augmentcode's profile picture. The only AI coding Agent built for professional software engineers and large codebases.

Augment Code

@augmentcode

The only AI coding Agent built for professional software engineers and large codebases.

Pinned

Claude Haiku 4.5 is now live in Augment Code. In our internal evals, we found it delivered ~90% of Sonnet 4.5’s performance at 34% faster response speed. We're giving up to 20 free user messages over the next two weeks to everyone subscribed to a monthly paid plan once the…


Haiku 4.5 advances small-model performance significantly. In our internal tests, Haiku 4.5 came close to Sonnet 4.5 in terms of quality, yet at a much higher speed: → Head-to-head preference: In testing across identical prompts, internal testers preferred responses from Sonnet…

augmentcode's tweet image. Haiku 4.5 advances small-model performance significantly. In our internal tests, Haiku 4.5 came close to Sonnet 4.5 in terms of quality, yet at a much higher speed:

→ Head-to-head preference: In testing across identical prompts, internal testers preferred responses from Sonnet…

Introducing Claude Haiku 4.5: our latest small model. Five months ago, Claude Sonnet 4 was state-of-the-art. Today, Haiku 4.5 matches its coding performance at one-third the cost and more than twice the speed.

claudeai's tweet image. Introducing Claude Haiku 4.5: our latest small model.

Five months ago, Claude Sonnet 4 was state-of-the-art. Today, Haiku 4.5 matches its coding performance at one-third the cost and more than twice the speed.


Augment Code reposted

Vibe coding is a big topic these days -- as is the generic looking designs that result. Something to think about for designers looking to inject more of their taste into 0-1: how do we talk to agents about design so they listen? augmentcode.com/blog/brand-gui…


Augment Code reposted

On the flip side, AI coding works less well on properly architected codebases. But context engineering can bridge the gap.

The reason AI coding works so well is that the source code of the median app was already slop before LLMs.



Upcoming webinar: How Collectors learnt to assess AI coding tools Most teams are experimenting with AI coding tools. Very few have a clear way to tell which ones actually help. At @collectors, CTO Dan Van Tran built a framework for evaluating these tools in real engineering…


Augment Code reposted

Excited for Augment to be featured at OpenAI Dev Day! If your codebase is large and messy (who’s isn’t??), you should try augmentcode.com

guygr's tweet image. Excited for Augment to be featured at OpenAI Dev Day! If your codebase is large and messy (who’s isn’t??), you should try augmentcode.com

Today we announced an upcoming change to how Augment Code is priced. Starting October 20, 2025, plans will move from user messages per month → to a pool of credits that can be spent across activities. 📖 Read more on our blog → augmentcode.com/blog/augment-c…


United States Trends

Loading...

Something went wrong.


Something went wrong.