Pinned
Kilo
4,183 posts
Kilo
@kilocode
Kilo is the all-in-one agentic engineering platform.
3M+ Kilo Coders. Open source since day one!
- Kilo reposted
Lineup for Inside The AI Coding Stack (7/1): @nvidia — Harry Kim on GPU infra for AI-native workloads @friendliai — Gon Chun on frontier AI inference for agents @MiniMax_AI — Victor Su-Ortiz on M3 + reasoning @kilocode — Brian Turcotte on agentic coding in production Spots are
It's live now, and the benchmark it routes on is public, so you can check it before trusting it. Full writeup from @coldopn here:- Having a great experience with @kilocode today. Discovered it due to the claudy meltdown. Awesome and cheap tech!
- Focusing exclusively on token spend is like measuring a race car by its fuel consumption. Engineering output, deployment frequency, and ROI are what matter. Our CEO, @s_breitenother, explains why.If an engineer costs $25K/month and AI costs $2K/month, why are so many companies focused on optimizing the smaller number? AI isn’t a cost center. It’s a productivity multiplier. Here's my latest @Forbes piece on why ROI matters more than token spend: forbes.com/councils/forbe…
- Next-Edit is live in Kilo, powered by Mercury Edit 2 from @_inception_ai. Autocomplete predicts the next few tokens ahead of your cursor. Next-Edit predicts your next actual edit anywhere in the file. Hit Tab to accept. And it's free for everyone for 30 days!
It's free through July 23 via the Kilo Gateway. No trial, no card. Read the full post here: - Augment is sunsetting its JetBrains IDE extensions with about a month of notice. The recommended replacement is a CLI or an enterprise agent platform. For most PyCharm and IntelliJ devs, that's not what they signed up for. Kilo's JetBrains plugin already does what those users
We broke down what changed, why Augment is moving upmarket, and how the Version 7 native plugin maps to what JetBrains users have been asking for: - Last week, we had 8 AI models review the same code, with the same prompt, at the same time. The cheapest one was the only model that caught every bug, including one that 7 of the others missed completely. The full test, the bugs, and which model found everything:
