Qwen3.5-35B-A3B-Unredacted-MAX

Qwen3.5-35B-A3B-Unredacted-MAX is an optimized release built on top of huihui-ai/Huihui-Qwen3.5-35B-A3B-abliterated. This version focuses on improved repository packaging, updated shard layout, and compatibility enhancements for modern Transformers and inference ecosystems, while preserving the reasoning and instruction-following capabilities of the base model. The result is a powerful 35B parameter Mixture-of-Experts language model designed for stable inference, scalable deployment, and research-oriented experimentation.

This model is intended strictly for research and learning purposes. Any outputs generated by this model are the sole responsibility of the user. The authors and hosting platform disclaim all liability for generated content. Users must ensure safe, ethical, and lawful usage.

---

Base Model Signatures:

This model has been re-sharded and optimized for the latest Transformers version from the base model: https://huggingface.co/huihui-ai/Huihui-Qwen3.5-35B-A3B-abliterated

---

Key Highlights

• Optimized Repository Packaging Improved structure for cleaner downloads, versioning, and deployment workflows.

• Updated Sharding Layout Enhanced weight distribution format for better loading performance and compatibility.

• 35B MoE Architecture (A3B) Built on Qwen3.5-35B-A3B, leveraging Mixture-of-Experts for scalable reasoning capacity.

• Stable Transformers Integration Designed for smoother compatibility with modern Transformers versions and inference pipelines.

• Preserved Model Behavior No changes to weights or architecture; performance remains consistent with the upstream base model.

• Improved Deployment Reliability Reduced friction in initialization, multi-GPU usage, and distributed inference setups.

---

Quick Start with Transformers

pip install transformers==5.3.0
# or
pip install git+https://github.com/huggingface/transformers.git

from transformers import Qwen3_5MoeForConditionalGeneration, AutoProcessor
import torch

model = Qwen3_5MoeForConditionalGeneration.from_pretrained(
    "prithivMLmods/Qwen3.5-35B-A3B-Unredacted-MAX",
    torch_dtype="auto",
    device_map="auto"
)

processor = AutoProcessor.from_pretrained(
    "prithivMLmods/Qwen3.5-35B-A3B-Unredacted-MAX"
)

messages = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "Explain how transformer models work in simple terms."}
        ],
    }
]

text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)

inputs = processor(
    text=[text],
    padding=True,
    return_tensors="pt"
).to("cuda")

generated_ids = model.generate(**inputs, max_new_tokens=256)

generated_ids_trimmed = [
    out_ids[len(in_ids):] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]

output_text = processor.batch_decode(
    generated_ids_trimmed,
    skip_special_tokens=True,
    clean_up_tokenization_spaces=False
)

print(output_text)

---

Intended Use

• Multimodal and Language Research Studying large-scale MoE behavior and inference performance.

• Red-Teaming & Evaluation Testing robustness under adversarial or complex prompting scenarios.

• High-Performance Deployment Running large models on optimized multi-GPU or cloud inference systems.

• Research Prototyping Experimentation with transformer scaling, architecture behavior, and deployment strategies.

---

Limitations & Risks

Important Note: This model inherits behavior from its base model with minimal modification.

• Output Variability Results may vary depending on decoding settings and prompt structure.

• Resource Requirements A 35B MoE model requires substantial GPU memory or optimized inference techniques such as quantization or tensor parallelism.

• Deployment Complexity Performance depends heavily on hardware configuration and runtime optimization.

• General Model Limitations May still produce incorrect, incomplete, or inconsistent outputs in complex scenarios.