasVouch":0,"biasNull":0,"blockHeight":null,"confirmations":0,"timestamp":"2026-10-24T23:05:22.267Z","tripFp":"649ec9f882d9","viewCount":0,"powDifficulty":0,"relation":0,"ephemeral":true,"nonce":"51e1eedba1a13d14","sig":"3b645030f0f93abab1649aa93eacfcc4aefaeea050b299793dbf484c804148d0d06060fa74b474c14a009d2c75c495fbcd383a48cc3a309d8d73a69ced015354","ephStamp":{"blockHashHex":"00000000000000000000d59e03442bf185da5ea729235bf0568fa382fb68363b","nonceHex":"cd90000000000000"},"replyCount":0,"chainReplyCount":0,"vaporReplyCount":0,"clusterSize":1,"receiptHeight":972985,"expiryHeight":977305,"authorMarks":{"since":"2026-10-24T18:05:09.908Z","via":"sight","inscribed":0,"likes":0,"biases":0},"lang":"en","servedBy":{"host":"6af58b1822e0cabfe57d76540f745d059b438f489a796ec6a705c6540233845e","t":"2026-11-08T20:17:14.359Z","sig":"3be1e6e7fe0722efaadc68d5d2bf31471e75c5870a7e7a12fbbc44723d39822df18fbd2e3267a608a9dc5f71defe6c648183a53067d00dc18e3119d0c511c950","h":"00000000000000000001c6fee9ce4c94f2939fc431bac9a783a15f129c2c58b1"}}