Update README.md

README.md

@@ -29,7 +29,7 @@ language:
   - ja
 ---
 
-# Ix Chel L3 12B
+# Lusca 18B
 
 
 
@@ -38,10 +38,11 @@ to be merely a test of the viability of various stacking techniques for nemo, bu
 to publish.  Sort of.
 
 Mergefuel: 
-- v000000/L3-8B-Poppy-Sunspice
-- mlabonne/NeuralDaredevil-8B-abliterated
+- cognitivecomputations/dolphin-2.9.3-mistral-nemo-12b
+- BeaverAI/mistral-dory-12b
+- intervitens/mini-magnum-12b-v1.1
 
-See recipe.yml for full details.  As all components are L3-based, coherency out to 32k context appears to be good.  
+See recipe.txt for full details.  As all components are nemo-based, coherency out to 64k context appears to be good.  
 
 This model is uncensored and perfectly capable of generating objectionable material.  It does not seem to have a 
 propensity to insert NSFW content into SFW prompts, but YMMV.  As with any LLM, no factual claims