view post Post 3060 We collaborated with NVIDIA to teach you how we made LLM training ~25% faster! πLearn how 3 optimizations help your home GPU train models faster:1. Packed-sequence metadata caching2. Double-buffered checkpoint reloads3. Faster MoE routingGuide: https://unsloth.ai/blog/nvidia-collabGitHub: https://github.com/unslothai/unsloth See translation π₯ 12 12 π 3 3 π€ 2 2 π 1 1 + Reply
drbaph/HiDream-O1-Image-Dev-FP8 Image-Text-to-Image β’ 9B β’ Updated about 21 hours ago β’ 12 β’ 2
drbaph/HiDream-O1-Image-Dev-FP8 Image-Text-to-Image β’ 9B β’ Updated about 21 hours ago β’ 12 β’ 2
drbaph/HiDream-O1-Image-Dev-BF16 Image-Text-to-Image β’ 9B β’ Updated about 21 hours ago β’ 14 β’ 1
drbaph/HiDream-O1-Image-Dev-BF16 Image-Text-to-Image β’ 9B β’ Updated about 21 hours ago β’ 14 β’ 1