Adina Yakefu
AdinaY
AI & ML interests
None yet
Recent Activity
posted
an
update
about 2 hours ago
GLM-Image from Z.ai is out π₯
It was fully trained on Ascend Atlas 800T A2 with MindSpore, probably the first SOTA multimodal model fully trained on domestic chips π
https://huggingface.co/zai-org/GLM-Image
β¨ Hybrid Architecture: combined autoregressive + diffusion design delivers strong semantic alignment with high-fidelity details
β¨ Strong performance in long, dense, and multilingual text rendering
β¨ MIT licensed (VQ tokenizer & ViT weights under Apache 2.0)
β¨ Now live on Hugging Face inference provider π€
updated
a collection
about 3 hours ago
2026 JanuaryβοΈ - China Open Source Highlights
liked
a model
about 3 hours ago
zai-org/GLM-Image
Organizations
