On December 5th, the Byte Bean Bag model team launched the latest code model evaluation benchmark - FullStack Bench, covering over 11 real-world scenarios, supporting 16 programming languages, and including 3374 questions. Compared to previous evaluation standards, this benchmark can more accurately assess the code development capabilities of large models across a broader programming domain, facilitating the optimization of models in real-world programming tasks. Current mainstream code evaluation benchmarks, such as HumanEval and MBPP, typically focus on basic and advanced.

Ant Group has released the 4bits quantized version of CodeFuse-CodeLlama-34B. The quantized model can be loaded on a single A10 or RTX4090. The quantized model achieves a performance of 73.8% on the Humaneval pass@1 metric. CodeFuse is Ant's self-developed large model for code generation, aimed at enhancing developers' efficiency by providing intelligent suggestions and real-time support.

Meta has released the open-source LLM Code Llama, built on Llama 2, claiming superior performance on coding tasks compared to other LLMs. Code Llama comes in three variants: a base model, a model specifically for Python, and a model for natural language instructions. Code Llama provides stable generation and supports multiple programming languages, suitable for tasks such as code completion, debugging, and real-time code assistance.

Silicon Heart Technology has launched a new intelligent software development system, aiXcoder Europa, focusing on solving the challenges of implementing large code models in enterprises. The new version includes features such as code defect detection and repair, automatic generation of unit test code, automatic generation of code comments, and code explanation, covering a wider range of coding scenarios. aiXcoder has provided intelligent software development solutions for multiple enterprises and has explored a mature methodology for private deployment and personalized training.