Model Card: Qwen3.5-Qwen3.6-plus-Reasoning-Distilled

Overview

This model is a distilled reasoning-enhanced variant of Qwen3.5-2B, designed to improve:

• Structured reasoning
• Step-by-step problem solving
• Decision stability
• Output efficiency (token usage)

The model is trained via distillation from a stronger reasoning model (Qwen3.6-plus), transferring:

• Clean reasoning trajectories
• Better stopping behavior
• Reduced reasoning noise

Key Improvements Over Base Model

Reasoning Efficiency

Compared to the base model, this model:

• Produces shorter and more relevant reasoning chains
• Avoids repetitive self-verification loops
• Maintains high signal-to-noise ratio

Stability

The base model often exhibits:

• Overthinking
• Infinite or near-infinite reasoning loops
• Hypothesis explosion

This distilled model:

• Converges faster to a solution
• Maintains deterministic reasoning paths
• Avoids reasoning drift

Decision-Making

• Improved reasoning termination policy
• Clearer final answers
• Better alignment between reasoning and output

Known Failure Modes

• Occasional hallucinated justifications
• Overconfidence in incorrect options
• Missing rare edge-case interpretations
• Limited deep domain reasoning beyond training distribution

How to Get Started with the Model

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("khazarai/Qwen3.5-2B-Qwen3.6-plus-Distilled")
model = AutoModelForCausalLM.from_pretrained(
    "khazarai/Qwen3.5-2B-Qwen3.6-plus-Distilled",
    device_map={"": 0}
)

question = """
An 8-year-old boy is brought to the pediatrician by his mother with nausea, vomiting, and decreased frequency of urination. He has acute lymphoblastic leukemia for which he received the 1st dose of chemotherapy 5 days ago. His leukocyte count was 60,000/mm3 before starting chemotherapy. The vital signs include: pulse 110/min, temperature 37.0°C (98.6°F), and blood pressure 100/70 mm Hg. The physical examination shows bilateral pedal edema. Which of the following serum studies and urinalysis findings will be helpful in confirming the diagnosis of this condition? ?
{'A': 'Hyperkalemia, hyperphosphatemia, hypocalcemia, and extremely elevated creatine kinase (MM)', 'B': 'Hyperkalemia, hyperphosphatemia, hypocalcemia, hyperuricemia, urine supernatant pink, and positive for heme', 'C': 'Hyperuricemia, hyperkalemia, hyperphosphatemia, lactic acidosis, and urate crystals in the urine', 'D': 'Hyperuricemia, hyperkalemia, hyperphosphatemia, and urinary monoclonal spike', 'E': 'Hyperuricemia, hyperkalemia, hyperphosphatemia, lactic acidosis, and oxalate crystals'}
"""

messages = [
    {
        "role": "user",
        "content": [
            {"type": "text", "text": question}
        ]
    },
]

inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
    enable_thinking = True,
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=2048, do_sample = True, top_p=0.95, top_k=20, temperature=1.0, min_p=0.0, repetition_penalty=1.0)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))