Instructions to use Zapivara/pyqwen-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- Unsloth Desktop
π PyQwen-4B v0.0.1
A fine-tuned version of Qwen/Qwen3.5-4B specialized in Python development.
π§ What it knows
- Pydantic v2 (validators, aliases, model_validate)
- typing (Generic, ParamSpec, TypeVar)
- collections (ChainMap, defaultdict)
- Decorators & context managers
π Quick start
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("Zapivara/pyqwen-4B")
tokenizer = AutoTokenizer.from_pretrained("Zapivara/pyqwen-4B")
π Training
| Parameter | Value |
|---|---|
| Base model | Qwen/Qwen3.5-4B |
| Dataset | 498 examples |
| Method | QLoRA (LoRA r=16) |
| Epochs | 3 |
| Final loss | 0.380 |
| Hardware | Tesla T4 |
| Time | 52 min |