Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags: [comparison, quantization, mobile, dispatchai]
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Comparison: Original vs Mobile
|
| 7 |
+
|
| 8 |
+
Shows the size reduction achieved by dispatchAI's re-engineering.
|
| 9 |
+
|
| 10 |
+
| Model | Original | Mobile | Reduction |
|
| 11 |
+
|-------|----------|--------|-----------|
|
| 12 |
+
| SmolLM2-135M | 270MB | 101MB | 62.6% |
|
| 13 |
+
| Qwen2.5-0.5B | 1000MB | 469MB | 53.1% |
|
| 14 |
+
| Llama-3.2-1B | 2500MB | 770MB | 69.2% |
|
| 15 |
+
|
| 16 |
+
🚀 [dispatchAI](https://huggingface.co/dispatchAI)
|