AITOPAI热报
Products

Hugging Face 上线训练直播：实时围观 LFM2.5、LFM3 与 Twil-9B 的 RL 训练

官方账号Hugging Face@huggingface
Post
View the post
Why we picked it

Hugging Face 把模型训练过程直接直播出来了，正在用 RL 微调 LiquidAI 的 LFM2.5、LFM3 和 Twil-9B，点进去能实时看 loss 曲线，做微调的可以去围观。

Hugging Face 推出 Harness adapters 训练直播，目前正用 RL 对 LiquidAI 的 LFM2.5、LFM3 和 thewebAI 的 Twil-9B 做 LoRA 适配器训练。目标是让这几个模型在 agentic 任务上更高效、更准确。训练过程可通过 hf.space 页面实时观看，配套博客已发布。

Original · Hugging Face

Stream live training runs on Hugging Face! younes @yb2698 Harness adapters training is now live! We are training LoRA adapters on LFM2.5 / LFM3 from @liquidai and the recent Twil-9B model from @thewebAI using RL to make them more efficient and accurate on agentic tasks! You can follow the training going live here: …ness-hub-harness-adapters-rl.hf.space huggingface.co/blog/harness-h… Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 27 🔄 7 ❤️ 57 👀 7252 📊 26 ⚡

#Hugging Face#LFM2.5#Twil-9B#LoRA#开源模型

Keep reading

Recent stories with the same tag or section
How-toBuildersSelected04:42
Hugging Face publishes 30k-word MoE training handbook

The handbook distills hard-won engineering lessons on MoE efficiency that labs usually keep internal, making sparse-model training accessible to smaller teams. Engineers planning distributed pretraining or fine-tuning runs can use its profiling guidance and interactive figures to diagnose bottlenecks without burning hundreds of thousands of GPU-hours themselves. It complements Hugging Face's earlier open training playbooks rather than vendor whitepapers.

#Hugging Face#MoE#fine-tuning#open-source models
Thomas Wolf@Thom_WolfSource
How-toBuildersSelected04:41
Hugging Face Publishes Comprehensive Guide to Training MoE Models

MoE architectures power many frontier open-weight models, yet training guidance at scale is scarce, so a step-by-step playbook with visual explanations fills a real gap for engineers. Practitioners planning pretraining or fine-tuning runs can use it alongside the team's earlier Ultra Scale Playbook. No regional angle applies; the closest comparison is existing distributed-training documentation from PyTorch and DeepSpeed.

#Hugging Face#Mixture of Experts#fine-tuning#open-source models
Thomas Wolf@Thom_WolfSource