Field	Value
Base model	`Qwen/Qwen2.5-3B-Instruct`
Method	Online Policy Distillation (OPD)
Teacher	`Qwen/Qwen2.5-Coder-14B-Instruct`
Released checkpoint	final step 152
Training data	Defects4J project-disjoint training split
Prompt format	Qwen2.5 chat template
Global training seed	42

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "tomhu/RL4TG-Qwen2.5-3B-OPD-14B-Teacher"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

The model is intended for research on Java unit-test generation. Generated tests must still be compiled and executed in the target project's own build environment.