Nex AGI

N2 (Nex-N2-Pro)

Free local frontier model — runs in Hermes for $0.

ContextSpecs not yet public
PricingFree — runs locally inside Hermes
Tasks tested0
Avg scorecurrently unranked
Medals🥇0 🥈0 🥉0
Release2026-06-12

What is N2 (Nex-N2-Pro)?

N2 (Nex-N2-Pro) is the Nex AGI frontier model with a Specs not yet public context window, released 2026-06-12. Tagline: Free local frontier model — runs in Hermes for $0..

Pricing detail. Nex AGI's flagship N2 (full name Nex-N2-Pro) runs free for individuals inside Hermes on your own Mac. Vendor-reported SWE-bench Verified: 80.8% — the highest published coding score among the free dragons (Kimi K2.6 / GLM / N2).

How I use it inside the Agent OS. Wired into Hermes as the free local frontier coder. Dispatched from Agent OS for any task where the per-token bill would dominate.

What I built with N2 (Nex-N2-Pro)

Every model on Goldie Bench gets the same fixed prompt set — one shot, single HTML file out — and I score the result 0–10 inside the Agent Operating System. Here's what N2 (Nex-N2-Pro) shipped on the bench: 0 one-shot demos across Specs not yet public of context. Of those, 0 are scored against the field with my honest 0–10 from the source guides at agentos.guide.

Strengths

  • Vendor-reported 80.8% on SWE-bench Verified — top of the free open-weights field
  • Runs locally inside Hermes — no token meter, no rate limits
  • Builds two playable 3D games in one-shot per Julian's three-dragons writeup

Trade-offs

  • Vendor-reported benchmark (not independently verified yet)
  • No goldiebench per-task scores yet

Best for

  • Local agent loops that need a free frontier-class coder
  • Privacy-sensitive workflows (no cloud round-trip)
  • Operators experimenting with Hermes who want to compare local models against the cloud field

Every demo by N2 (Nex-N2-Pro)

0 live demos, sorted by category. Click any tile to play the actual one-shot result. Verdicts and 0–10 scores are pulled from the source guides where I posted them publicly.

Head-to-heads with N2 (Nex-N2-Pro)

Direct comparisons against every other scored model on the bench:

N2 (Nex-N2-Pro) vs Opus 4.8 N2 (Nex-N2-Pro) vs GLM-5.2 N2 (Nex-N2-Pro) vs Grok N2 (Nex-N2-Pro) vs Qwen 3.7 N2 (Nex-N2-Pro) vs Kimi K2.7

Read more on agentos.guide: /n2-hermes, /n2-freeclaude, /three-dragons

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