procedure2012/Aurora-Coder-Base

1---2license: mit3library_name: transformers4---5# Aurora-Coder-Base6<!-- markdownlint-disable first-line-h1 -->7<!-- markdownlint-disable html -->8<!-- markdownlint-disable no-duplicate-header -->9 10<div align="center">11  <img src="figures/fig1.png" width="60%" alt="Aurora-Coder-Base" />12</div>13<hr>14 15<div align="center" style="line-height: 1;">16  <a href="LICENSE" style="margin: 2px;">17    <img alt="License" src="figures/fig2.png" style="display: inline-block; vertical-align: middle;"/>18  </a>19</div>20 21## 1. Introduction22 23Aurora-Coder-Base is a code-specialised checkpoint fine-tuned on a deduplicated multi-language corpus. It targets repository-level code completion and agentic tool use.24 25## 2. Evaluation Results26 27### Comprehensive Benchmark Results28 29<div align="center">30 31| | Benchmark | StarBase | CodeNova | Aurora-mini | Aurora-Coder-Base |32|---|---|---|---|---|---|33| **Core Reasoning Tasks** | Math Reasoning | 0.579 | 0.535 | 0.552 | 0.592 |34|  | Logical Reasoning | 0.810 | 0.831 | 0.824 | 0.846 |35|  | Common Sense | 0.731 | 0.734 | 0.750 | 0.761 |36| **Language Understanding** | Reading Comprehension | 0.681 | 0.711 | 0.673 | 0.732 |37|  | Question Answering | 0.607 | 0.592 | 0.591 | 0.628 |38|  | Text Classification | 0.833 | 0.808 | 0.832 | 0.843 |39|  | Sentiment Analysis | 0.785 | 0.780 | 0.756 | 0.806 |40| **Generation Tasks** | Code Generation | 0.676 | 0.665 | 0.642 | 0.692 |41|  | Creative Writing | 0.609 | 0.623 | 0.639 | 0.656 |42|  | Dialogue Generation | 0.657 | 0.616 | 0.618 | 0.673 |43|  | Summarization | 0.762 | 0.780 | 0.748 | 0.787 |44| **Specialized Capabilities** | Translation | 0.808 | 0.778 | 0.808 | 0.816 |45|  | Knowledge Retrieval | 0.655 | 0.645 | 0.691 | 0.697 |46|  | Instruction Following | 0.725 | 0.743 | 0.742 | 0.779 |47|  | Safety Evaluation | 0.715 | 0.700 | 0.703 | 0.759 |48 49</div>50 51### Overall Performance Summary52The Aurora-Coder-Base demonstrates strong performance across all evaluated benchmark categories, with particularly notable results in reasoning and generation tasks.53 54## 3. Chat Website & API Platform55We offer a chat interface and API for you to interact with Aurora-Coder-Base. Please check our official website for more details.56 57## 4. How to Run Locally58 59Please refer to our code repository for more information about running Aurora-Coder-Base locally.60 61### Temperature62We recommend setting the temperature parameter to 0.6.63 64## 5. License65This repository is released under the mit license. The model supports commercial use.66 67## 6. Contact68If you have any questions, please contact us at hello@aurora.dev.69