---
title: Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM
canonical_url: "https://www.modelscope.cn/models/XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM"
md_url: "https://www.modelscope.cn/models/XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM.md"
repository: XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM
last_updated: 2026-04-03
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3_5
architectures:
  - Qwen3_5ForConditionalGeneration
base_model:
  - Qwen/Qwen3.5-27B
  - Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled
base_model_relation: quantized
parameters: 6.3B
tensor_type:
  - BF16
  - I32
  - F16
library_name:
  - safetensors
  - pytorch
frameworks:
  - PyTorch
language:
  - en
  - zh
downloads: 73
stars: 0
tags:
  - unsloth
  - qwen
  - qwen3.5
  - reasoning
  - chain-of-thought
  - Dense
  - vLLM
  - SGLang
---

# Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM

> Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM - XingyuZheng 在 ModelScope 开源的模型。Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled通过FOEM量化的INT4版本

XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM 是 ModelScope 魔搭社区上的 6.3B 参数text-generation模型，采用 apache-2.0 许可，基于 Qwen/Qwen3.5-27B、Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled 构建。

- **Repository**: XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 6.3B
- **Base model**: Qwen/Qwen3.5-27B, Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled
- **Tags**: unsloth, qwen, qwen3.5, reasoning, chain-of-thought, Dense, vLLM, SGLang
- **Downloads**: 73
- **Stars**: 0
- **Last updated**: 2026-04-03

Source: https://www.modelscope.cn/models/XingyuZheng/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-INT4-FOEM

---

# 🌟 Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-int4-foem

This is an unofficial quantized version of Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled.

## 🧠 Quantization Framework

[GPTQModel](https://github.com/ModelCloud/GPTQModel)

## 🗺️ Quantization Method

[FOEM (AAAI 2026)](https://ojs.aaai.org/index.php/AAAI/article/view/40123)

FOEM is an improved quantization method over GPTQ. The resulting model preserves the same inference structure as GPTQ, ensuring compatibility with existing deployment pipelines while achieving better accuracy.

## 📚 Calibration Dataset

We randomly sampled 512 examples from [nohurry/Opus-4.6-Reasoning-3000x-filtered](https://huggingface.co/datasets/nohurry/Opus-4.6-Reasoning-3000x-filtered).

## 📋 Usage Example

This model can be deployed using standard frameworks such as **vLLM**, just like other GPTQModel-quantized models.

Example evaluation command:

```bash
lm-eval --model vllm --model_args pretrained=models/gptqmodel/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-int4-foem,tensor_parallel_size=1,dtype=auto,gpu_memory_utilization=0.45 --tasks wikitext --batch_size 1
```

**Note:** Running the latest version of vLLM directly may lead to errors. Please refer to the following discussion for a workaround: https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled/discussions/68

## ⚠️ Limitations & Intended Use

*(Adapted from the original repository of Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled)*

- **Hallucination Risk:** While reasoning is strong, the model remains an autoregressive LLM; external facts provided during the thinking sequence may occasionally contain hallucinations if verifying real-world events.
- **Intended Scenario:** Best suited for offline analytical tasks, coding, math, and heavy logic-dependent prompting where the user needs to transparently follow the AI's internal logic.
- **Preview Version Notice:** Because this model is relatively new and intentionally lightweight, the surrounding ecosystem — including inference templates, fine-tuning pipelines, routing configurations, and tooling integrations — may not yet be fully mature or standardized. As a result, users may encounter occasional bugs, compatibility inconsistencies, or integration edge cases. The current release should be considered a preview build while the broader architectural stack and supporting utilities continue to stabilize and improve.

## 🙏 Acknowledgements
Special thanks to [Jackrong](https://huggingface.co/Jackrong) for providing the original model:
 [Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled](https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled)

## 📖 Citation

If you use this model in your research or projects, please cite:

```bibtex
@misc{jackrong_qwen35_opus_distilled,
  title        = {Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled},
  author       = {Jackrong},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled}}
}
```

```bibtex
@misc{qubitium2024gptqmodel,
  author = {ModelCloud.ai and qubitium@modelcloud.ai},
  title = {GPT-QModel},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/modelcloud/gptqmodel}},
  note = {Contact: qubitium@modelcloud.ai},
  year = {2024},
}
```

```bibtex
@inproceedings{zheng2026first,
  title={First-order error matters: Accurate compensation for quantized large language models},
  author={Zheng, Xingyu and Qin, Haotong and Li, Yuye and Chu, Haoran and Wang, Jiakai and Guo, Jinyang and Magno, Michele and Liu, Xianglong},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={40},
  number={34},
  pages={28883--28891},
  year={2026}
}
```