Inference framework Archon promises to make LLMs quicker, without additional costs https://t.co/2R4jTYw3hg https://t.co/1CKWEVi7SH on https://twitter.com/AcerboLivio/status/1841240935436345595