dispatchAI/Qwen2.5-Coder-1.5B-mobile

1---2license: apache-2.03language:4  - en5library_name: transformers6tags:7  - mobile8  - on-device9  - quantized10  - gguf11  - dispatchai12pipeline_tag: text-generation13---14 15# Qwen2.5-Coder-1.5B-mobile16 17✅ **Verified on real phone hardware** — Snapdragon 865, June 2026.18 19## Phone Benchmark (Samsung S20 FE, Snapdragon 865)20 21| Metric | Value |22|--------|-------|23| **Phone Speed** | **16.5 tokens/sec** |24| **CPU Speed** | 7.8 tokens/sec |25| **File Size** | 940 MB |26| **Chat Format** | chatml |27| **Test Output** | "Paris" ✅ (correct) |28 29## Usage30 31### Python (llama-cpp-python)32```python33from llama_cpp import Llama34 35llm = Llama(model_path="model.gguf", chat_format="chatml", n_ctx=512, n_threads=4, verbose=False)36response = llm.create_chat_completion(37    messages=[{"role": "user", "content": "What is the capital of France?"}],38    max_tokens=50,39)40print(response["choices"][0]["message"]["content"])41```42 43### dispatchAI SDK44```python45from dispatchai import load_model46model = load_model("Qwen2.5-Coder-1.5B-mobile", backend="gguf")47print(model.chat("What is the capital of France?"))48```49 50### On Android (via ADB)51```bash52hf download dispatchAI/Qwen2.5-Coder-1.5B-mobile model.gguf53MSYS_NO_PATHCONV=1 adb push model.gguf /data/local/tmp/54MSYS_NO_PATHCONV=1 adb shell "cd /data/local/tmp && LD_LIBRARY_PATH=/data/local/tmp ./llama-cli -m model.gguf -p 'Hello' -n 30 -t 4 -st"55```56 57## Model Details58 59| Attribute | Value |60|-----------|-------|61| **Base Model** | unknown |62| **File Size** | 940 MB |63| **Format** | GGUF |64| **Chat Format** | chatml |65| **License** | apache-2.0 |66 67## About dispatchAI68 69[dispatchAI](https://huggingface.co/dispatchAI) — Small. Mobile. Free. UAE-built.70