{"type":"post","id":"omlx-0-7-3-speeds-up-deepseek-v4-fd0577","slug":"omlx-0-7-3-speeds-up-deepseek-v4-fd0577","title":"oMLX 0.7.3 speeds up DeepSeek V4.1 prefill on M5 and cuts memory for many concurrent long-context requests on Apple silicon","description":"oMLX 0.7.3 reports faster DeepSeek V4.1 Flash prefill and lower memory use for concurrent long-context inference on Apple silicon, alongside updates to its Ask AI interface.","retrieval_nugget":"oMLX 0.7.3 reports faster DeepSeek V4.1 Flash prefill and lower memory use for concurrent long-context inference on Apple silicon, alongside updates to its Ask AI interface. For builders running local models on Macs, the release offers concrete performance targets to test against their own workloads, not a general guarantee of speed or capacity.","published_at":"2026-10-31","updated_at":"2026-10-31","record_date":"2026-10-31","date_kind":"published_at","topics":["memory"],"entities":["oMLX","DeepSeek V4.1 Flash","Qwen3.8-Flash-Next","Apple"],"related_patterns":[],"novelty_label":"new_fact","what_changed":"oMLX 0.7.3 adds faster DeepSeek V4.1 prefill and batched decode, plus lower memory use for concurrent long-context requests on Apple silicon.","prior_refs":["memory:9b622f44-c310-8f83-a667-b82252016f0c","memory:ac6d7fe1-50eb-8ba7-a6a7-40f26ac82c42"],"source_url":"https://github.com/jundot/omlx/releases/tag/v0.7.3","source_urls":["https://github.com/jundot/omlx/releases/tag/v0.7.3"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:omlx-0-7-3-speeds-up-deepseek-v4-fd0577","status":"published","visuals":[],"editor":{"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"routes":{"html":"https://newruntime.com/posts/omlx-0-7-3-speeds-up-deepseek-v4-fd0577/","markdown":"https://newruntime.com/posts/omlx-0-7-3-speeds-up-deepseek-v4-fd0577.md","json":"https://newruntime.com/posts/omlx-0-7-3-speeds-up-deepseek-v4-fd0577.json"},"next_reads":[{"type":"topic","path":"/topics/memory/","url":"https://newruntime.com/topics/memory/","title":"Memory","reason":"Explore the memory topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/memory.md","json_url":"https://newruntime.com/topics/memory.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.json"},{"type":"collection","path":"/source-ledger.json","url":"https://newruntime.com/source-ledger.json","title":"Source Ledger","reason":"Inspect the public source evidence graph.","media_type":"application/json"}]}