{"type":"post","id":"kapitsa-0-2-22-runs-flash-next-from-9160f3","slug":"kapitsa-0-2-22-runs-flash-next-from-9160f3","title":"Kapitsa 0.2.22 runs Flash-Next from EXL3 packs at 78.6 tok/s on an M5 Max and loads 8-bit expert packs","description":"Kapitsa 0.2.22 expands an Apple Silicon-oriented local inference runtime with EXL3 support for a Flash-Next pack, additional expert-weight import paths, and maintainer-reported benchmarks.","retrieval_nugget":"Kapitsa 0.2.22 expands an Apple Silicon-oriented local inference runtime with EXL3 support for a Flash-Next pack, additional expert-weight import paths, and maintainer-reported benchmarks. For teams testing local mixture-of-experts inference, the release makes format choice, download size, device memory, and runtime compatibility more concrete operational variables.","published_at":"2026-10-28","updated_at":"2026-10-28","record_date":"2026-10-28","date_kind":"published_at","topics":["github"],"entities":["Kapitsa","Flash-Next","EXL3","Sushi","MLX","GGUF"],"related_patterns":["harness-architecture-outlives-model-choice"],"novelty_label":"new_fact","what_changed":"Kapitsa 0.2.22 adds EXL3 support, reports Flash-Next at 78.6 tok/s on an M5 Max, and loads 8-bit expert packs—new capabilities and benchmarks beyond v0.2.15.","prior_refs":["memory:b1983391-b1a5-8fb5-ac96-81bf579d72ec","memory:63f020bf-f96b-8941-a401-750ae224ada3"],"source_url":"https://github.com/nace-ai/kapitsa/releases/tag/v0.2.22","source_urls":["https://github.com/nace-ai/kapitsa/releases/tag/v0.2.22"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:kapitsa-0-2-22-runs-flash-next-from-9160f3","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/kapitsa-0-2-22-runs-flash-next-from-9160f3/","markdown":"https://newruntime.com/posts/kapitsa-0-2-22-runs-flash-next-from-9160f3.md","json":"https://newruntime.com/posts/kapitsa-0-2-22-runs-flash-next-from-9160f3.json"},"next_reads":[{"type":"topic","path":"/topics/github/","url":"https://newruntime.com/topics/github/","title":"Github","reason":"Explore the github topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/github.md","json_url":"https://newruntime.com/topics/github.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"}]}