phi3.5-phunction-calling Model Card

Model Overview

Model Name: phi3.5-phunction-calling

Description: This model is a fine-tuned version of the phi3.5 model, specifically designed for function calling tasks. It has been optimized to understand and execute function calls accurately and efficiently.

Intended Use

Primary Use Case: This model is intended for use in applications where function calling is a critical component, such as automated assistants, code generation, and API interaction.

Limitations: While the model is highly accurate, it may still produce errors or misunderstandings in complex or ambiguous function calls.

from unsloth.chat_templates import get_chat_template

tokenizer = get_chat_template(
    tokenizer,
    chat_template = "phi-3", # Supports zephyr, chatml, mistral, llama, alpaca, vicuna, vicuna_old, unsloth
    mapping = {"role" : "from", "content" : "value", "user" : "human", "assistant" : "gpt"}, # ShareGPT style
)

FastLanguageModel.for_inference(model) # Enable native 2x faster inference

messages = [
    {"from": "system", "value": "You are a function calling AI model. You are provided with function signatures within <tools> </tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n<tools>\n[{'type': 'function', 'function': {'name': 'search_recipes', 'description': 'Searches for recipes based on given criteria.', 'parameters': {'type': 'object', 'properties': {'cuisine': {'type': 'string', 'description': 'The type of cuisine to search for.'}, 'dietary_restriction': {'type': 'string', 'description': 'Any dietary restrictions to consider.', 'enum': ['vegetarian', 'vegan', 'gluten-free', 'none']}}, 'required': ['cuisine']}}}, {'type': 'function', 'function': {'name': 'get_recipe_details', 'description': 'Retrieves detailed information about a specific recipe.', 'parameters': {'type': 'object', 'properties': {'recipe_id': {'type': 'string', 'description': 'The unique identifier for the recipe.'}}, 'required': ['recipe_id']}}}, {'type': 'function', 'function': {'name': 'calculate_nutrition', 'description': 'Calculates nutritional information for a given recipe.', 'parameters': {'type': 'object', 'properties': {'recipe_id': {'type': 'string', 'description': 'The unique identifier for the recipe.'}, 'serving_size': {'type': 'integer', 'description': 'The number of servings to calculate nutrition for.', 'default': 1}}, 'required': ['recipe_id']}}}]\n</tools>\nFor each function call return a json object with function name and arguments within <tool_call> </tool_call> tags with the following schema:\n<tool_call>\n{'arguments': <args-dict>, 'name': <function-name>}\n</tool_call>\n"},
    {"from": "human", "value": "I'm planning a dinner party and I'm looking for some Italian recipes to try. Can you help me find some vegetarian Italian dishes? Once we have a list, I'd like to get more details about the first recipe in the search results. Finally, I want to calculate the nutritional information for that recipe, assuming I'm cooking for 4 people. Can you please perform these tasks for me?"},
]

inputs = tokenizer.apply_chat_template(
    messages,
    tokenize = True,
    add_generation_prompt = True, # Must add for generation
    return_tensors = "pt",
).to("cuda")

outputs = model.generate(input_ids = inputs, max_new_tokens = 256, use_cache = True)
tokenizer.batch_decode(outputs)
Downloads last month
29
Safetensors
Model size
3.82B params
Tensor type
BF16
·
Inference API
Unable to determine this model's library. Check the docs .

Model tree for ayan-sh003/phi3.5-phunction-calling

Finetuned
(50)
this model

Dataset used to train ayan-sh003/phi3.5-phunction-calling