Instructions to use Rumiii/Qwen3-0.6B-BrowserAgent with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Rumiii/Qwen3-0.6B-BrowserAgent with Transformers:

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

pipe = pipeline("text-generation", model="Rumiii/Qwen3-0.6B-BrowserAgent")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Rumiii/Qwen3-0.6B-BrowserAgent")
model = AutoModelForCausalLM.from_pretrained("Rumiii/Qwen3-0.6B-BrowserAgent", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Rumiii/Qwen3-0.6B-BrowserAgent with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Rumiii/Qwen3-0.6B-BrowserAgent"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Rumiii/Qwen3-0.6B-BrowserAgent",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Rumiii/Qwen3-0.6B-BrowserAgent

SGLang

How to use Rumiii/Qwen3-0.6B-BrowserAgent 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 "Rumiii/Qwen3-0.6B-BrowserAgent" \
    --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": "Rumiii/Qwen3-0.6B-BrowserAgent",
		"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 "Rumiii/Qwen3-0.6B-BrowserAgent" \
        --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": "Rumiii/Qwen3-0.6B-BrowserAgent",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Rumiii/Qwen3-0.6B-BrowserAgent with Docker Model Runner:
```
docker model run hf.co/Rumiii/Qwen3-0.6B-BrowserAgent
```

Qwen3-0.6B-BrowserAgent

A compact browser agent. Given a task goal and a BrowserGym page observation, it outputs only the next action, for example click('37').

It is Qwen/Qwen3-0.6B fine-tuned with LoRA (merged into the weights) on saital/browser-agent-phase1-sft-action-only, a set of successful MiniWoB steps generated by a larger teacher model.

Results

Our fine-tuned Qwen3-0.6B Browser Agent model demonstrates a substantial performance improvement over the base Qwen3-0.6B model across all evaluation metrics.

Most notably, exact match accuracy surged from 0.83% to 89.58%, representing an 88.75 percentage point increase (approximately 108× improvement).

The model achieved perfect 100% action type matching (up from 84.17%) and maintained 100% valid format compliance (up from 98.33%).

Additionally, target element matching improved dramatically from 4.17% to 92.5% (+88.33 pp), while fill text matching increased from 20.0% to 97.5% (+77.5 pp).

These results confirm that our fine-tuning approach effectively enhanced the model's capability to perform precise browser automation tasks with high reliability and accuracy.

Usage

Use non-thinking mode (enable_thinking=False) with the BrowserGym system prompt from the dataset.

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

repo = "Rumiii/Qwen3-0.6B-BrowserAgent"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype=torch.float16, device_map="auto")

messages = [
    {"role": "system", "content": SYSTEM_PROMPT},
    {"role": "user", "content": USER_PROMPT},  # task, goal, URL, history, observation
]
text = tok.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, enable_thinking=False)
inputs = tok(text, return_tensors="pt", add_special_tokens=False).to(model.device)
out = model.generate(**inputs, max_new_tokens=64, do_sample=False)
print(tok.decode(out[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

Training

LoRA SFT (r=64, alpha=128) on all attention and MLP projections, then merged
Loss on action tokens only
3 epochs, learning rate 1e-4, batch size 16, single T4 GPU
Best checkpoint by validation loss

Limitations

Trained on about 30 synthetic MiniWoB task types; little transfer to real-world websites is expected.
Trained on successful trajectories only, so it has not learned to recover from mistakes.
Weakest on tasks that require choosing among many similar elements or copying text exactly.

Downloads last month: 129

Safetensors

Model size

0.6B params

Tensor type

F16

Model tree for Rumiii/Qwen3-0.6B-BrowserAgent

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Adapter

(660)

this model

Adapters

2 models

Rumiii
/

Qwen3-0.6B-BrowserAgent