README

README.md

@@ -6,6 +6,11 @@ tags:
 - trl
 - sft
 - mlx
+- apple-silicon
+- on-device
+- tiny-llm
+- smollm
+- quantized
 datasets:
 - Magpie-Align/Magpie-Pro-300K-Filtered
 - bigcode/self-oss-instruct-sc2-exec-filter-50k
@@ -16,3 +21,32 @@ language:
 - en
 pipeline_tag: text-generation
 ---
+
+# SmolLM-360M-Instruct (MLX 3-bit)
+
+An **3-bit MLX quantized** build of `HuggingFaceTB/SmolLM-360M-Instruct` for ultra-low memory usage on Apple Silicon.
+
+##  Benchmark Environment
+- Device: MacBook Pro (M3 Pro)
+- Runtime: MLX
+- Quantization: ~3.5 bits per weight
+
+## Tiny Footprint (Measured)
+- Disk size: ~155 MB
+- Peak memory: ~0.20 GB
+- Generation speed: ~458 tokens/sec (short generation)
+
+> These numbers were measured on macOS (M3 Pro).  
+> This is an **extreme compression** build and may reduce output quality vs 4/5-bit.
+
+## Usage
+```bash
+mlx_lm.generate \
+  --model Irfanuruchi/SmolLM-360M-Instruct-MLX-3bit \
+  --prompt "Reply with exactly 3 bullet points, 4-8 words each: what can you do offline?" \
+  --max-tokens 80
+```
+
+## License
+Upstream SmolLM is released under **Apache-2.0**. Preserve attribution and the original license terms.
+