Hugging Face's logo
Moon's picture

Moon

shakhmoon
·

AI & ML interests

None yet

Recent Activity

replied to danielhanchen's post about 20 hours ago
We collaborated with Hugging Face to enable you to train MoE models 12× faster with 35% less VRAM via our new Triton kernels (no accuracy loss). 🤗 Train gpt-oss locally on 12.8GB VRAM with our free notebooks: https://unsloth.ai/docs/new/faster-moe
reacted to danielhanchen's post with 🔥 about 20 hours ago
We collaborated with Hugging Face to enable you to train MoE models 12× faster with 35% less VRAM via our new Triton kernels (no accuracy loss). 🤗 Train gpt-oss locally on 12.8GB VRAM with our free notebooks: https://unsloth.ai/docs/new/faster-moe
View all activity

Organizations

None yet
shakhmoon (Moon)
Hugging Face's logo
Moon's picture

Moon

shakhmoon
·

AI & ML interests

None yet

Recent Activity

replied to danielhanchen's post about 20 hours ago
We collaborated with Hugging Face to enable you to train MoE models 12× faster with 35% less VRAM via our new Triton kernels (no accuracy loss). 🤗 Train gpt-oss locally on 12.8GB VRAM with our free notebooks: https://unsloth.ai/docs/new/faster-moe
reacted to danielhanchen's post with 🔥 about 20 hours ago
We collaborated with Hugging Face to enable you to train MoE models 12× faster with 35% less VRAM via our new Triton kernels (no accuracy loss). 🤗 Train gpt-oss locally on 12.8GB VRAM with our free notebooks: https://unsloth.ai/docs/new/faster-moe
View all activity

Organizations

None yet