20M-parameter masked-diffusion chat models trained from scratch: encoder–decoder, ONNX export, 0-shot benchmarks.
quark28
quark28
AI & ML interests
None yet
Recent Activity
updated a collection 2 days ago
mini-llm updated a collection 2 days ago
mini-llm updated a model 2 days ago
quark28/mini-llm-thinking-sftOrganizations
None yet