Test build of uncensored GATED IQ MOE model for creative usage.

Subject to change, this is one of 9 test builds being considered.

This is a MOE model, with 4 active experts, 32B parameters.

This build requires system prompt(s) for reasoning and top performance.

For reference / system prompts / usage see:

https://huggingface.co/DavidAU/Llama3.1-MOE-4X8B-Gated-IQ-Multi-Tier-COGITO-Deep-Reasoning-32B-GGUF

This model works just like any other "GATED IQ" model, except the models and reasoning model have be altered for creative use cases including using a fully de-censored reasoning model and core models that have also been de-censored and/or have low censorship levels.

Downloads last month
238
GGUF
Model size
24.9B params
Architecture
llama
Hardware compatibility
Log In to view the estimation

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support