I plugged Qwen 3.8 27b into DeepSeek Harness and asked for a replica Age of Empires 2

Returned to my Studio 70 minutes later to find the Mac still purring... So far my local model tests on this prompt have produced unsharable results Let's see what we get

bartslodyczkaAug 241

DeepSeek Harness. Hands down the standout, and honestly not what we expected going in.

The final page had a level of polish we didn't see coming. It also took its time — this one clearly likes to think before it builds, and it ran way longer than Pi. Results:

egoAug 2410

We tested Ox Alpha in Pi and DeepSeek Harness.

Same task for both: use ego lite to scope out a few great sites, then build the most impressive single-file HTML landing page. The gap was bigger than we expected 👇

egoAug 24122

Every run of my NYC prompt since April.

24 of them, side by side. Same prompt every time. One HTML file, three.js, no assets, everything procedural. Opus 4.6 gave me boxes on a green plane. GPT-5.5 Standard gave me an empty sky. DeepSeek V4 came back almost black. Then Opus 5

aipulsedailyAug 243