gpt 5.5 feels so much faster than opus 4.7

just sent the StarScout extension for review to be publish in the Chrome Web Store, wish me luck!!

btw, this was cause my card bouncing, so it put me back in the free tier, which has separate usage from the paid one lol

Giving away Command Code Max subscription to someone at random who follows me and Command. RT. That’s more than 5 billion tokens of DeepSeek v4 pro. In 24hrs. LFG!! Read the eng deep dives below, good for all not just us.

openusage tells me I've spent just about 40$ of inference on my 24€ a month plan, so it doesn't look crazy

absolutely LOVE this

usei o Codex CLI como interventor no meu server de produção do StarScout e foi tudo como esperado, respeitou bem as regras que eu impus e resolveu o problema

it's been cool to start 'shipping' stuff I do before AI all of these would just die in a private repo

minha extensão tá com 14 usuários, o milhão vem

os meus modelos ultimamente com Codex de 20 dol e OpenCode Go - planejamento geral: GPT 5.5 High - planejamento frontend: Kimi K2.6 - execução geral: GLM 5.1, GPT 5.5 Low se for algo complexo - execução frontend: Kimi K2.6 e bem satisfeito so far