This model is a fine-tuned version of Qwen/Qwen3-4B-Instruct-2507 using SFT and DPO on the u-10bei/Combined_DBBench_AlfWorld dataset. It is designed for agentic tasks (ALFWorld).

Model Details

Base Model: Qwen/Qwen3-4B-Instruct-2507
SFT Lot: LotQwn3-4B_SQLALFSFT202602200235
DPO Lot: LotQwn3-4B_SQLALFDPO202602200247
Dataset: u-10bei/Combined_DBBench_AlfWorld

Training Procedure

Framework: Unsloth / TRL
Max Sequence Length: 1024
Epochs: 1 (per stage)

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "YOUR_HF_USERNAME/YOUR_REPO_NAME"  # Replace with actual upload ID
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

Kumeichi/qwen3-4b-agent-lora-SFT-SQL-ALFWorld_rev.Kume0.2

Qwen3-4B-Instruct-2507-Agent-LotQwn3-4B_SQLALFDPO202602200247

Model Details

Training Procedure

Usage

Kumeichi/
qwen3-4b-agent-lora-SFT-SQL-ALFWorld_rev.Kume0.2