Gru.ai

Software Agents from

[Gru.ai @BabelCloudAI]

DeepSeek V3 0324 has demonstrated remarkable capabilities; the power of the open-source community should not be underestimated.

[Gru.ai @BabelCloudAI]

Some practices on AI coding
https://t.co/INGu7MmoWZ

[Gru.ai @BabelCloudAI]

Simply install a Github bot, and it seamlessly integrates into your workflow to instantly boost your unit test coverage
https://t.co/Jn7jjCXcIL

[Gru.ai @BabelCloudAI]

Want to stay updated on AI's latest practices in software engineering?
Don't miss Hailong, CEO of Gru, sharing cutting-edge insights at the AIE Conference.
#AIinSoftwareEngineering #TechInnovation #AIEConference #Leadership
https://t.co/Lwh98uShTn

[Gru.ai @BabelCloudAI]

More and more agents will become contributors in your repos in 2025. https://t.co/mGmotorfCK

[PyCon China @PyConChina]

We hope you had a wonderful time at PyCon China 2024.🐍🎉

Huge thanks to all our sponsors, volunteers, partners, and everyone who made this event unforgettable!

See you next year for PyCon China 2025! #PyCon #PyConChina #PyConChina2024

[Hailong Zhang @zhlmmc]

o1 is not an upgrade of gpt-4o, it’s more like a sibling that performs better in “complicated” situations. Now, the problem is how do you identify these “complicated” scenarios in an automatical way?

[Hailong Zhang @zhlmmc]

@BitValentine @eyishazyer @BabelCloudAI No, actually AI boosts human innovation as human can focus on what to do while AI focus on how to do.

[Gru.ai @BabelCloudAI]

https://t.co/QNTzFJRqtT provides three more agents: Assistant Gru will Helps users solve standalone technical issues, which is now in public use. Test Gru can Generates unit test code automatically and Babel Gru will Assists in building end-to-end projects

[Gru.ai @BabelCloudAI]

https://t.co/QNTzFJRqtT ranked first with a high score of 45.2% in the latest data released by SWE-Bench-Verified Evaluation, the authoritative standard for AI model evaluation, which is a collaboration between SWE and OpenAI. #GruAI #OpenAI #SWEBench

[Gru.ai @BabelCloudAI]

Behind our winning score is Bug Fix Gru, an Agent designed to auto-fix bugs based on user issues. Here is a video about how Bug Fix Gru works. https://t.co/ByLZApkXMc

[Gru.ai @BabelCloudAI]

@nexteacc You can use Assistant GRU on https://t.co/QNTzFJRqtT, it can help you handle independent technical problems.
https://t.co/L0OITfeor7

[Gru.ai @BabelCloudAI]

Apple want to acquire Perplexity ?
We created our own version with an AI Agent in just 30 minutes. #Apple #AIAgents

[Gru.ai @BabelCloudAI]

@nexteacc Hello, thank you for your attention!
Babel Gru is still in the lab stage at the moment, and it is not yet publicly available, only a small amount of closed testing has been done.

[Gru.ai @BabelCloudAI]

Our score was just officially accepted by SWE bench, and https://t.co/QNTzFJRqtT received a score of 35.67%! This ranks us first among the teams that provided a trajectory. Well down Gru!
https://t.co/XaH6t51vQo
#SWEBench

[Gru.ai @BabelCloudAI]

As a developer, you often face a lot of tedious tasks. That's where https://t.co/QNTzFJRqtT comes in to help. Let's check out two examples.
https://t.co/YrGapFW73Y

[Gru.ai @BabelCloudAI]

12 Predictions for the Large Language Model Industry
https://t.co/WThQbJ4nyo

[歸藏(guizang.ai) @op7418]

另一个类似Devin的AI 软件构建 Agents，支持通过自主驱动的Agents共同努力解决编码、调试、测试、部署等问题。

Babel Agent 可根据需求文档自主设计和分解任务，并逐一执行。

Babel Agent 不仅能自主编写代码，还能自主编译代码，然后根据编译反馈的问题进行自主调试。

当需要集成新发布的 Claude 3（不在大型模型的固有知识范围内）时，Babel Agent 会自主搜索 SDK、查找文档、编写代码，然后进行测试和验证。

必要时，Babel Agent 将编写自动测试代码、执行测试并自我纠正问题。

遇到要求不明确或未提供必要信息时，巴别代理会寻求人工帮助。此外，如果尝试不同方法后仍无法完成任务，它也会这样做。

Babel Agent 支持对需求的迭代更改和在线问题的自主修正。

官网：https://t.co/R0DAw6zctr

[Gru.ai @BabelCloudAI]

@dotey 😍