codegen

Code LLMs Explained,
CodeGen

The CodeGen model is a pre-trained neural network-based model for program synthesis that generates code from natural language descriptions. It uses an encoder-decoder architecture with a transformer-based design to capture complex dependencies between different parts of the code expression.

Model Details View All Models

Technical Experts

50 Custom AI projects

4.8 Minimum Rating

An Overview of CodeGen

The CodeGen model is a pre-trained neural network-based model for program synthesis that generates code from natural language descriptions.

38.83 BLEU score and 0.376 F1 score on CoNaLa dataset

Score of 38.83 on BLEU

CodeGen-Multi 16.1B has achieved state-of-the-art performance on program synthesis tasks, with a BLEU score of 38.83 on the CoNaLa dataset.

CodeGen is trained on MTPB with 115 problems

Trained on 115 problems

CodeGen is trained on a Multi-Turn Programming Benchmark (MTPB) that has 115 expert-written problems, each with a multi-step description in natural language.

CODEGEN-NL models outperforms GPT-NEO and GPTJ

Outperforms GPT-NEO

CODEGEN-NL models (350M, 2.7B, and 6.1B) outperform or are comparable to GPT-NEO and GPT-J models.

Blockchain Success Starts here

  • About Model

  • Model Highlights

  • Training Details

  • Model Types

  • Key Results

  • Model Features

  • Model Tasks

  • Fine-tuning

  • Benchmark Results

  • Sample Codes

  • Limitations

  • Other LLMs

Model ParametersHighlight
CODEGEN-NL 350M350 millionGood balance between accuracy and efficiency
CODEGEN-NL 2.7B2.7 billionHigh accuracy with a wide range of programming languages
CODEGEN-NL 6.1B6.1 billionImproved accuracy and ability to handle more complex code expressions
CODEGEN-NL 16.1B16.1 billionState-of-the-art performance on program synthesis tasks
TaskDatasetScore
Pass@1HumanEval29.28
Pass@10HumanEval49.86
Pass@100HumanEval75
Multi-Turn ProgrammingPile30.33
Multi-Turn ProgrammingBigquery26.27
Multi-Turn ProgrammingBigpython47.34
Pass@1MBPP35.28
Pass@10MBPP67.32
Pass@100MBPP80.09