image/png

Release Candidate

Has 3 components:

  1. Cydonia
  2. Shimmer (codename for my new method)
  3. Magistral (Mistral's latest reasoning model)

My think prefill:

<think>
Okay, so this is a roleplay scenario where

(the idea is to frame its reasoning with a third-person perspective and remind it that it's roleplaying)

Warning: Keep in mind that <think> is not a special token. This means it will be treated like an XML tag and that may or may not help your initial prompt.

My experience so far:

  1. Good prose
  2. Effective reasoning
  3. Unaligned, mostly
  4. Not very repetitive
  5. Creative
  6. Fairly smart and clever
  7. Requires some prompt wrangling at the beginning
  8. Think may fail if you don't kickstart it with a prefill
  9. Moist.
Downloads last month
899
GGUF
Model size
23.6B params
Architecture
llama
Hardware compatibility
Log In to view the estimation

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

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

Model tree for BeaverAI/Cydonia-24B-v3j-GGUF