Datasetpublic

dispatchAI/model-comparison

Model Comparison: Original vs Mobile Shows the size reduction achieved by dispatchAI's re-engineering. Model Original Mobile Reduction SmolLM2-135M 270MB 101MB 62.6% Qwen2.5-0.5B 1000MB 469MB 53.1% Llama-3.2-1B 2500MB 770MB 69.2% 🚀 dispatchAI

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face

0likes11downloads

rootcomparison.json

comparison.json29 linesDownload Raw Back to root

1[2  {3    "mobile_model": "SmolLM2-135M-Instruct-mobile",4    "base_model": "HuggingFaceTB/SmolLM2-135M-Instruct",5    "original_size_mb": 270,6    "mobile_size_mb": 101,7    "size_reduction_pct": 62.6,8    "mobile_speed_tps": 59.7,9    "mobile_phone_tps": 46.010  },11  {12    "mobile_model": "Qwen2.5-0.5B-Instruct-mobile-int4",13    "base_model": "Qwen/Qwen2.5-0.5B-Instruct",14    "original_size_mb": 1000,15    "mobile_size_mb": 469,16    "size_reduction_pct": 53.1,17    "mobile_speed_tps": 12.5,18    "mobile_phone_tps": 23.219  },20  {21    "mobile_model": "Llama-3.2-1B-Instruct-Q4-mobile",22    "base_model": "meta-llama/Llama-3.2-1B-Instruct",23    "original_size_mb": 2500,24    "mobile_size_mb": 770,25    "size_reduction_pct": 69.2,26    "mobile_speed_tps": 11.3,27    "mobile_phone_tps": 5.428  }29]