Qwen3.8 Flash-Next Abliterated

Mixture-of-experts model with refusal behaviour removed; 131,072-token context, one request at a time.

qwen3.8-flash-next-abliterated

Capabilities

Context length: 131,072 tokens, prompt plus output
Parallel requests: 1 per account (max_parallel)
Thinking: On by default; off with chat_template_kwargs.enable_thinking: false
Tools and JSON output: tools, tool_choice and response_format are passed to the model unchanged
Refusals: Removed: "Abliterated" in the name means the model's refusal behaviour was removed. Replies carry no safety warnings; moderate replies before other people see them
Input and output: Text only

Request

The example streams the reply. A non-streamed request that sends no response byte for 100 s ends with 524, a gateway timeout. A streamed request for a model that is not loaded waits for the load and receives : loading comments meanwhile (Streaming).

curl

curl -N https://api.albastov.com/v1/chat/completions \
  -H "Authorization: Bearer $ALBASTAR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.8-flash-next-abliterated",
    "stream": true,
    "max_tokens": 1000,
    "messages": [
      {"role": "user",
       "content": "Write a three-sentence heist scene for a crime novel."}
    ]
  }'

Python

import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.albastov.com/v1",
    api_key=os.environ["ALBASTAR_API_KEY"],
)
stream = client.chat.completions.create(
    model="qwen3.8-flash-next-abliterated",
    stream=True,
    max_tokens=1000,
    messages=[{
        "role": "user",
        "content": "Write a three-sentence heist scene for a crime novel.",
    }],
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
print()

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.albastov.com/v1",
  apiKey: process.env.ALBASTAR_API_KEY,
});
const stream = await client.chat.completions.create({
  model: "qwen3.8-flash-next-abliterated",
  stream: true,
  max_tokens: 1000,
  messages: [{
    role: "user",
    content: "Write a three-sentence heist scene for a crime novel.",
  }],
});
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

Every parameter, response header and error: Chat completions reference.