Instructions to use apus-ailab/APUS-OpenJev-v1-9B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use apus-ailab/APUS-OpenJev-v1-9B with Transformers:

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

pipe = pipeline("text-generation", model="apus-ailab/APUS-OpenJev-v1-9B")
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)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("apus-ailab/APUS-OpenJev-v1-9B")
model = AutoModelForMultimodalLM.from_pretrained("apus-ailab/APUS-OpenJev-v1-9B", 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=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use apus-ailab/APUS-OpenJev-v1-9B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "apus-ailab/APUS-OpenJev-v1-9B"
# 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-OpenJev-v1-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/apus-ailab/APUS-OpenJev-v1-9B

SGLang

How to use apus-ailab/APUS-OpenJev-v1-9B 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-OpenJev-v1-9B" \
    --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-OpenJev-v1-9B",
		"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-OpenJev-v1-9B" \
        --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-OpenJev-v1-9B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use apus-ailab/APUS-OpenJev-v1-9B with Docker Model Runner:
```
docker model run hf.co/apus-ailab/APUS-OpenJev-v1-9B
```

APUS-OpenJev-v1-9B

File size: 4,098 Bytes

---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3.5-9B
base_model_relation: finetune
pipeline_tag: text-generation
language:
- en
- zh
tags:
- apus-openjev
- decision-model
- structured-output
- bf16
---
# APUS-OpenJev-v1-9B

[English](README.md) | [中文](README.zh-CN.md) · [Collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-6ab1ee888eb002fcdd3a2825) · [Model family](https://huggingface.co/apus-ailab/APUS-OpenJev-v1) · [Technical Report](https://huggingface.co/apus-ailab/APUS-OpenJev-v1/blob/main/TECHNICAL_REPORT.md) · [Runtime](RUNTIME.md)

A Qwen3.5-based decision model for browser action selection, workflow routing, and natural-language principle judgments. This repository contains **9B checkpoint-3000 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

```bash
python -m pip install huggingface_hub
hf download apus-ailab/APUS-OpenJev-v1-9B --local-dir ./APUS-OpenJev-v1-9B
cd APUS-OpenJev-v1-9B
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 **68/80 (85.00%)** at full depth on the [Frozen80 development panel](https://huggingface.co/datasets/apus-ailab/APUS-OpenJev-Eval-Frozen80), covering Browser, HelpSteer3, BoolQ, MNLI, and attribute decisions. See [merged-evaluation.json](merged-evaluation.json) and [training.md](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](RUNTIME.md). 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.

<!-- OPENJEV-QUANT:START -->
## GGUF and MLX versions

Quantized versions of this model (Frozen80 68/80) for Ollama / llama.cpp / LM Studio and Apple Silicon Macs: [GGUF collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-gguf-6ab39d5e724c4d8a1021198f) · [MLX collection](https://huggingface.co/collections/apus-ailab/apus-openjev-v1-mlx-6ab39d5fc988a1b1cb89dfc8).

| Version | Frozen80 | Decisions = BF16 |
|---|---:|---:|
| [GGUF Q8_0](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B-GGUF) | 68/80 | 80/80 |
| [GGUF Q4_K_M](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B-GGUF) | 68/80 | 78/80 |
| [MLX 8bit](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B-MLX-8bit) | 69/80 | 79/80 |
| [MLX 4bit](https://huggingface.co/apus-ailab/APUS-OpenJev-v1-9B-MLX-4bit) | 68/80 | 75/80 |

```bash
ollama run hf.co/apus-ailab/APUS-OpenJev-v1-9B-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.
<!-- OPENJEV-QUANT:END -->

## License and acknowledgments

We thank the [Qwen/Qwen3.5-9B](https://huggingface.co/Qwen/Qwen3.5-9B) team. See [LICENSE](LICENSE), [provenance](provenance/), and the pinned source and file identities in [release-manifest.json](release-manifest.json).

**Authors:** gumpcheng ([xDAN2099](https://huggingface.co/xDAN2099)), zhangxu, [APUS AI-LAB](https://github.com/APUS-AI-Lab).