Kimchi Blog

Same Model, Different Results: What Actually Separates AI Coding Agents

We ran GLM-5.2 through four AI coding harnesses on Terminal-Bench 2.1. Same model, same 89 tasks, same benchmark. The harness made a 13-point difference. Here's what that means.

Michal Ján Sallem · 8 min read

Kimchi now has native debugger support

Real breakpoints, real stack traces, live variable inspection. No more print statements. Kimchi speaks the Debug Adapter Protocol natively, so the agent observes runtime state instead of guessing at it.

Kimchi Team · 7 min read

A model weight file can't call home! The case for Open-weight models.

Seven countries have restricted the DeepSeek app on government devices. Zero have restricted the DeepSeek model weights. Here's why the distinction matters for your security review.

· 8 min read

Technology executives shouldn't give up control

AI coding is becoming the software factory. Enterprises need to own the enforcement path, the economics, and the definition of done.

Leon Kuperman, Founder of Cast AI · 11 min read

GLM-5.2 beat Claude Fable 5 on one leaderboard. Here's what the rest of the data says.

Open weights now sit within 1-3 points of the frontier on individual coding benchmarks at 6-30x lower cost. On composites, private benchmarks, and long-horizon agentic work, the frontier lead is intact.

Kimchi Team · 9 min read

A two-day hackathon on Kimchi inference, at 1/13th the frontier cost

~120,700 requests. 1,580 active sessions. ~98% success rate. ~$4.4k total spend - about 1/13th of the modeled frontier-model cost.

Kimchi Team · 5 min read

Claude Code costs $2,000/month per engineer. Here's what we built instead.

Why utility-priced AI coding tools blow budgets when deployed like SaaS, and how Kimchi's model orchestration and spend console fix it.

Laurent Gil · 6 min read

Kimchi Coding is GA, and why we built it

Today we take Kimchi Coding out of Early Access. It's live at kimchi.dev with full production support behind it — and this is the story of why we built it.

Laurent Gil · 6 min read

Why we use 3 models instead of 1 for every coding task

How Kimchi routes seven coding roles across Kimi K2.7, MiniMax M3, and Nemotron-3-Ultra - and why matching the model to the task cuts cost 40–60% without sacrificing quality.

· 6 min read

Inside the Coding Harness: Why Architecture Beats Models for AI Development

The difference between autocomplete tools and a coding harness is not the model. It's the architecture.

· 6 min read

Kimchi Coding becomes the first coding agent to integrate MiniMax M3

We added MiniMax M3 to the recipe. Kimchi Coding just got spicier!

· 5 min read

The $500 Million AI Bill

Why Generative AI is Breaking the SaaS Procurement Model

· 7 min read

Say hello to Kimchi

· 5 min read