apus-ailab

APUS-OpenJev-v1-9B

README

License: apache-2.0

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_hubhf download apus-ailab/APUS-OpenJev-v1-9B --local-dir ./APUS-OpenJev-v1-9Bcd APUS-OpenJev-v1-9Bpython -m pip install -r requirements.txtpython 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, 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 68/80) for Ollama / llama.cpp / LM Studio and Apple Silicon Macs: GGUF collection · MLX collection.

Table with columns: Version, Frozen80, Decisions = BF16
Version	Frozen80	Decisions = BF16
GGUF Q8_0	68/80	80/80
GGUF Q4_K_M	68/80	78/80
MLX 8bit	69/80	79/80
MLX 4bit

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.

License and acknowledgments

We thank the Qwen/Qwen3.5-9B team. See LICENSE, provenance, and the pinned source and file identities in release-manifest.json.

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

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

apus-ailab

Model Tree

Base

Qwen/Qwen3.5-9B

Fine-tuned

this model

Input Modalities

TextImageVideo

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

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_hubhf download apus-ailab/APUS-OpenJev-v1-9B --local-dir ./APUS-OpenJev-v1-9Bcd APUS-OpenJev-v1-9Bpython -m pip install -r requirements.txtpython examples.py . --device cuda:0 --effort high

Evaluation and training

Series and downloads

GGUF and MLX versions

Quantized versions of this model (Frozen80 68/80) for Ollama / llama.cpp / LM Studio and Apple Silicon Macs: GGUF collection · MLX collection.

Table with columns: Version, Frozen80, Decisions = BF16
Version	Frozen80	Decisions = BF16
GGUF Q8_0	68/80	80/80
GGUF Q4_K_M	68/80	78/80
MLX 8bit	69/80	79/80
MLX 4bit

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.

License and acknowledgments

We thank the Qwen/Qwen3.5-9B team. See LICENSE, provenance, and the pinned source and file identities in release-manifest.json.

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