Made with exllamav2 0.3.1 with default dataset.
These quants can be used with Nvidia RTX GPU on Windows or RTX/AMD ROCm on Linux with TabbyAPI or Text-Generation-WebUI.
Ensure they fit your GPU for optimal performance. For example on a 12GB GPU you might fit 5bpw quant with about 16k context.
With 16GB it might be possible to fit 32k context with 5bpw quant.
23.08.2025 Fixed corrupted generation_config.json file. If you downloaded the model earlier and were unable to load it, just redownload this file.
Original model card
The-Omega-Directive-Qwen3-14B-v1.1
Where Forbidden Knowledge Meets Unparalleled Immersion
⚡ Quantum Leap Forward
This evolution of Forgotten-Safeword delivers coherent depravity with unprecedented immersion: