Instructions to use tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2")
model = AutoModelForCausalLM.from_pretrained("tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2

SGLang

How to use tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2 with Docker Model Runner:
```
docker model run hf.co/tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2
```

Qwen3-4B-AgentBench-Merged-v2-2

This repository provides a merged full model fine-tuned from Qwen/Qwen3-4B-Instruct-2507 using LoRA + Unsloth. The LoRA adapter has been merged into the base model weights.

The model can be loaded directly without requiring a separate adapter.

Training Objective

This model is trained for multi-turn agent-style reasoning tasks, including structured tool use and database-oriented reasoning.

Loss is applied to all assistant turns within each trajectory.

Training Configuration

Base model: Qwen/Qwen3-4B-Instruct-2507
Method: LoRA (merged)
Max sequence length: 4096
Epochs: 2
Learning rate: 2e-06
LoRA config: r=64, alpha=128

Training Data

The model was trained on a merged dataset created by concatenating and shuffling the following datasets:

tussiiiii/openalex_dbbench_synth_v1
tussiiiii/openalex_dbbench_synth_v2
tussiiiii/alfworld_synth_v1

Dataset Details

tussiiiii/openalex_dbbench_synth_v1
tussiiiii/openalex_dbbench_synth_v2
tussiiiii/alfworld_synth_v1

The three datasets above were independently created by the author. They are fully synthetic and were generated from scratch. No benchmark evaluation data was used in their creation.

License & Compliance

Users must comply with:

The license of each dataset listed above
The license of the base model: Qwen/Qwen3-4B-Instruct-2507

This repository does not claim ownership of third-party datasets. Synthetic datasets were independently generated.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "your_id/your-repo"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

Downloads last month: 6

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for tussiiiii/Qwen3-4B-AgentBench-Merged-v2-2

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

(2084)

this model

tussiiiii
/

Qwen3-4B-AgentBench-Merged-v2-2