Models / Qwen

28 Qwen builds: 8 abliterated, 9 uncensored, 9 roleplay, 2 security.

Get a key

Every Qwen abliterated, uncensored and roleplay build PentestLLM serves, on one key and priced per token. The table lists each build’s price and context window from the live catalog as of 2026-11-18; each name opens its own page with the settings for SillyTavern and Janitor.

Here to chat, not to build? The same builds run as chat, images and video in the browser at Unrestrictable, no key.

Which Qwen build to pick

Largest context
Lowest output price
Abliterated
Qwen 2.5 32B Abliterated, $1.75 per million output tokens
Roleplay
Qwen3.8 27B Antislop, $2.10 per million output tokens
Prices as of
2026-11-18
Uncensored models and their prices per million tokens
Wire formats
OrcaCyber Qwen3.8 Flash Nextqwen3.8-flash-next-cybersecuritySecurity$2.00$4.00262Kopenai
Qwen 2.5 32B Abliteratedqwen2.5-32b-abliteratedAbliterated$1.75$1.7533Kopenai
Qwen 3.8 27B Uncensored Fastqwen3.8-27b-uncensored-fastUncensored$2.50$7.50524Kopenai
Qwen 3.8 27B Uncensored Thinkingqwen3.8-27b-uncensored-thinkingUncensored$2.50$7.501049Kopenai
Qwen3 Coder Next Uncensoredqwen3-coder-next-uncensoredUncensored$0.005 a requestflat120Kopenai
Qwen3 Next 80B-A3B Instruct Uncensoredqwen3-next-80b-a3b-instruct-uncensoredUncensored$0.005 a requestflat120Kopenai
Qwen3 VL 235B-A22B Instruct Uncensoredqwen3-vl-235b-a22b-instruct-uncensoredUncensored$0.005 a requestflat120Kopenai
Qwen3.5 27B Blossom V6.4 Derestrictedqwen3.5-27b-blossom-v6.4-derestrictedAbliterated$1.071$1.071262Kopenai
Qwen3.5 27B BlueStar v3 Derestrictedqwen3.5-27b-bluestar-v3-derestrictedAbliterated$1.071$1.071262Kopenai
Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestrictedqwen3.5-27b-claude-4.6-opus-distilled-derestrictedAbliterated$1.0389$1.0389262Kopenai
Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted Liteqwen3.5-27b-claude-4.6-opus-distilled-derestricted-liteAbliterated$1.071$1.071262Kopenai
Qwen3.5 27B Queen Derestrictedqwen3.5-27b-queen-derestrictedAbliterated$1.071$1.071262Kopenai
Qwen3.5 397B-A17B Uncensoredqwen3.5-397b-a17b-uncensoredUncensored$0.09 a requestflat120Kopenai
Qwen3.8 2.4T-A95B Uncensoredqwen3.8-2.4t-a95b-uncensoredUncensored$0.07 a requestflat120Kopenai
Qwen3.8 27B Abliteratedqwen3.8-27b-abliteratedAbliterated$1.0037$4.5326262Kopenai
Qwen3.8 27B Antislopqwen3.8-27b-antislopRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Cybersecurityqwen3.8-27b-cybersecuritySecurity$0.40$2.40262Kopenai
Qwen3.8 27B DarkIdolqwen3.8-27b-darkidolRoleplay$0.40$2.40262Kopenai
Qwen3.8 27B Fableqwen3.8-27b-fableRoleplay$0.875$5.25524Kopenai
Qwen3.8 27B Fimiqwen3.8-27b-fimiRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Kimikoqwen3.8-27b-kimikoRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Kiwenqwen3.8-27b-kiwenRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Novelistqwen3.8-27b-novelistRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Obliteratedqwen3.8-27b-obliteratedAbliterated$0.582$7.27524Kopenai
Qwen3.8 27B Queenqwen3.8-27b-queenRoleplay$0.875$5.25524Kopenai
Qwen3.8 27B Synthiaqwen3.8-27b-synthiaRoleplay$0.35$2.10262Kopenai
Qwen3.8 27B Uncensoredqwen3.8-27b-uncensoredUncensored$1.2125$3.637533Kopenai
Qwen3.8 Max Uncensoredqwen3.8-max-uncensoredUncensored$1.00$3.00991Kopenai

Qwen abliterated vs uncensored, roleplay and security builds

Abliterated
8 builds, output from $1.0389 per million tokens, context up to 524,288 tokens
Uncensored
9 builds, output from $3.00 per million tokens, context up to 1,048,576 tokens
Roleplay
9 builds, output from $2.10 per million tokens, context up to 524,288 tokens
Security
2 builds, output from $2.40 per million tokens, context up to 262,144 tokens

What the Qwen build types mean

  • Abliterated (8): a abliterated build is the stock model with its refusal direction removed from the weights. It answers prompts the stock model declines, and there is no jailbreak prompt to keep alive.
  • Uncensored (9): a uncensored build is a fine-tune trained to answer without the stock model’s safety refusals.
  • Roleplay (9): a roleplay build is a fine-tune trained on long-form fiction and character dialogue. It holds a character’s voice, follows a scenario and writes at length.
  • Security (2): a security build is set up for offensive-security, malware-analysis and red-team questions the stock models refuse. It runs under the acceptable-use terms like everything else here.

Run Qwen abliterated without a GPU

Nothing is downloaded and no card is needed: the builds run on the API. Create a key in the console, set the base URL to https://api.pentestllm.ai/v1, and send the model id from the table as model. Any client that speaks the OpenAI chat format works, from the SDK to SillyTavern and Janitor; the docs page has the setup for each.

Each build is billed to the token from a prepaid balance. The daily real-model check sends a fixed prompt to every model each day and records whether the answer still matches, so the build behind an id is the one on its label.

Qwen questions

Yes. PentestLLM lists 8 abliterated Qwen builds: Qwen 2.5 32B Abliterated, Qwen3.5 27B Blossom V6.4 Derestricted, Qwen3.5 27B BlueStar v3 Derestricted, Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted, Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted Lite, Qwen3.5 27B Queen Derestricted, Qwen3.8 27B Abliterated and Qwen3.8 27B Obliterated.

Qwen 3.8 27B Uncensored Thinking, with a 1,048,576-token context window.

Output runs from $1.0389 to $7.27 per million tokens across the abliterated Qwen builds here, with input billed separately at each build’s own rate. There is no subscription: a prepaid balance from $10 pays per token.

An abliterated build is the stock model with its refusal direction removed from the weights. It answers prompts the stock model declines, and there is no jailbreak prompt to keep alive. An uncensored build is a fine-tune trained to answer without the stock model’s safety refusals.

Yes. Every build here speaks the OpenAI chat format: point the client at https://api.pentestllm.ai/v1, paste a key and type the model id from the table.

No. An Ollama or GGUF build runs on your own machine and needs the memory for it; these run on the API and are billed per token, so there is nothing to download.

Every abliterated model · Other families: DeepSeek abliterated models · Gemma abliterated models · GLM abliterated models · Llama abliterated models

Help