metadata

language:
  - en
  - cn
pipeline_tag: text-generation
inference: false
license: other
license_name: deepseek
license_link: LICENSE
tags:
  - deepseek
  - deepseek coder
  - pytorch
  - functions
  - function calling
  - sharded
extra_gated_prompt: >-
  Access to this model requires the purchase of a license
  [here](https://buy.stripe.com/9AQbJubSda9Z8EM00A)
extra_gated_fields:
  Name: text
  Affiliation: text
  Email: text
  I agree to the terms of the license described on the model card: checkbox
  I have purchased a license (see repo for payment links): checkbox

Function Calling Deepseek Coder Instruct

Extends the model with function calling capabilities.
The model responds with a structured json argument with the function name and arguments.

Recent Updates

November 6th 2023 -> added Deepseek Coder 1.3B, 6.7B and 33B
October 11th 2023 -> added Mistral 7B with function calling
October 11th 2023 -> new models pushed, trained on an improved underlying dataset

Improvements with v2

Shortened syntax: Only function descriptions are needed for inference and no added instruction is required.
Function descriptions are moved outside of the system prompt. This avoids the behaviour of function calling being affected by how the system prompt had been trained to influence the model.

Performance and Tips

Larger models are better at handling function calling. The cross entropy training losses are approximately 0.5 for 7B, 0.4 for 13B, 0.3 for 70B. The absolute numbers don't mean anything but the relative values offer a sense of relative performance.
Provide very clear function descriptions, including whether the arguments are required or what the default values should be.
Make sure to post-process the language model's response to check that all necessary information is provided by the user. If not, prompt the user to let them know they need to provide more info (e.g. their name, order number etc.)

Check out this video overview of performance here

Licensing

Llama-7B with function calling is licensed according to the Meta Community license.

Mistral-7B, Llama-13B, DeepSeek models, Code-llama-34b, Llama-70B and Falcon-180B with function calling require the purchase of access.

Commercial license purchase required per user.
Licenses are not transferable to other users/entities.

Use of all Llama models with function calling is further subject to terms in the Meta license. Deepseek coder models are subject to the Deepseek license.

Dataset

The dataset used for training this model can be found at Trelis Function Calling Extended Dataset.

Inference

!!! Make sure to check the prompt format below and adjust inference accordingly !!!

Quick Start in Google Colab Try out this notebook fLlama_Inference notebook

Commercial Applications You can this model with text-generation-interface and chat-ui

Here is the github for setup

And here is a video showing it working with llama-2-7b-chat-hf-function-calling-v2 (note that we've now moved to v2)

Note that you'll still need to code the server-side handling of making the function calls (which obviously depends on what functions you want to use).

Run on your laptop Run on your laptop video and juypter notebook

After running llama.cpp server, you can call the server with this command, with thanks to @jdo300:

import requests
import json

# Define the roles and markers
B_FUNC, E_FUNC = "<FUNCTIONS>", "</FUNCTIONS>\n\n"
B_INST, E_INST = "[INST] ", " [/INST]" #Llama style
# B_INST, E_INST = "\n### Instruction:\n", "\n### Response:\n" #DeepSeek Coder Style

# Define the function metadata
function_metadata = {
    "function": "search_bing",
    "description": "Search the web for content on Bing. This allows users to search online/the internet/the web for content.",
    "arguments": [
        {
            "name": "query",
            "type": "string",
            "description": "The search query string"
        }
    ]
}

# Define the user prompt
user_prompt = 'Search for the latest news on AI.'

# Format the function list and prompt
function_list = json.dumps(function_metadata, indent=4)
prompt = f"{B_FUNC}{function_list.strip()}{E_FUNC}{B_INST}{user_prompt.strip()}{E_INST}\n\n"

# Define the API endpoint
url = "http:/localhost:8080/completion"

# Send the POST request to the API server
response = requests.post(url, json={"prompt": prompt})

# Print the response
print(response.json())

Syntax

Prompt Templates

The function descriptions must be wrapped within a function block. You can put this function below before or after the system message block.

Example without a system message:

  # Define the roles and markers
  B_FUNC, E_FUNC = "<FUNCTIONS>", "</FUNCTIONS>\n\n"
  B_INST, E_INST = "[INST] ", " [/INST]" #Llama style
  # B_INST, E_INST = "\n### Instruction:\n", "\n### Response:\n" #DeepSeek Coder Style

  functionList = {function_1_metadata}{function_2_metadata}...
  user_prompt = '...'

  # Format your prompt template
  prompt = f"{B_FUNC}{functionList.strip()}{E_FUNC}{B_INST}{user_prompt.strip()}{E_INST}\n\n"

Example with a system message:

  # Define the roles and markers
  B_FUNC, E_FUNC = "<FUNCTIONS>", "</FUNCTIONS>\n\n"
  B_INST, E_INST = "[INST] ", " [/INST]" #Llama style
  # B_INST, E_INST = "\n### Instruction:\n", "\n### Response:\n" #DeepSeek Coder Style
  B_SYS, E_SYS = "<<SYS>>\n", "\n<</SYS>>\n\n"

  # assuming functionList is defined as above
  system_prompt = '...'
  user_prompt = '...'

  # Format your prompt template
  prompt = f"{B_FUNC}{functionList.strip()}{E_FUNC}{B_INST}{B_SYS}{system_prompt.strip()}{E_SYS}{user_prompt.strip()}{E_INST}\n\n"

Notice that the function block is placed at the very start of the sequence, before 'B_INST'.

Function Metadata Template

functionMetadata should be a string representation of a JSON object, like this:

"functionMetadata": {
        "function": "search_bing",
        "description": "Search the web for content on Bing. This allows users to search online/the internet/the web for content.",
        "arguments": [
            {
                "name": "query",
                "type": "string",
                "description": "The search query string"
            }
        ]
    }
'''

and the language model should respond with a json object formatted like this:

{
    "function": "function_name",
    "arguments": {
        "argument1": "argument_value",
        "argument2": "argument_value"
    }
}

It is recommended to handle cases where:

There is no json object in the response
The response contains text in addition to the json response

Sample functionList

{
    "function": "search_bing",
    "description": "Search the web for content on Bing. This allows users to search online/the internet/the web for content.",
    "arguments": [
        {
            "name": "query",
            "type": "string",
            "description": "The search query string"
        }
    ]
}

{
    "function": "search_arxiv",
    "description": "Search for research papers on ArXiv. Make use of AND, OR and NOT operators as appropriate to join terms within the query.",
    "arguments": [
        {
            "name": "query",
            "type": "string",
            "description": "The search query string"
        }
    ]
}

Training Set Argument Types

Models were fine-tuned on argument types including strings, numbers and arrays. The training set includes function calls with 0, 1, 2 or 3 arguments. The larger the model the better it will generalise beyond these types.

Here is a function call with an array:

{ "function": "delete_file", "arguments": { "fileNames": [ "Dissecting Transformer Length Extrapolation via The Lens of Receptive Field Analysis", "Luna- Linear Unified Nested Attention", "Substack_Inc_2021_2020_GAAP_Audited_Financials" ] } }

Here is a function call with three arguments:

{ "function": "save_chat", "arguments": { "fileName": "KiteDiscussion", "fileDescription": "Notes on one and two stringed kites", "fileContent": "--- **Types of Kite** There are one and two string kites. The two string ones are easier to control, although you can get the cords tangled. The one-stringed ones are sometimes used for kite fights, and you lose the kite and have to run after it if the string breaks. ---" } }

Below follows information on the original Deepseek coder model card...

1. Introduction of Deepseek Coder

Deepseek Coder is composed of a series of code language models, each trained from scratch on 2T tokens, with a composition of 87% code and 13% natural language in both English and Chinese. We provide various sizes of the code model, ranging from 1B to 33B versions. Each model is pre-trained on project-level code corpus by employing a window size of 16K and a extra fill-in-the-blank task, to support project-level code completion and infilling. For coding capabilities, Deepseek Coder achieves state-of-the-art performance among open-source code models on multiple programming languages and various benchmarks.

Massive Training Data: Trained from scratch on 2T tokens, including 87% code and 13% linguistic data in both English and Chinese languages.
Highly Flexible & Scalable: Offered in model sizes of 1.3B, 5.7B, 6.7B, and 33B, enabling users to choose the setup most suitable for their requirements.
Superior Model Performance: State-of-the-art performance among publicly available code models on HumanEval, MultiPL-E, MBPP, DS-1000, and APPS benchmarks.
Advanced Code Completion Capabilities: A window size of 16K and a fill-in-the-blank task, supporting project-level code completion and infilling tasks.

2. Model Summary

deepseek-coder-33b-instruct is a 33B parameter model initialized from deepseek-coder-33b-base and fine-tuned on 2B tokens of instruction data.

Home Page: DeepSeek
Repository: deepseek-ai/deepseek-coder
Chat With DeepSeek Coder: DeepSeek-Coder

3. How to Use

Here give some examples of how to use our model.

Chat Model Inference

from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/deepseek-coder-33b-instruct", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/deepseek-coder-33b-instruct", trust_remote_code=True).cuda()
messages=[
    { 'role': 'user', 'content': "write a quick sort algorithm in python."}
]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
# 32021 is the id of <|EOT|> token
outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, top_k=50, top_p=0.95, num_return_sequences=1, eos_token_id=32021)
print(tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True))

4. License

This code repository is licensed under the MIT License. The use of DeepSeek Coder models is subject to the Model License. DeepSeek Coder supports commercial use.

See the LICENSE-MODEL for more details.

5. Contact

If you have any questions, please raise an issue or contact us at [email protected].