model card

README.md

@@ -12,7 +12,7 @@ tags:
   - vext
 ---
 
-> **Weights are hosted on Cloudflare R2** (see the Download section below), not inside this Hugging Face repo. Download the files to a local folder, then load from that local path. The base weights (`juwel-sapphire`) are on R2 the same way.
+> **The adapter weights are hosted in this Hugging Face repo** — load them directly from the Hub. The **base** model (`VextLabsinc/juwel-sapphire`) is currently on Cloudflare R2 (see Download) pending migration to HF: download the base, load it from the local path, then apply this adapter.
 
 # GEM Ruby · code specialist
 
@@ -101,11 +101,7 @@ See also `USE_POLICY.md`.
 
 <!-- legal_tier: T1 -->
 
-## Weights — download (Cloudflare R2, public, BF16)
+## Weights — download
 
-Weights are hosted on Cloudflare R2 (BF16, no quantization):
-
-- Adapter: https://pub-a6ae0476e46849f98f1746a61dc4c106.r2.dev/gem-ruby/adapter_model.safetensors
-- Config:  https://pub-a6ae0476e46849f98f1746a61dc4c106.r2.dev/gem-ruby/adapter_config.json
-
-Load on the base `VextLabsinc/juwel-sapphire`. (r2.dev is the launch host; a custom domain can be attached later without changing the repo.)
+- **Adapter (this model):** hosted in this HF repo — `adapter_model.safetensors` + `adapter_config.json`. Loads with `PeftModel.from_pretrained(base, "VextLabsinc/gem-ruby")`. (R2 mirror: https://pub-a6ae0476e46849f98f1746a61dc4c106.r2.dev/gem-ruby/)
+- **Base (`juwel-sapphire`, 144L, BF16):** currently on Cloudflare R2 (migrating to HF). Download all shards from https://pub-a6ae0476e46849f98f1746a61dc4c106.r2.dev/juwel-sapphire/ and load from the local path, then apply this adapter.