models(gallery): add lumimaidv2 (#2312)

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
mudler · May 13, 2024 · 2db2208 · 2db2208
1 parent fa7b2ae
commit 2db2208
Showing 1 changed file with 23 additions and 0 deletions.
diff --git a/gallery/index.yaml b/gallery/index.yaml
@@ -572,6 +572,29 @@
     - filename: Llama-3-Lumimaid-8B-v0.1-OAS-Q4_K_M-imat.gguf
       sha256: 1199440aa13c55f5f2cad1cb215535306f21e52a81de23f80a9e3586c8ac1c50
       uri: huggingface://Lewdiculous/Llama-3-Lumimaid-8B-v0.1-OAS-GGUF-IQ-Imatrix/Llama-3-Lumimaid-8B-v0.1-OAS-Q4_K_M-imat.gguf
+- !!merge <<: *llama3
+  name: "llama-3-lumimaid-v2-8b-v0.1-oas-iq-imatrix"
+  urls:
+    - https://huggingface.co/Lewdiculous/Llama-3-Lumimaid-8B-v0.1-OAS-GGUF-IQ-Imatrix
+  icon: https://cdn-uploads.huggingface.co/production/uploads/65d4cf2693a0a3744a27536c/JUxfdTot7v7LTdIGYyzYM.png
+  license: cc-by-nc-4.0
+  description: |
+    This model uses the Llama3 prompting format.
+
+    Llama3 trained on our RP datasets, we tried to have a balance between the ERP and the RP, not too horny, but just enough.
+
+    We also added some non-RP dataset, making the model less dumb overall. It should look like a 40%/60% ratio for Non-RP/RP+ERP data.
+
+    "This model received the Orthogonal Activation Steering treatment, meaning it will rarely refuse any request."
+
+    This is v2!
+  overrides:
+    parameters:
+      model: v2-Llama-3-Lumimaid-8B-v0.1-OAS-Q4_K_M-imat.gguf
+  files:
+    - filename: v2-Llama-3-Lumimaid-8B-v0.1-OAS-Q4_K_M-imat.gguf
+      sha256: b00b4cc2ea4e06db592e5f581171758387106626bcbf445c03a1cb7b424be881
+      uri: huggingface://Lewdiculous/Llama-3-Lumimaid-8B-v0.1-OAS-GGUF-IQ-Imatrix/v2-Llama-3-Lumimaid-8B-v0.1-OAS-Q4_K_M-imat.gguf
 - !!merge <<: *llama3
   name: "suzume-llama-3-8B-multilingual"
   urls: