How to use from the
Use from the
Transformers library
# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="RUC-AIBOX/AutoMaster-bc-4b")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("RUC-AIBOX/AutoMaster-bc-4b")
model = AutoModelForMultimodalLM.from_pretrained("RUC-AIBOX/AutoMaster-bc-4b", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))
Quick Links

AutoMaster-bc-4b

AutoMaster: a teacher robot shares a skill with a student robot

AutoMaster-bc-4b is the 4B AutoMaster checkpoint for multi-step web search and evidence-based question answering. Based on Qwen3.5-4B, it is trained with Skill-Guided Fine-Tuning (Skill-FT) on demonstrations produced by a skill-guided teacher.

Paper Project homepage AutoMaster model collection Code on GitHub

Model Overview

Property Description
Base model Qwen3.5-4B
Domain Search
Benchmark BrowseComp (Lite-185)
Training Full-parameter supervised fine-tuning with Skill-FT
Release format Full model weights in Hugging Face Safetensors format
Inference framework AweAgent

This repository contains the fine-tuned student model. Generated skills shape the teacher's training demonstrations and are absent during student evaluation.

Method: Skill-Guided Fine-Tuning

Successful trajectories can still demonstrate weak habits, such as unsupported assumptions or superficial verification. AutoMaster uses an LLM agent to study trajectories, develop natural-language skills, and select demonstrations that teach more effective behavior.

  • Inner loop: analyze behavior, design and test skills, and curate teacher demonstrations.
  • Outer loop: fine-tune the student and use evaluation feedback to guide the next round.

AutoMaster inner loop for skill design and demonstration selection, and outer loop for student training and evaluation

Framework figure from the AutoMaster paper, distributed with the public project repository.

Training

The paper uses 512 training trajectories and three epochs of full-parameter SFT per run. Skill instructions remain in the training context. The Vanilla SFT comparison uses the same teacher, task pool, training set size, and SFT recipe, with successful unguided teacher trajectories.

Evaluation

Results on BrowseComp (Lite-185), as reported in Table 1 of the paper:

Model Accuracy (%) ↑ (mean ± SD)
Qwen3.5-4B 33.69 ± 0.82
Qwen3.5-4B + Vanilla SFT 43.24 ± 0.54
AutoMaster-bc-4b 46.13 ± 2.98

Scores are the mean ± sample standard deviation over three evaluation runs per checkpoint. Evaluation uses the domain-specific AweAgent configuration, without generated skill prompts. See the paper and inference guide for task assets, tools, and evaluation settings.

Using the Model

Use a Qwen3.5-compatible serving backend and keep the checkpoint's tokenizer and chat template with its weights. The public repository provides a serving script and benchmark configurations:

# Run from the AutoMaster repository after setting up its inference environment.
export MODEL_PATH="RUC-AIBOX/AutoMaster-bc-4b"
export SERVED_MODEL_NAME=automaster
bash recipes/inference/serve_qwen35.sh

In a separate terminal with AweAgent installed and the benchmark assets configured:

awe-agent run --config recipes/inference/browsecomp.yaml --max-instances 1

Follow the inference guide to configure the endpoint and required tools or sandbox.

Related Models

bc-4b · bc-9b · doc2repo-4b · doc2repo-9b · tb21-4b · tb21-9b

Citation

@misc{chen2026automasterautonomousposttrainingmaking,
  title={AutoMaster: Autonomous Post-Training by Making the Same Model a Better Teacher},
  author={Guoxin Chen and Shuang Sun and Fanzhe Meng and Xinyu Tang and Ruihua Song and Wayne Xin Zhao},
  year={2026},
  eprint={2611.26510},
  archivePrefix={arXiv},
  primaryClass={cs.CL},
  url={https://arxiv.org/abs/2611.26510}
}
Downloads last month
16
Safetensors
Model size
5B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for RUC-AIBOX/AutoMaster-bc-4b

Finetuned
Qwen/Qwen3.5-4B
Finetuned
(1,545)
this model

Collection including RUC-AIBOX/AutoMaster-bc-4b

Paper for RUC-AIBOX/AutoMaster-bc-4b