{"type":"post","id":"unsloth-v0-1-914-beta-3-9x-faster-768648","slug":"unsloth-v0-1-914-beta-3-9x-faster-768648","title":"Unsloth v0.1.914-beta: 3.9x faster Transformers-backend generation, CPU training and more parallelism options","description":"Unsloth’s v0.1.914-beta release adds CPU text fine-tuning, more multi-GPU training configurations and faster generation for supported Llama and Qwen3 models.","retrieval_nugget":"Unsloth’s v0.1.914-beta release adds CPU text fine-tuning, more multi-GPU training configurations and faster generation for supported Llama and Qwen3 models. In the project’s A100 test, 4-bit Qwen3-8B reached 99.1 tokens per second, up from 25.7; the result is specific to that setup and backend.","published_at":"2026-11-04","updated_at":"2026-11-04","record_date":"2026-11-04","date_kind":"published_at","topics":["agent-economics"],"entities":["Unsloth","Unsloth v0.1.914-beta","Qwen3-8B","Llama"],"related_patterns":["agent-economics-moves-to-completed-work"],"novelty_label":"new_fact","what_changed":"Unsloth v0.1.914-beta adds 3.9× faster Transformers-backend generation, CPU training, and more parallelism options; the archive covers a separate Transformers training release.","prior_refs":["https://newruntime.com/posts/transformers-5-22-adds-lora-peft-training-on-d1c416/"],"source_url":"https://github.com/unslothai/unsloth/releases","source_urls":["https://github.com/unslothai/unsloth/releases"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:unsloth-v0-1-914-beta-3-9x-faster-768648","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/unsloth-v0-1-914-beta-3-9x-faster-768648/","markdown":"https://newruntime.com/posts/unsloth-v0-1-914-beta-3-9x-faster-768648.md","json":"https://newruntime.com/posts/unsloth-v0-1-914-beta-3-9x-faster-768648.json"},"markdown_url":"https://newruntime.com/posts/unsloth-v0-1-914-beta-3-9x-faster-768648.md","next_reads":[{"type":"topic","path":"/topics/agent-economics/","url":"https://newruntime.com/topics/agent-economics/","title":"Agent Economics","reason":"Explore the agent-economics topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/agent-economics.md","json_url":"https://newruntime.com/topics/agent-economics.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":"/sources/","url":"https://newruntime.com/sources/","title":"Sources","reason":"See the original sources this archive cites.","media_type":"text/html"}]}