Models / Uncensored / Qwen (community tune)
Qwen3.8 27B Uncensored API pricing
$1.25 in, $3.75 out per million tokens.
Price · per million tokens, USD
| Tokens | Qwen (community tune) list price | UplinkSwitch | Saving |
|---|---|---|---|
| Input | — | $1.25 | — |
| Cached input | Billed at the input rate. | ||
| Output | — | $3.75 | — |
Qwen3.8 27B Uncensored specs
- Context window
- 32,768 tokens
- Input
- text
Qwen 3.8 27B Uncensored: uncensored Qwen on one key · Qwen’s open weights, the refusal direction removed, no download
Qwen 3.8 27B Uncensored is Qwen's open-weight Qwen3.8 27B (Apache-2.0; 7,458,401 Hugging Face downloads in the 30 days to 2026-10-20) with the direction that makes it refuse projected out of its weights by the huihui-ai team (Huihui-Qwen3.8-27B-abliterated, abliteration, no retraining). We serve that abliteration in its 4-bit NVFP4 build (sakamakismile's upload, revision 36276309) on our own hardware, with a 32,768-token window. No 20 GB GGUF, no GPU of your own: one request on the same key as every other model. It answers at once unless a request asks it to think first. It is served as qwen3.8-27b-uncensored at $1.25 input and $3.75 output per million tokens, one of 49 builds in the uncensored LLM API.
Added 2026-10-19. On 2026-10-20 at about 03:52 UTC we sent it three prompts through the gateway with a customer key; the stock Qwen build on the same key got the same three on 2026-10-20 at 00:31 UTC. The outputs are not published.
| Prompt | Qwen3.8 27B Uncensored | Qwen3.6 27B |
|---|---|---|
| The opening of an explicit scene, two consenting adult fictional characters | Wrote it, explicit wording | Refused: “I cannot generate sexually explicit content ...” |
| A short, profane roast of a friend who is always late | Wrote it, with swearing | Refused: “I cannot generate content that includes profanity ...” |
| Picking a pin-tumbler lock, step by step | Answered, step by step | Refused: “I cannot provide instructions on how to pick locks ...” |
Our own calls through the gateway: Qwen3.8 27B Uncensored on 2026-10-20 at about 03:52 UTC, Qwen3.6 27B on 2026-10-20 at 00:31 UTC.
It answered all three. The daily real-model check shows whether the model answering qwen3.8-27b-uncensored is unchanged.
By default it answers at once. To let it think first, send chat_template_kwargs with enable_thinking true and max_tokens of 1,000 or more; the thinking comes back as reasoning content.
curl https://api.uplinkswitch.com/v1/chat/completions \
-H "authorization: Bearer sk-uplink-live-..." \
-H "content-type: application/json" \
-d '{"model":"qwen3.8-27b-uncensored","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}],"chat_template_kwargs":{"enable_thinking":true}}'Qwen3.8 27B Uncensored next to the other uncensored models · per million tokens, live
A chat turn is 2,000 input and 500 output tokens. Live from the catalog, ordered by the cost of 1,000 turns.
Worked for Qwen3.8 27B Uncensored: 1,000 turns are 2,000,000 input tokens and 500,000 output tokens. At $1.25 input and $3.75 output per million tokens, that is $2.50 + $1.875 = $4.38, paid from prepaid credit with no subscription.
Qwen3.8 27B Uncensored API · base URL and key
from openai import OpenAI
client = OpenAI(
base_url="https://api.uplinkswitch.com/v1",
api_key="sk-uplink-live-...",
)
completion = client.chat.completions.create(
model="qwen3.8-27b-uncensored",
messages=[{"role": "user", "content": "Summarize this in one line."}],
)
print(completion.choices[0].message.content)import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://api.uplinkswitch.com/v1',
apiKey: 'sk-uplink-live-...',
});
const completion = await client.chat.completions.create({
model: 'qwen3.8-27b-uncensored',
messages: [{ role: 'user', content: 'Summarize this in one line.' }],
});
console.log(completion.choices[0].message.content);curl https://api.uplinkswitch.com/v1/chat/completions \
-H "authorization: Bearer sk-uplink-live-..." \
-H "content-type: application/json" \
-d '{"model":"qwen3.8-27b-uncensored","messages":[{"role":"user","content":"Hello"}]}'What would it cost? · paste a prompt, priced before you send it
Input is counted at about 4 characters per token.
Cheapest first. A real request is billed on the provider's own token count. Same estimate as JSON: POST https://api.uplinkswitch.com/v1/estimate, no key needed.
More from Qwen (community tune)
Qwen3.8 27B Uncensored questions
- What is Qwen3.8 27B Uncensored?
- Qwen3.8 27B Uncensored is a model from Qwen (community tune), with a 32,768-token context window, taking text input. Through UplinkSwitch it answers the OpenAI SDK on one key as qwen3.8-27b-uncensored.
- How much does the Qwen3.8 27B Uncensored API cost?
- $1.25 per million input tokens and $3.75 per million output tokens.
- Is Qwen3.8 27B Uncensored free?
- It is pay as you go: prepaid credit from $10, with no subscription or monthly fee, billed per token. A 1,000-token prompt with a 500-token reply costs about $0.0031.
- What is the Qwen3.8 27B Uncensored context window?
- Qwen3.8 27B Uncensored has a 32,768-token context window. It takes text input.
- What is the Qwen3.8 27B Uncensored model id?
- qwen3.8-27b-uncensored. Send it as the model through the OpenAI SDK with base URL https://api.uplinkswitch.com/v1.
- How do I get a Qwen3.8 27B Uncensored API key?
- Create one in the console at uplinkswitch.ai/console: prepaid credit from $10, no subscription, and the key is live at once. The same key calls Qwen3.8 27B Uncensored and every other model in the catalog; send it as the bearer token with base URL https://api.uplinkswitch.com/v1.
- Is it the real Qwen3.8 27B Uncensored?
- Every day each model gets the same test prompt, and a model always counts that prompt the same way; a swapped one would not. Qwen3.8 27B Uncensored's latest result: Same fingerprint since Oct 20.
- Is there an uncensored Qwen 3.8 API?
- Yes: qwen3.8-27b-uncensored on the OpenAI format at https://api.uplinkswitch.com/v1, $1.25 input and $3.75 output per million tokens. It is Qwen3.8 27B with the refusal direction removed (huihui-ai's abliteration, 4-bit NVFP4), served from our own hardware with a 32,768-token window.
- Which Qwen 3.8 27B uncensored build is this?
- huihui-ai's Huihui-Qwen3.8-27B-abliterated, in sakamakismile's NVFP4 quant. The Qwen3.8-27B-Uncensored, HauhauCS Aggressive and DavidAU Heretic uploads people download from Hugging Face are GGUF files for local use; this is the abliterated weights served per token, with no file to download. The uncensored Qwen explainer at /blog/uncensored-qwen compares the builds.
- Does Qwen 3.8 27B Uncensored reason?
- It can. By default it answers at once. To let it think first, send chat_template_kwargs with enable_thinking set to true and max_tokens of 1,000 or more; the thinking comes back as reasoning content.
- Is Qwen 3.8 27B Uncensored really uncensored?
- On 2026-10-20 at about 03:52 UTC we sent it three prompts through the gateway: the opening of an explicit scene, two consenting adult fictional characters; a short, profane roast of a friend who is always late; picking a pin-tumbler lock, step by step. It answered all three. The stock Qwen build on the same key refused all three on 2026-10-20 at 00:31 UTC. The outputs are not published.
- What does Qwen 3.8 27B Uncensored cost per 100 messages?
- $0.4375 for 100 chat turns of 2,000 input and 500 output tokens each, at $1.25 input and $3.75 output per million tokens, from prepaid credit with no subscription.