Release 1.0.21

Author: rwightman

README.md

@@ -12,16 +12,18 @@
 
 ## What's New
 
-## Oct 16, 2025
+## Oct 16-20, 2025
 * Add an impl of the Muon optimizer (based on https://github.com/KellerJordan/Muon) with customizations
   * extra flexibility and improved handling for conv weights and fallbacks for weight shapes not suited for orthogonalization
   * small speedup for NS iterations by reducing allocs and using fused (b)add(b)mm ops
   * by default uses AdamW (or NAdamW if `nesterov=True`) updates if muon not suitable for parameter shape (or excluded via param group flag)
   * like torch impl, select from several LR scale adjustment fns via `adjust_lr_fn`
   * select from several NS coefficient presets or specify your own via `ns_coefficients`
-* First 2 steps of 'meta' device model initalization supported
+* First 2 steps of 'meta' device model initialization supported
   * Fix several ops that were breaking creation under 'meta' device context
   * Add device & dtype factory kwarg support to all models and modules (anything inherting from nn.Module) in `timm`
+* License fields added to pretrained cfgs in code
+* Release 1.0.21
 
 ## Sept 21, 2025
 * Remap DINOv3 ViT weight tags from `lvd_1689m` -> `lvd1689m` to match (same for `sat_493m` -> `sat493m`)

timm/version.py

@@ -1 +1 @@
-__version__ = '1.0.21.dev0'
+__version__ = '1.0.21'