I'm excited to share that we've built the world's most capable AI software engineer, achieving 30.08% on SWE-Bench – ahead of Amazon and Cognition. This model is so much more than a benchmark score: it was trained from the start to think and behave like a human SWE.
I get asked every now and then about whether to use
@LangChainAI
or instructor
My general advice normally boils down to
1. How familiar are you with either library
2. How long are you maintaining this code for
This is kind of how I've been thinking about it.
SmartCrawl is insane - we're seeing some really cool use cases 🤯
One of my favorites is extracting data behind auth walls 🔒
Check out this demo where it logs into
@firecrawl_dev
, gets my usage data in seconds, and turns it into an API!
@wordware_ai
is the best toolkit for building AI.
Today we say hello to 🇯🇵
We are officially viral with in Japan 🚀
🚨 87k users in the last hour total 🚨 4.6 mm users in total 🚨
All proudly powered by Claude Sonnet 3.5 and
@AnthropicAI
helping us
I stopped writing code entirely in 2024.
I only copy-paste code generated by AI ✌️🤓
How do I do that? 10 hints:
Hint 1: if you have a creative task such as code architecture, you want to use so called chain of thoughts. You add "Think step-by-step" to your prompt and enjoy a
The team is on it! Updated to Gemma 2 in GroqCloud and now available to you the developer community. Let us know if this new version of Gemma is useful to your apps. Send pics... 😁
How the hell Phi-3.5 is even possible?
Phi-3.5-3.8B (Mini) somehow beats LLaMA-3.1-8B..
(trained only on 3.4T tokens)
Phi-3.5-16x3.8B (MoE) somehow beats Gemini-Flash
(trained only on 4.9T tokens)
Phi-3.5-V-4.2B (Vision) somehow beats GPT-4o
(trained on 500B tokens)
how? lol
New in the Cohere Toolkit!
Add interactive HTML app rendering, configurable authentication, and multi-step tool use to your RAG applications built with toolkit.
Check it out: