Real head-to-head · same prompt, one shot

Kilo Code vs North Mini Code

Fable 5-class intelligence at ~59% less. The split-the-cost play. vs Cohere's free coder that beats models 4× its size, runs on your own Mac.

Kilo Code · contextVaries (Kilo dispatches across models)
North Mini Code · contextTBD
Kilo Code · price~59% less than Fable 5 solo
North Mini Code · priceFree — local
Kilo Code · vendorKilo
North Mini Code · vendorCohere

What I tested — same prompt, two models

I run the same fixed prompt set through every new model the day it drops — same string, one shot, single HTML file out — and I score the result 0–10 on whether it ran, how close it hit the brief, and how good it looked. Below is what came out when I gave the exact same prompts to Kilo Code and North Mini Code, side by side, on 0 shared tasks inside the Agent Operating System.

Both models were given identical prompts inside the Agent Operating System — no help, no iteration, no "best of N" tricks. I run each prompt once, save the HTML file the model produces, and score it 0–10 on whether it ran, how close it hit the brief, and how good it looked. The scoring is mine. The verdicts below are pulled from my source comparison guides at agentos.guide where I publish every score and the reasoning behind it.

Kilo Code · Used inside Agent OS as a routing layer: Fable 5 generates the plan, cheaper models execute. Bench scoring pending a head-to-head comparison.

North Mini Code · Wired into the Agent OS as the local-first coder for offline workflows. Bench scoring pending.

Side-by-side on 0 shared tasks

Click any cell to play that model's actual one-shot attempt. Medals are derived from my 0–10 scores per task (highest = 🥇, second = 🥈, third = 🥉).

Task ↓
Kilo Code
North Mini Code

Strengths & weaknesses I logged

Kilo Code

Strengths

  • Kilo's own rubric: Fable 5 plan = 9.1/10, GPT-5.5 plan = 8.3/10 — Kilo isolates where the intelligence actually lives
  • Plan quality stays high while execution cost drops
  • Drop-in for Agent OS — Kilo Split framework already wired

Trade-offs

  • Adds routing complexity — two model providers in one workflow
  • No per-task goldiebench head-to-heads yet

North Mini Code

Strengths

  • Tiny + fast — runs locally with no GPU farm required
  • Out-scores models 4× its parameter count on agent-coding benchmarks (Cohere's own number)
  • Zero cost — free for individuals, runs offline

Trade-offs

  • Cohere's own number, not independently verified
  • No goldiebench per-task scores yet

Pricing & context — the spec sheet

Spec Kilo Code North Mini Code
VendorKiloCohere
Context windowVaries — Kilo splits planning from execution across multiple modelsSpecs not yet public
Price~59% less than Fable 5 soloFree — local
Pricing detailKilo Code is a routing layer that splits planning (heavy model) from execution (cheaper model) so you get Fable-5-class plans driving GPT-5.5-class builds. Total spend lands at ~59% less than running Fable 5 end-to-end.Cohere's free coding model — small parameter count, runs locally, designed to out-punch models four times its size on agent-coding benchmarks. No token bill, no API key required.
Release2026-06-162026-06-19
Bench coverage0/0 scored · avg —0/0 scored · avg —

The verdict — which should you pick?

Not enough scored shared tasks yet for a head-to-head average. The live demos for both are on the matrix above — play them and form your own opinion.

If you only run one of these inside your stack, the head-to-head average above is the call. If you can run both, my honest play is to wire Kilo Code and North Mini Code both into the Agent Operating System and dispatch each from the kanban by task type — cost-conscious operators who run high-volume agent loops → Kilo Code, local coding loops on a mac where you don't want a token meter → North Mini Code. That's the same setup I run for the 3,600+ founders inside the AI Profit Boardroom.

FAQ — Kilo Code vs North Mini Code

Which is better, Kilo Code or North Mini Code?

On Goldie Bench, Kilo Code averages no scored verdicts yet across the shared tasks, with 0 gold, 0 silver, 0 bronze overall. North Mini Code averages no scored verdicts yet, with 0 gold, 0 silver, 0 bronze. Not enough scored shared tasks yet to call a winner.

How much does Kilo Code cost vs North Mini Code?

Kilo Code: Kilo Code is a routing layer that splits planning (heavy model) from execution (cheaper model) so you get Fable-5-class plans driving GPT-5.5-class builds. Total spend lands at ~59% less than running Fable 5 end-to-end. North Mini Code: Cohere's free coding model — small parameter count, runs locally, designed to out-punch models four times its size on agent-coding benchmarks. No token bill, no API key required.

What's the context window for Kilo Code vs North Mini Code?

Kilo Code has a Varies — Kilo splits planning from execution across multiple models context window. North Mini Code has a Specs not yet public context window.

When should I pick Kilo Code over North Mini Code?

Pick Kilo Code for: Cost-conscious operators who run high-volume agent loops; Multi-step workflows where the plan is the expensive part; Teams already paying for Fable 5 who want to keep the plan but drop the execution bill. The trade-off is the weaknesses we logged on the bench: Adds routing complexity — two model providers in one workflow; No per-task goldiebench head-to-heads yet.

When should I pick North Mini Code over Kilo Code?

Pick North Mini Code for: Local coding loops on a Mac where you don't want a token meter; Offline / air-gapped agent workflows; Operators who want to compare a small-but-mighty model against the larger field. The trade-off is the weaknesses we logged on the bench: Cohere's own number, not independently verified; No goldiebench per-task scores yet.

How does Goldie Bench score Kilo Code vs North Mini Code?

Every demo on this page was built by Julian Goldie inside the Agent Operating System — same fixed prompt for both models, one shot, single HTML file out. Each result gets a 0–10 score on whether it ran, how close it hit the brief, and how good it looked. The highest score on each task gets gold; second gets silver; third gets bronze. See methodology for full provenance.

The same stack Julian uses

Run this stack yourself.

Every demo on this bench was built inside the Agent Operating System — one prompt, one shot, single HTML file out. The Agent OS, the prompts, the templates, the weekly walkthroughs and 3,600+ founders shipping with it every day all live inside the AI Profit Boardroom.

3,600+founders
258documented wins
38countries
$100k+/mocommunity MRR