Hugging Face releases LFM2.5 series DSpark draft models, boosting inference speed by up to 3.18x
Original title:Up to 3.2x Faster Inference with LFM2.5-DSpark
AI summary
Hugging Face released DSpark draft model checkpoints for three LFM2.5 series models, using speculative decoding to boost GPU throughput by up to 3.18x and on-device performance by up to 2.87x without changing output quality. The draft models have about 300M parameters, and LFM2.5-2.6B reduces function-calling latency by an average of 57%. They are open-sourced with support for llama.cpp and SGLang.
Key Takeaways
- Hugging Face released DSpark draft model checkpoints for three LFM2.
- 5 series models,using speculative decoding to boost GPU throughput by up to 3.
- 18x and on-device performance by up to 2.
- 87x without changing output quality.
- The draft models have about 300M parameters,and LFM2.
Quick Facts
- Published
- Updated
- Source
- Hugging Face 社区博客(混合发现)
- Category
- Model Releases
- AI score
- 61
约300M参数的draft模型换最高3.18倍吞吐,贪心输出不变,LFM2.5-2.6B在M4 Max达139 tok/s,本地agent门槛被拉低。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Hugging Face releases LFM2.5 series DSpark draft models, boosting inference speed by up to 3.18x. August 21, 2026. https://new.nat6.net/en/news/cmt1rv5n8066iroovaxgoej1b/