Downloads · 30 days
0
kasumi-nakano/qwen3-4b-agentbench-v9-dpo
qwen3-4b-agentbench-v9-dpo is a text generation model from kasumi-nakano. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This repository provides a LoRA adapter fine-tuned from kasumi-nakano/qwen3-4b-agentbench-v7-v6-plus-v1 using LoRA + Unsloth.
Downloads · 30 days
0
Access
Public
Updated Mar 1, 2026
Parameters
4.1B
2.7 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors2.7 GB · 99%
How the weights are stored.
U83.7B · 91%
From the Hugging Face model README
This repository provides a LoRA adapter fine-tuned from kasumi-nakano/qwen3-4b-agentbench-v7-v6-plus-v1 using LoRA + Unsloth.
This repository contains LoRA adapter weights only. The base model must be loaded separately.
This adapter is trained to improve multi-turn agent task performance on ALFWorld (household tasks) and DBBench (database operations).
Loss is applied to all assistant turns in the multi-turn trajectory, enabling the model to learn environment observation, action selection, tool use, and recovery from errors.
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
base = "kasumi-nakano/qwen3-4b-agentbench-v7-v6-plus-v1"
adapter = "your_id/your-repo"
tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
base,
torch_dtype=torch.float16,
device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)
Training data: kasumi-nakano/agentbench-dpo-v1-sql-error-recovery
Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License. Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.