Qwen3.8 27B uncensored: chat and API

Model and id	Context	Input / 1M	Cached / 1M	Output / 1M
Qwen3.8 27Bqwen3.8-27b-xploded	1M	$1.50	$0.50	$5.00
Qwen3.8 27B Ultrafastqwen3.8-27b-fast-xploded	1M	$1.75	$0.50	$5.50

If you were about to download a GGUF

Most people searching for an uncensored or abliterated Qwen want the weights to run on their own machine. If you have the hardware, that is a fine way to do it. This page is for the other case: you want the model answering in the next minute, on a phone or a laptop, with a million tokens of context that a home GPU cannot hold.

What it is good for

A cheap agent model. It calls tools reliably for its size.
Everyday chat and roleplay where a small model is enough.
Pictures. It reads images.

It is the smallest model here. Its base model scores 34 on the Artificial Analysis Intelligence Index, so for hard problems GLM 5.3 or Kimi K3.1 Pro will do better.

Standard or Ultrafast

Ultrafast is the same model on quicker hardware, for a little more: $1.75 in and $5.50 out per million tokens against $1.50 and $5.00. Choose it when you are watching the words appear. For background jobs the standard build is the better deal.

Uncensored, not abliterated

This Qwen is an uncensored build, not an abliterated one. The difference is explained here.

Calling it

curl https://uncensoredchats.com/v1/chat/completions \
  -H "Authorization: Bearer $UNCENSORED_AI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen3.8-27b-xploded",
       "messages": [{"role": "user", "content": "Hello"}]}'

Keep reading

Updated 12 November 2026. Prices are read from the live price list.