README.md

---
pipeline_tag: text-generation
license: apache-2.0
tags:
- text generation
- Deci AI
- DeciCoder
programming_language:
  - Java
  - JavaScript
  - Python
  - Rust
  - Go
  - C++
  - C
  - C#
metrics:
- code_eval
inference: true
widget:
- text: 'def print_hello_world():'
  example_title: Hello world
  group: Python
model-index:
- name: DeciCoder-6b
  results:
  - task:
      type: text-generation
    dataset:
      type: nuprl/MultiPL-E
      name: MultiPL-HumanEval (Python)
    metrics:
    - name: pass@1
      type: pass@1
      value: 0.34
      verified: false
  - task:
      type: text-generation
    dataset:
      type: nuprl/MultiPL-E
      name: MultiPL-HumanEval (JavaScript)
    metrics:
    - name: pass@1
      type: pass@1
      value: 0.29
      verified: false
  - task:
      type: text-generation
    dataset:
      type: nuprl/MultiPL-E
      name: MultiPL-HumanEval (Java)
    metrics:
    - name: pass@1
      type: pass@1
      value: 0.30
      verified: false
datasets:
- bigcode/starcoderdata
---

# Model Card for DeciCoder 6B

DeciCoder 6B is a 6 billion parameter decoder-only code completion model
trained on the Python, Java, Javascript, Go, Rust, C++, C, and C# subset of [Starcoder Training Dataset](https://huggingface.co/datasets/bigcode/starcoderdata)..
The model uses variable Grouped Query Attention and has a context window of 4096
tokens. It was trained using a Fill-in-the-Middle training objective. The model's
architecture was generated by Deci's proprietary Neural Architecture
Search-based technology, AutoNAC.

## Model Details

- **Developed by:** Deci 
- **Model type:** DeciCoder is an auto-regressive language model based on the transformer decoder architecture, using variable Grouped Query Attention.
- **Language(s):** Python, Java, JavaScript, Go, Rust, C++, C, C#
- **License:** Model checkpoints are licensed under the [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)

## Model Architecture

| Parameters | Layers | Heads  | Sequence Length  | GQA num_key_value_heads  | Hidden Size  |
|:----------|:----------|:----------|:----------|:----------|:----------|
| 6B    | 32    | 32    | 4096   | Variable  | 4096 |  |


- **Decoder layer:** Variable Grouped Query Attention. Grouped Query Attention was introduced in [Ainslie et al., 2023](https://arxiv.org/abs/2305.13245)
- **Position Embeddings:** Rotary Position Embeddings [Su et al., 2021](https://arxiv.org/abs/2104.09864)

## Uses

The model is intended to do single/multiline code completion from a
context window of up to 4096k tokens. It is *not* an instruction model
and commands like \"Write a function that computes the absolute value of
an integer,\" won't yield the desired results. A more effective approach
is to frame instructions in the style of source code comments (e.g. \#
this function calculates the absolute value of an integer) or to present
a function signature and docstring, enabling the model to complete the
function's body.

### How to Use

```bibtex
# pip install -q transformers
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

checkpoint = "Deci/DeciCoder-6b"
device = "cuda" # for GPU usage or "cpu" for CPU usage

tokenizer = AutoTokenizer.from_pretrained(checkpoint)
model = AutoModelForCausalLM.from_pretrained(checkpoint, torch_dtype=torch.bfloat16, trust_remote_code=True).to(device)

inputs = tokenizer.encode("def print_hello_world():", return_tensors="pt").to(device)
outputs = model.generate(inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0]))

### Attribution

DeciCoder was trained on StarCoder Training Dataset, filtered for
Python, Java, JavaScript, Rust, Go, C++, C, and C#. For additional information, please
refer to [https://huggingface.co/datasets/bigcode/starcoderdata](https://huggingface.co/datasets/bigcode/starcoderdata).

```

### Limitations

The model has undergone training with source code from Python, Java,
JavaScript, Go, Rust, C++, C, and C#. While the primary language in the source is English, it does
contain other languages. Therefore, the model can produce code snippets
given some context. However, there\'s no assurance that the resulting
code will function as expected. It might be suboptimal, contain bugs, or
even exploits.

## Evaluation

Below are DeciCoder's pass@1 on MultiPL HumanEval scores

| Python | JavaScript | Java  | C++  | C#  | Rust  | Go  | C  | 
|:----------|:----------|:----------|:----------|:----------|:----------|:----------|:----------|
| 33.5%    | 29.3%    | 30.3%    |29.93%    |20.31%    |20.5%    |77.47%    |xx%    |


### Runtime Benchmarks

|Inference Tool/Hardware | Qualcomm AI 100 (tokens/sec) |
|:----------|:----------|
| Infery LLM | xxx   | 

- Throughput (tokens/sec) - Measured with an optimal batch size of 96

## Documentation

- [Notebook](https://colab.research.google.com/drive/1JCxvBsWCZKHfIcHSMVf7GZCs3ClMQPjs) CHANGE
- Blog post: [Introducing DeciCoder: The New Gold Standard in Efficient and Accurate Code Generation](https://deci.ai/blog/decicoder-efficient-and-accurate-code-generation-llm/)CHANGE
- Questions:Feel free to contact us via our [Discord Community!](https://discord.com/invite/p9ecgRhDR8/)CHANGE

## How to Cite

Please cite this model using this format.

```bibtex
@misc{DeciFoundationModels,
title = {DeciCoder},
author = {DeciAI Research Team},
year = {2023}
url={[https://huggingface.co/deci/decicoder-6b](https://huggingface.co/deci/decicoder-6b)},
}
```