Update README.md

README.md

@@ -47,18 +47,20 @@ Mistralai's newest "Mistral-Small-3.1-24B-Instruct-2503" model with "Neo Imatrix
 
 Like almost all of Mistralai's models this model is mostly uncensored.
 
+I have included 3 different system prompts to help this model come alive.
+
 NOTE: This model is text generation only, it does not contain "vision" components.
 
 5 examples provided below with prompts at IQ4XS.
 
 Context: 128k.
 
-"MAXED"
+<B>"MAXED"</B>
 
 This means the embed and output tensor are set at "BF16" (full precision) for all quants.
 This enhances quality, depth and general performance at the cost of a slightly larger quant.
 
-"NEO IMATRIX"
+<B>"NEO IMATRIX"</B>
 
 A strong, in house built, imatrix dataset built by David_AU which results in better overall function, 
 instruction following, output quality and stronger connections to ideas, concepts and the world in general.
@@ -80,11 +82,13 @@ F16
 </pre>
 </small>
 
-Recommend quants IQ3s / IQ4XS / IQ4NL / Q4s for best results for creative.
+Recommend quants IQ3s / IQ4XS / IQ4NL / Q4s for best results for creative use cases. 
+I also suggest using "System Prompt #3" below to really power up this model for creative uses.
 
 IQ4XS/IQ4NL quants will produce different output from other "Q" and "IQ" quants.
+I like these best for creative.
 
-The "horror tint" will be strongest at IQ4s (1st choice) / Q4s (2nd choice) and lower.
+The will be strongest at IQ4s (1st choice) / Q4s (2nd choice) and lower.
 
 Recommend q5s/q6/q8 for general usage.
 
@@ -136,7 +140,7 @@ You are a deep thinking AI, you may use extremely long chains of thought to deep
 
 Option #3 - Creative Plus
 
-This will enhance the creative power of the model.
+This will enhance the creative power of the model. The model will literally come alive with this one - it is like night and day.
 
 Copy and paste exactly as shown, including line breaks.