agent: nits

examples/agent/fastify-requirements.txt

@@ -2,4 +2,4 @@ fastapi[all]
 pydantic
 sse-starlette
 uvicorn[all]
-typer[all]
+typer[all]

examples/agent/requirements.txt

@@ -5,4 +5,4 @@ pydantic
 requests
 sse-starlette
 uvicorn[all]
-typer[all]
+typer[all]

examples/agent/run_sandboxed_tools.sh

@@ -71,4 +71,4 @@ docker run \
     --mount "type=bind,source=$( realpath "$script_folder" ),target=/src,readonly" \
     --mount "type=bind,source=$( realpath "$DATA_DIR" ),target=/data" \
     -p "$PORT:$PORT" \
-    -it "$LLAMA_IMAGE_NAME"
+    -it "$LLAMA_IMAGE_NAME"

examples/openai/api.py

@@ -87,4 +87,4 @@ class ChatCompletionResponse(BaseModel):
     choices: list[Choice]
     usage: Usage
     system_fingerprint: str
-    error: Optional[CompletionError] = None
+    error: Optional[CompletionError] = None

examples/openai/llama_cpp_server_api.py

@@ -9,4 +9,4 @@ class LlamaCppServerCompletionRequest(LlamaCppParams):
     cache_prompt: Optional[bool] = None
 
     grammar: Optional[str] = None
-    json_schema: Optional[Json] = None
+    json_schema: Optional[Json] = None

examples/openai/prompt1.txt

@@ -1,43 +0,0 @@
-<|im_start|>system
-Role:
-  You are a function calling AI agent with self-recursion.
-  You can call only one function at a time and analyse data you get from function response.
-  You are provided with function signatures within <tools></tools> XML tags.
-  The current date is: March 25, 2024.
-
-Objective:
-  You may use agentic frameworks for reasoning and planning to help with user query.
-  Please call a function and wait for function results to be provided to you in the next iteration.
-  Don't make assumptions about what values to plug into function arguments.
-  Once you have called a function, results will be fed back to you within <tool_response></tool_response> XML tags.
-  Don't make assumptions about tool results if <tool_response> XML tags are not present since function hasn't been executed yet.
-  Analyze the data once you get the results and call another function.
-  At each iteration please continue adding the your analysis to previous summary.
-  Your final response should directly answer the user query with an anlysis or summary of the results of function calls.
-
-Tools:
-  Here are the available tools:
-  <tools> 
-  {"type":"function","function":{"name":"get_current_weather","description":"Get the current weather","parameters":{"type":"object","properties":{"location":{"type":"string","description":"The city and state, e.g. San Francisco, CA"},"format":{"type":"string","enum":["celsius","fahrenheit"],"description":"The temperature unit to use. Infer this from the users location."}},"required":["location","format"]}}}
-  {"type":"function","function":{"name":"get_n_day_weather_forecast","description":"Get an N-day weather forecast","parameters":{"type":"object","properties":{"location":{"type":"string","description":"The city and state, e.g. San Francisco, CA"},"format":{"type":"string","enum":["celsius","fahrenheit"],"description":"The temperature unit to use. Infer this from the users location."},"num_days":{"type":"integer","description":"The number of days to forecast"}},"required":["location","format","num_days"]}}} 
-  </tools>
-  If the provided function signatures doesn't have the function you must call, you may write executable python code in markdown syntax and call code_interpreter() function as follows:
-  <tool_call>
-  {"arguments": {"code_markdown": <python-code>, "name": "code_interpreter"}}
-  </tool_call>
-  Make sure that the json object above with code markdown block is parseable with json.loads() and the XML block with XML ElementTree.
-
-Instructions:
-  At the very first turn you don't have <tool_results> so you shouldn't not make up the results.
-  Please keep a running summary with analysis of previous function results and summaries from previous iterations.
-  Do not stop calling functions until the task has been accomplished or you've reached max iteration of 10.
-  Calling multiple functions at once can overload the system and increase cost so call one function at a time please.
-  If you plan to continue with analysis, always call another function.
-  For each function call return a valid json object (using doulbe quotes) with function name and arguments within <tool_call></tool_call> XML tags as follows:
-  <tool_call>
-  {"arguments": <args-dict>, "name": <function-name>}
-  </tool_call>
-<|im_end|>
-<|im_start|>user
-what is the weather going to be like in San Francisco and Glasgow over the next 4 days (temperature in celsius for both)<|im_end|>
-<|im_start|>assistant

examples/openai/requirements.txt

@@ -4,4 +4,4 @@ jinja2
 pydantic
 sse-starlette
 uvicorn[all]
-typer[all]
+typer[all]

examples/openai/server.py

@@ -139,7 +139,7 @@ def main(
         messages = chat_request.messages
 
         prompt = chat_handler.render_prompt(messages)
-        
+
         if verbose:
             sys.stderr.write(f'\n# REQUEST:\n\n{chat_request.model_dump_json(indent=2)}\n\n')
             # sys.stderr.write(f'\n# MESSAGES:\n\n{TypeAdapter(list[Message]).dump_json(messages)}\n\n')

examples/openai/test_chat_handlers.py

@@ -202,14 +202,14 @@ if __name__ == "__main__":
                 ch = get_chat_handler(args, parallel_calls=True, tool_style=style)
 
                 print(f'\n### {model_name} / {style.name} / {tool_situation}\n')
-                
+
                 print(f'\nPrompt:\n\n```js\n{ch.render_prompt(TEST_MESSAGES_THOUGHT)}\n```\n')
 
                 print(f'\nOutput format prompt:\n\n```json\n{ch.output_format_prompt.content}\n```\n')
 
                 print(f'\nGrammar:\n\n```js\n{ch.grammar}\n```\n')
 
-                        
+
                 # if model_name == 'hermes_2_pro_mistral':
                 #     print("Skipping hermes_2_pro_mistral")
                 #     continue
@@ -233,38 +233,3 @@ if __name__ == "__main__":
             print(f'{f}\n\n')
 
         assert not failures
-    # test_templates([
-    #     Message(**{
-    #         "role": "user",
-    #         "name": None,
-    #         "tool_call_id": None,
-    #         "content": "What is the sum of 2535 squared and 32222000403 then multiplied by one and a half. What's a third of the result?",
-    #         "tool_calls": None
-    #     }),
-    #     Message(**{
-    #         "role": "assistant",
-    #         # "name": None,
-    #         "tool_call_id": None,
-    #         "content": "?",
-    #         "tool_calls": [
-    #             {
-    #                 # "id": "call_531873",
-    #                 "type": "function",
-    #                 "function": {
-    #                     "name": TOOL_NAME,
-    #                     "arguments": {
-    #                         "a": 2535,
-    #                         "b": 32222000403
-    #                     }
-    #                 }
-    #             }
-    #         ]
-    #     }),
-    #     Message(**{
-    #         "role": "tool",
-    #         "name": TOOL_NAME,
-    #         "tool_call_id": "call_531873",
-    #         "content": "32222002938",
-    #         "tool_calls": None
-    #     })
-    # ])

gguf-py/examples/reader.py

@@ -3,7 +3,6 @@ import sys
 from pathlib import Path
 from gguf.gguf_reader import GGUFReader
 
-
 sys.path.insert(0, str(Path(__file__).parent.parent))
 
 

gguf-py/gguf/gguf_reader.py

@@ -50,6 +50,7 @@ class ReaderField(NamedTuple):
     types: list[GGUFValueType] = []
 
 
+
 class ReaderTensor(NamedTuple):
     name: str
     tensor_type: GGMLQuantizationType