Salesforce AI的CODA-1.7B：离散扩散代码生成的飞跃

探索Salesforce AI的CODA-1.7B，这是代码生成的离散扩散模型，提供较少的参数和可重现管道的竞争性能。

Salesforce AI is making waves with CoDA-1.7B, a compact yet powerful discrete-diffusion model designed for code generation. This model leverages bidirectional context and parallel token updates, marking a significant advancement in the field.

Salesforce AI正在使用CODA-1.7B进行浪潮，这是一种专为代码生成而设计的紧凑而强大的离散扩散模型。该模型利用双向上下文和平行的令牌更新，标志着该领域的显着进步。

Understanding CoDA-1.7B's Architecture and Training

了解CODA-1.7B的架构和培训

CoDA-1.7B adapts a 1.7B-parameter backbone to discrete diffusion for text. It iteratively denoises masked sequences using full-sequence attention, enabling native infilling and non-autoregressive decoding. The model card documents a three-stage pipeline: pre-training with bidirectional masking, supervised post-training, and progressive denoising at inference. Reproducible scripts for TPU pre-training, GPU fine-tuning, and evaluation are also provided.

CODA-1.7B适应1.7B参数骨干，以使文本离散扩散。它迭代地使用全面的注意来掩盖了掩盖序列，从而实现了本地填充和非自动向解析的解码。该模型卡记录了三阶段的管道：双向掩蔽，监督后训练和推理时进行性降解的预训练。还提供了用于TPU预训练，GPU微调和评估的可再现脚本。

Benchmark Performance

基准性能

CoDA-1.7B-Instruct demonstrates impressive results on standard code-gen suites, including HumanEval (54.3%), HumanEval+ (47.6%), MBPP (47.2%), MBPP+ (63.2%), and EvalPlus aggregate (55.4%) pass@1. These results are competitive with some 7B diffusion models, such as Dream-7B-Instruct (57.9% HumanEval), while utilizing significantly fewer parameters.

CODA-1.7B教学结构在标准代码套件上表现出令人印象深刻的结果，包括Humaneval（54.3％），HumaneVal+（47.6％），MBPP（47.2％），MBPP+（63.2％）和Evalplus croxpregate（55.4％）Pass@1。这些结果与某些7B扩散模型（例如Dream-7B - 教学（57.9％）人类Val）具有竞争力，同时使用明显较少的参数。

Inference Behavior and Deployment

推理行为和部署

Generation cost in CoDA is governed by the number of diffusion steps. Users can tune latency/quality trade-offs using parameters like STEPS and ALG="entropy". The model updates tokens in parallel under full attention, which targets lower wall-clock latency at small scale compared with larger diffusion models. The release includes a FastAPI server with OpenAI-compatible APIs and an interactive CLI for local inference. Model cards and a Hugging Face collection centralize artifacts, with checkpoints published under CC BY-NC 4.0 on Hugging Face.

CODA中的发电成本受扩散步骤的数量约束。用户可以使用步骤和ALG =“熵”等参数来调整潜伏期/质量权衡。该模型在全部注意力下并行更新标记，与较大的扩散模型相比，该模型以较小的壁式潜伏期为目标。该版本包括具有兼容openAI的API的FastAPI服务器和用于本地推断的交互式CLI。模型卡和拥抱的脸部收集集中式人工制品，在CC BY-NC 4.0下出版了检查点。

Our Take

我们的看法

CoDA-1.7B is a valuable reference for discrete-diffusion code generation at a smaller scale. Its bidirectional denoising with parallel token updates and reproducible pipeline from pre-training to SFT and serving make it an accessible and practical tool. The ability to tune throughput/quality using step count and decoding knobs is also operationally advantageous. I believe CoDA-1.7B is a step toward making AI code generation more efficient and accessible to developers.

CODA-1.7B是较小规模的离散扩散代码生成的宝贵参考。它的双向denoings通过平行令牌更新和可再现的管道从预培训到SFT和服务，使其成为一种易于访问且实用的工具。使用步骤计数和解码旋钮调整吞吐量/质量的能力在操作上也有利。我认为CODA-1.7B是使AI代码生成更有效且可为开发人员访问的一步。

So, what are you waiting for? Dive into the world of CoDA-1.7B and see how it can revolutionize your code generation workflow!

那么，您还在等什么？潜入Coda-1.7b的世界，看看它如何彻底改变您的代码生成工作流程！