Instructions to use melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep with Transformers:

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

pipe = pipeline("text-generation", model="melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep")
model = AutoModelForCausalLM.from_pretrained("melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep")
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 melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

SGLang

How to use melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep 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 "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep" \
    --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": "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep",
		"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 "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep" \
        --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": "melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep with Docker Model Runner:
```
docker model run hf.co/melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep
```

agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

A full model fine-tuned from melon1891/agentbench-qwen3-4b-2stage-reasoning-20260228 using LoRA + Unsloth, with the adapter merged into the base model.

Training Objective

This model 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.

Training Configuration

Base model: melon1891/agentbench-qwen3-4b-2stage-reasoning-20260228
Method: LoRA (merged into base)
Max sequence length: 8192
Epochs: 10
Learning rate: 1e-06
LoRA: r=16, alpha=32

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep")
tokenizer = AutoTokenizer.from_pretrained("melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep")

Sources & Terms (IMPORTANT)

Training data: melon1891/alfworld-distilled-sft-dataset-20260301

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.

Downloads last month: 2

Safetensors

Model size

4B params

Tensor type

BF16

Model tree for melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

Base model

Qwen/Qwen3-4B-Instruct-2507

Finetuned

melon1891/agentbench-qwen3-4b-lr5e6-20260224v2

Finetuned

melon1891/agentbench-qwen3-4b-2stage-reasoning-20260228

Finetuned

(6)

this model

melon1891
/

agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

Training Objective

Training Configuration

Usage

Sources & Terms (IMPORTANT)

Model tree for melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep

Dataset used to train melon1891/agentbench-qwen3-4b-2stage-alfw-db-20260301-lr1e6-10ep