Professional model card

README.md

@@ -1,24 +1,24 @@
 ---
 license: llama3.2
 base_model: meta-llama/Llama-3.2-1B-Instruct
+library_name: basert
+pipeline_tag: text-generation
 tags:
   - basert
-  - base-format
   - apple-silicon
   - quantized
 ---
 
-# Llama-3.2-1B-Instruct (BaseRT `.base`)
+# Llama-3.2-1B-Instruct
 
-[BaseRT](https://github.com/basecompute/baseRT) `.base` conversions of
-[`meta-llama/Llama-3.2-1B-Instruct`](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) for the BaseRT Apple-Silicon (Metal)
-inference engine. Converted **directly from the bf16 source** (no
-quant-from-quant) to avoid compounding quantization error.
+[BaseRT](https://github.com/basecompute/baseRT) `.base` builds of [`meta-llama/Llama-3.2-1B-Instruct`](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) for fast local inference on Apple Silicon (Metal).
 
-| File | Quant | Size |
-|------|-------|------|
-| `Llama-3.2-1B-Instruct-Q4.base` | `base_q4` | 675M |
-| `Llama-3.2-1B-Instruct-Q8.base` | `base_q8` | 1.2G |
+## Files
+
+| File | Precision | Size |
+|------|-----------|------|
+| `Llama-3.2-1B-Instruct-Q4.base` | 4-bit | 702 MB |
+| `Llama-3.2-1B-Instruct-Q8.base` | 8-bit | 1.3 GB |
 
 ## Usage
 
@@ -28,4 +28,4 @@ basert pull basecompute/Llama-3.2-1B-Instruct
 basert chat basecompute/Llama-3.2-1B-Instruct
 ```
 
-Context: 128k · License inherited from the base model.
+Released under the llama3.2 license, inherited from the base model.