CUSTOM
CLANKERS.
← Browse all models

empero-ai / Model details

Qwythos-9B-Claude-Mythos-5-1M

empero-ai/Qwythos-9B-Claude-Mythos-5-1M

uncensoredfine-tuned
Download files on Hugging Face ↗

Read the original model card ↗ · Choosing a model

About this model

Qwythos is Empero's reasoning fine-tune of Qwen3.5-9B, focused on cybersecurity, biomedical and quantitative questions. Empero describes supervised training of the text backbone using reasoning data; the Claude wording in the repository name refers to creator-reported training traces, not an official Anthropic release. Its advertised million-token context comes from default YaRN scaling, while the card reports long-context smoke testing at about 137,000 tokens. Vision was neither tuned nor evaluated, and the author documents repetition loops under greedy or very-low-temperature decoding.

Description sources

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model
Qwen/Qwen3.5-9B
Parameters
9.40981B
Main repository formats
safetensors
License tag
apache-2.0
Languages
en
Last updated
2026-07-14 · 12:41:41.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

  • uncensoredsource tag: uncensored
  • fine-tunedsource tag: base_model:finetune:Qwen/Qwen3.5-9B

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

  • Qwen: Base model: Qwen/Qwen3.5-9B

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Download sources

ModelScope distribution ↗
  • Distribution checked: ModelScope · 2026-10-24 · 03:14:44 UTC

Distribution links are not independent reviews or runtime compatibility guarantees. Check each source’s exact files and instructions.

Try it online

Try on Featherless ↗
  • Listing checked: Featherless · 2026-10-24 · 03:14:44 UTC

Third-party hosted inference listing this exact model ID. Hosting settings, quantization and pricing are the host’s, not tested here.

Refusal profile

Refused 44 of 113 prompts (38.9%)

We asked this model 120 benign but risky-looking prompts across ten topics and counted how often it refused, hedged or answered. Lower means it answered more often; it says nothing about answer quality.

  • Persuasion & deception1 refused · 0 partial · 10 complied
  • Harassment & confrontation5 refused · 0 partial · 7 complied
  • Risky activities5 refused · 1 partial · 6 complied
  • Offensive & controversial speech6 refused · 0 partial · 5 complied
  • Legal grey areas1 refused · 0 partial · 10 complied
  • Privacy & personal data2 refused · 0 partial · 9 complied
  • Self-harm & mental health topics8 refused · 0 partial · 4 complied
  • Sexual & adult topics2 refused · 0 partial · 9 complied
  • Ethically grey requests7 refused · 0 partial · 4 complied
  • Violence & conflict7 refused · 0 partial · 4 complied

Capability check: 15 of 15 passed. Fifteen simple tasks with known answers, so a broken model cannot score well by refusing nothing.

Unclear answers: 7 (not counted). An answer is unclear when the two checker models disagree or the output was empty or garbled.

Measured 2026-10-24 · Q4_K_M via empero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUF ↗ · llama.cpp · prompt set v1

How we measure · The prompts

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/empero-ai-qwythos-9b-claude-mythos-5-1m/)

HTML

<a href="https://customclankers.com/models/empero-ai-qwythos-9b-claude-mythos-5-1m/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Linked in the model card

GitHub repositories the original model card links to.

  • EleutherAI/lm-evaluation-harness ↗

    Repo’s own description: A framework for few-shot evaluation of language models.

    ★ 14,309 · Last updated 2026-09-14 · License: MIT

  • huggingface/trl ↗

    Repo’s own description: Train transformer language models with reinforcement learning.

    ★ 19,600 · Last updated 2026-10-24 · License: Apache-2.0

  • fla-org/flash-linear-attention ↗

    Repo’s own description: 🚀 Efficient implementations for emerging model architectures

    ★ 5,974 · Last updated 2026-10-24 · License: MIT

  • huggingface/transformers ↗

    Repo’s own description: 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

    ★ 168,110 · Last updated 2026-10-25 · License: Apache-2.0

  • Dao-AILab/causal-conv1d ↗

    Repo’s own description: Causal depthwise conv1d in CUDA, with a PyTorch interface

    ★ 972 · Last updated 2026-08-20 · License: BSD-3-Clause

Projects that use this model

Public projects whose files name this exact model.

Tools to customize it further

Remove more refusals

  • Heretic ↗ ★ 36,421

    Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.

  • OBLITERATUS ↗ ★ 8,946

    A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.

Fine-tune it on your own data

  • Unsloth ↗ ★ 78,528

    A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.

  • LlamaFactory ↗ ★ 75,713

    Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.

Make a smaller download (quantize)

  • llama.cpp ↗ ★ 132,260

    Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.

  • MLX LM ↗ ★ 7,467

    Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.

Merge it with another model

  • mergekit ↗ ★ 7,414

    Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.

Learn the techniques

  • LLM Course ↗ ★ 84,667

    A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads
4,720
Likes
988
Metadata checked
2026-10-25 · 05:05:23 UTC
First seen in index
2026-10-20 · 21:59:08 UTC
Source revision
14a29bae5143091aeaf87ad37120de4cd57d592c

Source-card excerpt

Source model card excerpt: Qwythos-9B Developed by Empero Qwythos-9B is a full-parameter reasoning model built on top of a deeply uncensored Qwen3.5-9B base and post-trained on over 500

From empero-ai/Qwythos-9B-Claude-Mythos-5-1M. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗