enhanced backends

.gitignore

@@ -165,4 +165,5 @@ web/.env.dev
 web/.env.development
 node_modules/
 
+# Google chrome files
 *.crdownload

backends/c_transformers/__init__.py

@@ -57,7 +57,31 @@ class GPTJ(GPTBackend):
             self.model = AutoModelForCausalLM.from_pretrained(
                     f"./models/c_transformers/{self.config['model']}", model_type=model_type, lib = "avx"
                     )
+            
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
 
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return self.model.tokenize(prompt.encode())
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return self.model.detokenize(tokens_list)
+    
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,

backends/gpt_4all/__init__.py

@@ -40,6 +40,30 @@ class GPT4ALL(GPTBackend):
                 model_path=f"./models/gpt_4all/{self.config['model']}",
         )
 
+
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
+
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return None
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return None
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,
@@ -63,6 +87,7 @@ class GPT4ALL(GPTBackend):
                                             repeat_penalty=self.config['repeat_penalty'],
                                             repeat_last_n = self.config['repeat_last_n'],
                                             # n_threads=self.config['n_threads'],
+                                            streaming=True
                                            ):
                 if not new_text_callback(tok):
                     return

backends/gpt_4all/requirements.txt

@@ -0,0 +1 @@
+gpt4all

backends/gpt_j_a/__init__.py

@@ -38,7 +38,29 @@ class GptJ(GPTBackend):
                 model_path=f"./models/gpt_j/{self.config['model']}",
                 prompt_context="", prompt_prefix="", prompt_suffix=""
                 )
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
 
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return None
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return None
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,

backends/gpt_j_m/__init__.py

@@ -38,7 +38,29 @@ class GPTJ(GPTBackend):
         self.model = Model(
                 model=f"./models/llama_cpp/{self.config['model']}", avx2 = self.config["use_avx2"]
                 )
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
 
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return None
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return None
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,

backends/gptq/__init__.py

@@ -37,7 +37,29 @@ class GPTQ(GPTBackend):
         # load quantized model, currently only support cpu or single gpu
         self.model = AutoGPTQForCausalLM.from_pretrained(self.model_dir, BaseQuantizeConfig())
         self.tokenizer = AutoTokenizer.from_pretrained(self.model_dir, use_fast=True    )
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
 
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return None
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return None
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,

backends/llama_cpp_official/__init__.py

@@ -40,6 +40,31 @@ class LLAMACPP(GPTBackend):
             
         self.model = Llama(model_path=f"./models/llama_cpp_official/{self.config['model']}", n_gpu_layers=40, seed=seed)
 
+
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
+
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return self.model.tokenize(prompt.encode())
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return self.model.detokenize(tokens_list).decode()
+
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,
@@ -82,4 +107,6 @@ class LLAMACPP(GPTBackend):
         with open(file_path, 'r') as file:
             yaml_data = yaml.safe_load(file)
         
-        return yaml_data
+        return yaml_data
+    
+

backends/py_llama_cpp/__init__.py

@@ -40,7 +40,30 @@ class PyLLAMACPP(GPTBackend):
                 n_ctx=self.config['ctx_size'], 
                 seed=self.config['seed'],
                 )
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
 
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        return self.model.tokenize(prompt)
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        return self.model.detokenize(tokens_list)
+    
     def generate(self, 
                  prompt:str,                  
                  n_predict: int = 128,

backends/py_llama_cpp/requirements.txt

@@ -1 +1 @@
-pyllamacpp
+pyllamacpp>=2.1.3

gpt4all_api/backend.py

@@ -43,6 +43,30 @@ class GPTBackend:
             verbose (bool, optional): If true, the code will spit many informations about the generation process. Defaults to False.
         """
         pass
+    def tokenize(self, prompt):
+        """
+        Tokenizes the given prompt using the model's tokenizer.
+
+        Args:
+            prompt (str): The input prompt to be tokenized.
+
+        Returns:
+            list: A list of tokens representing the tokenized prompt.
+        """
+        pass
+
+    def detokenize(self, tokens_list):
+        """
+        Detokenizes the given list of tokens using the model's tokenizer.
+
+        Args:
+            tokens_list (list): A list of tokens to be detokenized.
+
+        Returns:
+            str: The detokenized text as a string.
+        """
+        pass
+
     @staticmethod
     def list_models(config:dict):
         """Lists the models for this backend
@@ -58,4 +82,5 @@ class GPTBackend:
         with open(file_path, 'r') as file:
             yaml_data = yaml.safe_load(file)
         
-        return yaml_data
+        return yaml_data
+

web/src/views/SettingsView.vue

@@ -58,7 +58,7 @@
                     <label for="model" class="block mb-2 text-sm font-medium text-gray-900 dark:text-white">
                         Model:
                     </label>
-                    <select id="model" @change="update_model($event.target.value)"
+                    <select id="model" @change="update_model($event.target.value)" @mouseup="update_model($event.target.value)" 
                         class="bg-gray-50 border border-gray-300 text-gray-900 text-sm rounded-lg focus:ring-blue-500 focus:border-blue-500 block w-full p-2.5 dark:bg-gray-700 dark:border-gray-600 dark:placeholder-gray-400 dark:text-white dark:focus:ring-blue-500 dark:focus:border-blue-500">
 
                         <option v-for="item in modelsArr" :selected="item === configFile.model">{{ item }}</option>