Upload 2 files

app.py

@@ -48,16 +48,16 @@ with gr.Blocks(theme="NoCrypt/miku@>=1.2.2", fill_width=True, css="") as app:
                 chat_topp = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p")
                 chat_topk = gr.Slider(minimum=0, maximum=100, value=40, step=1, label="Top-k")
                 chat_rp = gr.Slider(minimum=0.0, maximum=2.0, value=1.1, step=0.1, label="Repetition penalty")
+            with gr.Accordion("Add models", open=False):
+                chat_add_text = gr.Textbox(label="URL or Repo ID", placeholder="https://huggingface.co/mradermacher/MagnumChronos-i1-GGUF/blob/main/MagnumChronos.i1-Q4_K_M.gguf", lines=1)
+                chat_add_format = gr.Dropdown(choices=get_llm_formats(), value=get_llm_formats()[0], label="Message format")
+                chat_add_submit = gr.Button("Update lists of models")
             with gr.Accordion("Modes", open=True):
                 chat_model = gr.Dropdown(choices=get_dolphin_models(), value=get_dolphin_models()[0][1], allow_custom_value=True, label="Model")
                 chat_model_info = gr.Markdown(value=get_dolphin_model_info(get_dolphin_models()[0][1]), label="Model info")
                 with gr.Row():
                     chat_mode = gr.Dropdown(choices=get_dolphin_sysprompt_mode(), value=get_dolphin_sysprompt_mode()[0], allow_custom_value=False, label="Mode")
                     chat_lang = gr.Dropdown(choices=get_dolphin_languages(), value="English", allow_custom_value=True, label="Output language")
-                with gr.Accordion("Add models", open=False):
-                    chat_add_text = gr.Textbox(label="URL or Repo ID", placeholder="https://huggingface.co/mradermacher/MagnumChronos-i1-GGUF/blob/main/MagnumChronos.i1-Q4_K_M.gguf", lines=1)
-                    chat_add_format = gr.Dropdown(choices=get_llm_formats(), value=get_llm_formats()[0], label="Message format")
-                    chat_add_submit = gr.Button("Update lists of models")
         with gr.Row():
             with gr.Group():
                 output_text = gr.TextArea(label="Output tags", interactive=False, show_copy_button=True)

llmdolphin.py

@@ -1,5 +1,5 @@
-import gradio as gr
 import spaces
+import gradio as gr
 from llama_cpp import Llama
 from llama_cpp_agent import LlamaCppAgent, MessagesFormatterType
 from llama_cpp_agent.providers import LlamaCppPythonProvider
@@ -44,6 +44,15 @@ llm_models = {
     "guns-and-roses-r1-Q4_K_L-imat.gguf": ["Reiterate3680/guns-and-roses-r1-GGUF", MessagesFormatterType.MISTRAL],
     "MagnumChronos.Q4_K_M.gguf": ["mradermacher/MagnumChronos-GGUF", MessagesFormatterType.CHATML],
     "Trinas_Nectar-8B-model_stock.i1-Q4_K_M.gguf": ["mradermacher/Trinas_Nectar-8B-model_stock-i1-GGUF", MessagesFormatterType.MISTRAL],
+    "ChatWaifu_v1.3.1.Q4_K_M.gguf": ["mradermacher/ChatWaifu_v1.3.1-GGUF", MessagesFormatterType.MISTRAL],
+    "kainaticulous-rp-7b.Q5_K_M.gguf": ["mradermacher/kainaticulous-rp-7b-GGUF", MessagesFormatterType.MISTRAL],
+    "OpenCrystal-12B-L3.1-128K.Q4_K_S.gguf": ["mradermacher/OpenCrystal-12B-L3.1-128K-GGUF", MessagesFormatterType.MISTRAL],
+    "L3.1-12B-Niitama-v1.1.Q4_K_S.gguf": ["mradermacher/L3.1-12B-Niitama-v1.1-GGUF", MessagesFormatterType.MISTRAL],
+    "L3.1-12B-Celeste-V1.5.Q4_K_S.gguf": ["mradermacher/L3.1-12B-Celeste-V1.5-GGUF", MessagesFormatterType.MISTRAL],
+    "L3.1-Sthenorm-8B.Q4_K_S.gguf": ["mradermacher/L3.1-Sthenorm-8B-GGUF", MessagesFormatterType.MISTRAL],
+    "MN-12B-Estrella-v2.2.Q4_K_M.gguf": ["mradermacher/MN-12B-Estrella-v2.2-GGUF", MessagesFormatterType.MISTRAL],
+    "WoonaV1.2-9b.Q4_K_M.gguf": ["QuantFactory/WoonaV1.2-9b-GGUF", MessagesFormatterType.MISTRAL],
+    "kainaticulous-rp-7B-Q5_K_M.gguf": ["kainatq/kainaticulous-rp-7b-gguf", MessagesFormatterType.MISTRAL],
     "experiment-wip-q4_k_m.gguf": ["DreadPoor/EXPERIMENT-WIP-Q4_K_M-GGUF", MessagesFormatterType.CHATML],
     "eye_of_the_storm-q4_k_m.gguf": ["Daemontatox/Eye_of_the_Storm-Q4_K_M-GGUF", MessagesFormatterType.MISTRAL],
     "Fireball-3.1-8B-ORPO.i1-Q5_K_M.gguf": ["mradermacher/Fireball-3.1-8B-ORPO-i1-GGUF", MessagesFormatterType.LLAMA_3],
@@ -348,7 +357,7 @@ llm_formats = {
     "Qwen2": MessagesFormatterType.OPEN_CHAT,
 }
 # https://github.com/Maximilian-Winter/llama-cpp-agent
-llm_languages = ["English", "Japanese", "Chinese"]
+llm_languages = ["English", "Japanese", "Chinese", "Korean", "Spanish", "Portuguese", "German", "French", "Finnish", "Russian"]
 llm_models_tupled_list = []
 default_llm_model_filename = list(llm_models.keys())[0]
 override_llm_format = None
@@ -414,7 +423,8 @@ def download_llm_model(filename):
     if not filename in llm_models.keys(): return default_llm_model_filename
     try:
         hf_hub_download(repo_id = llm_models[filename][0], filename = filename, local_dir = llm_models_dir)
-    except Exception:
+    except Exception as e:
+        print(e)
         return default_llm_model_filename
     update_llm_model_tupled_list()
     return filename
@@ -444,7 +454,6 @@ def select_dolphin_format(format_name):
     return gr.update(value=format_name)
 
 
-#download_llm_models()
 download_llm_model(default_llm_model_filename)
 
 
@@ -494,11 +503,14 @@ def add_dolphin_models(query, format_name):
             if not api.repo_exists(repo_id = repo) or not api.file_exists(repo_id = repo, filename = filename): return gr.update(visible=True)
             add_models[filename] = [repo, format]
         else: return gr.update(visible=True)
-    except Exception:
+    except Exception as e:
+        print(e)
         return gr.update(visible=True)
-    print(add_models)
+    #print(add_models)
     llm_models = (llm_models | add_models).copy()
-    return gr.update(choices=get_dolphin_models())
+    update_llm_model_tupled_list()
+    choices = get_dolphin_models()
+    return gr.update(choices=choices, value=choices[-1])
 
 
 dolphin_output_language = "English"