Upload README.md with huggingface_hub

README.md

@@ -1,18 +1,17 @@
 ---
-license: other
+license: apache-2.0
 tags:
 - merge
-- mergekit
 - not-an-original-model
 ---
 
 # Mini-Merge-35B-A3B
 
-**This is a model *merge*, not an original training run.** It was produced by
-element-wise merging two existing open-weight checkpoints with
-[mergekit](https://github.com/arcee-ai/mergekit) -- zero pretraining and zero
-post-training of our own. We publish the exact recipe (`merge.yaml`, included in
-this repo) so the "how" is not a mystery.
+**This is a model *merge*, not an original training run.** It was produced by a plain
+element-wise weighted average of two existing open-weight checkpoints -- a ~40-line
+safetensors script, no mergekit (its catalog doesn't know this architecture). Zero
+pretraining, zero post-training of our own. The exact recipe ships as `merge.yaml`
+alongside the weights, so the "how" is not a mystery.
 
 ## Ingredients
 - `nex-agi/Nex-N2-mini`
@@ -45,5 +44,6 @@ modal run modal_app.py::sweep      # or ::rematch for the 397B build
 ```
 
 ## License
-Inherits the licenses of **both** parents (`nex-agi/Nex-N2-mini` and `Qwen/Qwen3.6-35B-A3B`). Check and comply with
-both before use. This card must travel with the weights.
+Both parents (`nex-agi/Nex-N2-mini` and `Qwen/Qwen3.6-35B-A3B`) are Apache-2.0, so this merge is released under
+**Apache-2.0**. The merge used a small original script, not mergekit, so no LGPL applies.
+This card must travel with the weights.