{
 "schema_version": 1,
 "generated": "2026-10-20",
 "id": "yamz-labs-kyojin--load-less-restricted-model-for-testing",
 "repo": "Yamz-Labs/kyojin",
 "repo_url": "https://github.com/Yamz-Labs/kyojin",
 "repo_slug": "yamz-labs-kyojin",
 "title": "Load a less-restricted model variant locally",
 "page_title": "How to load a less-restricted model variant locally with Kyojin",
 "summary": "Run an optional uncensored mode on GLM, MiMo or the Qwen preset for controlled red-team and research tests.",
 "who": "A security researcher or AI safety tester who needs a model that refuses fewer requests to study its behavior on their own machine.",
 "problem": "Hosted models refuse many requests, which blocks some legitimate adversarial testing and evaluation work.",
 "outcome": "A local model load with the uncensored mode applied at load time, with the original weights left unchanged.",
 "steps": [
  "Read doc/refusal_hook.md to understand how the mode works.",
  "Download the uncensored pack for GLM-5.3-Flash or MiMo-V2.6-Flash from the links in the README, or enable the preset in the Qwen3.8-Flash-Next pack.",
  "Follow the serve commands in doc/models.md for that model.",
  "Run only prompts that are part of an approved test plan and log every request and response."
 ],
 "agent_prompt": "In the Yamz-Labs/kyojin repo, read doc/refusal_hook.md and load the uncensored variant of <target model> following doc/models.md. Expected output: the server starts with the mode applied and returns responses to a scoped test prompt. Verify that the test is covered by an approved plan and that all requests are logged before running it.",
 "agents": [
  "any"
 ],
 "difficulty": "Advanced",
 "time": "3 hours",
 "basis": "readme",
 "caveat": "The README says the mode is off by default for the Qwen preset and that the other packs are first versions, so behavior may change between releases.",
 "category": "Software and DevOps",
 "tags": [
  "Software and DevOps"
 ],
 "lang": "python",
 "requirements": [
  "AMD Ryzen AI Max+ 395 mini PC with 128 GB of memory (Strix Halo, gfx1151)",
  "ROCm build of torch, installed per doc/install.md",
  "Python and the hf command-line tool to download model packs",
  "Roughly 100 GB of free disk space for a model pack (Qwen pack is 95 GB)"
 ],
 "risk": "dual-use",
 "seen_in_the_wild": false,
 "mentioned_publicly": false,
 "risks": [
  {
   "tag": "reverse-engineering",
   "notice": "The uncensored mode is designed to make the model refuse far less, so outputs must be kept inside an approved test plan and not used to generate harmful content."
  }
 ],
 "best": false,
 "score": 0.6,
 "evidence": [],
 "url": "https://dedupedai.com/use-cases/yamz-labs-kyojin/load-less-restricted-model-for-testing/",
 "md": "https://dedupedai.com/use-cases/yamz-labs-kyojin/load-less-restricted-model-for-testing/index.md",
 "disclaimer": "Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked \"Seen in the wild\", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email hello@dedupedai.com and we will look at it promptly.",
 "api": "https://dedupedai.com/api/v1/usecases/yamz-labs-kyojin--load-less-restricted-model-for-testing.json",
 "api_version": "v1"
}