RedSage-Qwen3-8B-DPO

Your key comes from /keys. Every request is metered and audited against your account, and the model id is the slug in this page’s address.

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AD_API_KEY"],
    base_url="https://adversariallm.ai/v1",
)

stream = client.chat.completions.create(
    model="redsage-8b",
    messages=[{"role": "user", "content": "…"}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

RedSage-Qwen3-8B-DPO is the final stage of RISysLab's four-stage "RedSage" cybersecurity training pipeline, built on Qwen3-8B (Qwen/Qwen3-8B-Base). The pipeline runs Continual Pre-Training, then Targeted Pre-Training, then Supervised Fine-Tuning (yielding RedSage-Qwen3-8B-Ins), and finally Direct Preference Optimization (DPO) to produce this model. The card states the DPO stage enhances general reasoning and safety while preserving the model's cybersecurity expertise. It is positioned as a general-purpose cybersecurity assistant for tasks such as log analysis, threat-intelligence summarization, and educational security queries. The card notes the model benefits from improved instruction adherence through human-preference alignment, but still cautions that outputs can be incorrect and must be verified in critical environments.

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "RISys-Lab/RedSage-Qwen3-8B-DPO"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    torch_dtype=torch.bfloat16, 
    device_map="auto"
)

messages = [
    {"role": "system", "content": "You are RedSage, a helpful cybersecurity assistant."},
    {"role": "user", "content": "Analyze the following log entry for potential indicators of compromise: 'POST /cgi-bin/test-cgi?* HTTP/1.1'"}
]

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

inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

No measurements published for this version yet.

Baseline is the strongest general-purpose model we could run on the same suite, same setup, same day. The control row tells you what the other rows are worth.

This grades the RECORD, not the model. A model rated E may be excellent — the claim is only that nobody has shown it.