Add corrections, implementation notes, pricing changes, or usage caveats for other readers.
Last updated
Aug 5, 2026
Input modalities
Output modalities
Capabilities
1,048,576 tokens
Recent tweets and retweets from Kilo Gateway
🚨 FREE MODEL ALERT!
@TencentGlobal's Hy3 is completely free in Kilo Code for 1 week.
Just pull up Kilo for VS Code or the Kilo CLI, select Tencent Hy3 [Free], and start building!
Install Kilo for VS Code here: kilo.ai/install
Link
Install Kilo Code in 60 Seconds — VS Code, JetBrains & CLI
Install Kilo Code in under a minute and start coding with AI in your editor. One-click setup for VS Code, Cursor, JetBrains, Windsurf, and the command line.
kilo.ai
We analyzed 10,643 code reviews.
Every finding classified by severity and category, then normalized per review so volume leaders don't win by default.
Here's how reviewers actually behave, model by model.
Authoring and reviewing are different jobs, and teams are already splitting them.
32.3% of attributed reviews used a different model than the one that wrote the code.
Most common pairing: Step 3.7 Flash writing, Laguna M.1 reviewing.
Full table: kilo.ai/articles/open-weight…
Link
State of Open-Weight Models for AI Code Review Workflows
Data from 10,643 Kilo Code Reviewer runs: open-weight models matched closed leaders on critical findings, while one outlier drove the security gap.
kilo.ai
Discuss this model
Add corrections, implementation notes, pricing changes, or usage caveats for other readers.