Qwen3.5 9B Claude 4.6 OS Auto Variable HERETIC UNCENSORED THINKING MAX NEOCODE Imatrix GGUF EU AI Act Compliance Profile

Fine tune via Unsloth of Qwen 3.5 9B dense model using Claude-4.6-OS dataset (4 Claude datasets) on local hardware. Every attempt was made to ensure the training was "mild" and did not negatively affect the model's already incrediblely strong benchmarks.

Training on 4 datasets improves reasoning, and output generation and exceeds the root model on all benchmarks too.

This is also a HERETIC model, trained post "Heretic'ing" -> this model does what you want, no questions asked.

Fully uncensored.

Example generation(s) at the bottom of this page.

GGUF POWER UPS:

• DI-MATRIX (duel imatrix) of NEO and NEO-CODE imatrix datasets (by DavidAU).
• All Unsloth tensor enhancements.
• BF16 output tensor to improve both thinking and output.
• Custom Jinja template (by DavidAU/Nightmedia) to address some issues with the Qwen 3.5 model series (stablize and reduce overthinking).

A radically stronger, more potent GGUF for all use cases.

VISION:

• Vision (images) tested -> working with new training.
• You need an "mmproj" (just one) of these downloaded too, and placed in the same folder as the GGUF for images.

BENCHMARKS:

Instruct version (as noted above in "benchmarks") is here:

https://huggingface.co/DavidAU/Qwen3.5-9B-Claude-4.6-OS-HERETIC-UNCENSORED-INSTRUCT

DE-CENSORING:

Performance

KLD of less than 1 is excellent, zero is perfect.

| Metric | This model | Original model (Qwen/Qwen3.5-9B) | | :----- | :--------: | :---------------------

Your obligations as a Qwen3.5 9B Claude 4.6 OS Auto Variable HERETIC UNCENSORED THINKING MAX NEOCODE Imatrix GGUF deployer

Track your progress

Sign up to track obligations, receive deadline reminders, and generate audit packages.

Tags

Languages

Library

Base Model