upgraded new configurations

configs/config.yaml

@@ -1,5 +1,5 @@
 # =================== Lord Of Large Language Multimodal Systems Configuration file =========================== 
-version: 81
+version: 82
 binding_name: null
 model_name: null
 model_variant: null
@@ -109,6 +109,10 @@ petals_device: cuda
 # lollms service
 enable_lollms_service: false
 lollms_base_url: http://localhost:1234
+activate_lollms_server: True
+activate_ollama_emulator: True
+activate_openai_emulator: True
+activate_mistralai_emulator: True
 
 # elastic search service
 elastic_search_service: false

lollms/configs/config.yaml

@@ -1,5 +1,5 @@
 # =================== Lord Of Large Language Multimodal Systems Configuration file =========================== 
-version: 81
+version: 82
 binding_name: null
 model_name: null
 model_variant: null
@@ -109,6 +109,10 @@ petals_device: cuda
 # lollms service
 enable_lollms_service: false
 lollms_base_url: http://localhost:1234
+activate_lollms_server: True
+activate_ollama_emulator: True
+activate_openai_emulator: True
+activate_mistralai_emulator: True
 
 # elastic search service
 elastic_search_service: false

lollms/server/configs/config.yaml

@@ -1,5 +1,5 @@
 # =================== Lord Of Large Language Multimodal Systems Configuration file =========================== 
-version: 81
+version: 82
 binding_name: null
 model_name: null
 model_variant: null
@@ -109,6 +109,10 @@ petals_device: cuda
 # lollms service
 enable_lollms_service: false
 lollms_base_url: http://localhost:1234
+activate_lollms_server: True
+activate_ollama_emulator: True
+activate_openai_emulator: True
+activate_mistralai_emulator: True
 
 # elastic search service
 elastic_search_service: false

lollms/server/endpoints/lollms_generator.py

@@ -414,6 +414,73 @@ class CompletionGenerationRequest(BaseModel):
     temperature: Optional[float] = -1
 
 
+@router.post("/instruct/generate")
+async def ollama_completion(request: CompletionGenerationRequest):
+    """
+    Executes Python code and returns the output.
+
+    :param request: The HTTP request object.
+    :return: A JSON response with the status of the operation.
+    """
+    try:
+        text = request.prompt
+        n_predict = request.max_tokens if request.max_tokens>=0 else elf_server.config.max_n_predict
+        temperature = request.temperature if request.temperature>=0 else elf_server.config.temperature
+        # top_k = request.top_k if request.top_k>=0 else elf_server.config.top_k
+        # top_p = request.top_p if request.top_p>=0 else elf_server.config.top_p
+        # repeat_last_n = request.repeat_last_n if request.repeat_last_n>=0 else elf_server.config.repeat_last_n
+        # repeat_penalty = request.repeat_penalty if request.repeat_penalty>=0 else elf_server.config.repeat_penalty
+        stream = request.stream
+        
+        if elf_server.binding is not None:
+            if stream:
+                output = {"response":""}
+                def generate_chunks():
+                    def callback(chunk, chunk_type:MSG_TYPE=MSG_TYPE.MSG_TYPE_CHUNK):
+                        # Yield each chunk of data
+                        output["response"] += chunk
+                        antiprompt = detect_antiprompt(output["text"])
+                        if antiprompt:
+                            ASCIIColors.warning(f"\n{antiprompt} detected. Stopping generation")
+                            output["response"] = remove_text_from_string(output["response"],antiprompt)
+                            return False
+                        else:
+                            yield {"response":chunk}
+                            return True
+                    return iter(elf_server.binding.generate(
+                                                text, 
+                                                n_predict, 
+                                                callback=callback, 
+                                                temperature=temperature,
+                            ))
+                
+                return StreamingResponse(generate_chunks())
+            else:
+                output = {"response":""}
+                def callback(chunk, chunk_type:MSG_TYPE=MSG_TYPE.MSG_TYPE_CHUNK):
+                    # Yield each chunk of data
+                    output["response"] += chunk
+                    antiprompt = detect_antiprompt(output["response"])
+                    if antiprompt:
+                        ASCIIColors.warning(f"\n{antiprompt} detected. Stopping generation")
+                        output["response"] = remove_text_from_string(output["response"],antiprompt)
+                        return False
+                    else:
+                        return True
+                elf_server.binding.generate(
+                                                text, 
+                                                n_predict, 
+                                                callback=callback,
+                                                temperature=request.temperature if request.temperature>=0 else elf_server.config.temperature
+                                            )
+                return output
+        else:
+            return None
+    except Exception as ex:
+        trace_exception(ex)
+        elf_server.error(ex)
+        return {"status":False,"error":str(ex)}
+
 @router.post("/v1/completions")
 async def v1_completion(request: CompletionGenerationRequest):
     """
@@ -471,7 +538,7 @@ async def v1_completion(request: CompletionGenerationRequest):
                                                 text, 
                                                 n_predict, 
                                                 callback=callback,
-                                                temperature=data.get("temperature", elf_server.config.temperature)
+                                                temperature=request.temperature if request.temperature>=0 else elf_server.config.temperature
                                             )
                 return output["text"]
         else:

lollms/server/endpoints/lollms_models_infos.py

@@ -99,3 +99,44 @@ def add_reference_to_local_model(data:ModelReferenceParams):
     else:        
         return {"status": False, "error":"Model not found"}       
 
+
+
+
+
+@router.get("/api/tags")
+async def ollama_list_models():
+    """
+    Retrieve a list of available models for the currently selected binding.
+
+    Returns:
+        List[str]: A list of model names.
+    """
+    if lollmsElfServer.binding is None:
+        return []
+    try:
+        model_list = lollmsElfServer.binding.get_available_models(lollmsElfServer)
+
+        md = {
+        "models": [
+            {
+            "name": model,
+            "modified_at": "2023-11-04T14:56:49.277302595-07:00",
+            "size": 7365960935,
+            "digest": "9f438cb9cd581fc025612d27f7c1a6669ff83a8bb0ed86c94fcf4c5440555697",
+            "details": {
+                "format": "gguf",
+                "family": "llama",
+                "families": None,
+                "parameter_size": "13B",
+                "quantization_level": "Q4_0"
+            }
+            }
+            for model in model_list
+        ]
+        }
+    except Exception as ex:
+        trace_exception(ex)
+        lollmsElfServer.error("Coudln't list models. Please reinstall the binding or notify ParisNeo on the discord server")
+        return []
+
+    return md