dream¶
Model type |
|
Class |
|
Task |
|
Source |
|
Description¶
Dream dense masked-diffusion transformer.
Dream uses Qwen2-style GQA with biased Q/K/V projections, bias-free output and SwiGLU projections, RMSNorm, RoPE, and full bidirectional attention.
Usage¶
mobius build --model <MODEL_ID> output_dir/
from mobius import build
model = build("<MODEL_ID>")