Instructions to use apus-ailab/APUS-Reflex-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use apus-ailab/APUS-Reflex-4B with Transformers:

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

pipe = pipeline("text-generation", model="apus-ailab/APUS-Reflex-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("apus-ailab/APUS-Reflex-4B")
model = AutoModelForMultimodalLM.from_pretrained("apus-ailab/APUS-Reflex-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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use apus-ailab/APUS-Reflex-4B with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/apus-ailab/APUS-Reflex-4B

SGLang

How to use apus-ailab/APUS-Reflex-4B 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 "apus-ailab/APUS-Reflex-4B" \
    --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": "apus-ailab/APUS-Reflex-4B",
		"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 "apus-ailab/APUS-Reflex-4B" \
        --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": "apus-ailab/APUS-Reflex-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use apus-ailab/APUS-Reflex-4B with Docker Model Runner:
```
docker model run hf.co/apus-ailab/APUS-Reflex-4B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

APUS Reflex 4B

Branding note: This model family is now APUS Reflex. Some weight filenames, script/module names, served-model aliases, and the separate evaluation dataset ID retain their historical identifiers for download and runtime compatibility; these do not indicate different model versions.

English | 中文 · Collection · Model family · Technical Report · Runtime

A Qwen3.5-based decision model for browser action selection, workflow routing, and natural-language principle judgments. This repository contains 4B checkpoint-5949 merged BF16 weights, ready to download independently without a separate LoRA adapter.

Highlights

Score dynamic candidates supplied with each request and return their distribution.
The included native runtime supports effort=low (16 layers) and effort=high (32 layers). Use high for text generation.
Reuse Qwen language representations and vocabulary projection; application code can assemble decisions into structured workflow outputs.

Quick start

python -m pip install huggingface_hub
hf download apus-ailab/APUS-Reflex-4B --local-dir ./APUS-Reflex-4B
cd APUS-Reflex-4B
python -m pip install -r requirements.txt
python examples.py . --device cuda:0 --effort high

Evaluation and training

The merged model in this repository scores 66/80 (82.50%) at full depth on the Frozen80 development panel, covering Browser, HelpSteer3, BoolQ, MNLI, and attribute decisions. See merged-evaluation.json and training.md. This reused engineering panel is not an independent blind benchmark or an end-to-end browser success rate.

Candidate probabilities express relative preference; calibration is required before interpreting them as correctness probabilities. BF16 merging changes some probabilities; see runtime evidence and limitations. The series 9B release selects checkpoint-3000, corresponding to the 85% development-panel result. Checkpoint-5949 remains available in the original family repository.

Series and downloads

The 4B, 9B, and 35B-A3B models use independent repositories grouped in a Collection. Hugging Face displays downloads per model. The original family repository and legacy paths remain available. The 35B-A3B merged release is subject to its own validation and publication status.

GGUF and MLX versions

Quantized versions of this model (Frozen80 66/80) for Ollama / llama.cpp / LM Studio and Apple Silicon Macs: GGUF collection · MLX collection.

Version	Frozen80	Decisions = BF16
GGUF Q8_0	67/80	79/80
GGUF Q4_K_M	67/80	77/80
MLX 8bit	66/80	80/80
MLX 4bit	67/80	79/80

ollama run hf.co/apus-ailab/APUS-Reflex-4B-GGUF:Q8_0 --think=false

Ollama needs thinking disabled (--think=false, or "think": false in the API). Per-question results and usage are in each repository.

License and acknowledgments

We thank the Qwen/Qwen3.5-4B team. See LICENSE, provenance, and the pinned source and file identities in release-manifest.json.

Authors: gumpcheng (xDAN2099), zhangxu, APUS AI-LAB.

Downloads last month: 1,473

Safetensors

Model size

5B params

Tensor type

BF16

Model tree for apus-ailab/APUS-Reflex-4B

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

(952)

this model

Quantizations

6 models

Space using apus-ailab/APUS-Reflex-4B 1

Collection including apus-ailab/APUS-Reflex-4B

APUS Reflex

Collection

Language-native decision models: 4B, 9B, and 35B-A3B. Standalone BF16 releases with candidate scoring, training provenance, and evaluation evidence. • 3 items • Updated 1 day ago