Luiz.js

25. Dev Frontend | Vue.js Seminarista.

[Luiz.js @luizfeer]

@wescld Boquinha santa hein …

[Luiz.js @luizfeer]

quem só vê essas telas o dia todo?

[jhey ʕ•ᴥ•ʔ @jh3yy]

heard you liked clean code 💦

[Luiz.js @luizfeer]

@DehumanoaDeus sei não se vem no preço de competir com o macbook pro hein

[Luiz.js @luizfeer]

@devVitorRibeiro @Gabuwp @BolhaDevs Vc e maluco de rodar sem split mano

[Luiz.js @luizfeer]

amo expo

[Krzysztof Magiera @kzzzf]

In case you ever wondered how to scan that QR code with your Simulator 😉

[Luiz.js @luizfeer]

@birobirobirodev sinistro

[Luiz.js @luizfeer]

@fabioricotta Brevo>>>>

[Luiz.js @luizfeer]

@djwesleyreal Pior, no Symbiam, Whatsapp S40

[Luiz.js @luizfeer]

@lucas_montano Namoral to sofrendo com tanta att em .env alguem recomenda um vault pra sincronizar?

[Luiz.js @luizfeer]

@SahilPanhotra @cursor_ai 2.0 > kimi 2.6
2.5 >>>>>>>

[Luiz.js @luizfeer]

@govbr

[Governo do Brasil @govbr]

A-CA-BOU a taxa das blusinhas!

Compras internacionais até 50 dólares não são mais tributadas pelo Governo do Brasil. O imposto de importação acabou.

O Governo do Brasil tá do lado do povo brasileiro.

[Luiz.js @luizfeer]

@lucas_montano é sobre mano, sucesso 🔥🚀

[Luiz.js @luizfeer]

@BolhaDevs seguindo as ultimas mudanças
Com ~500 conto compensa o Codex Pro ou Claude Max?
Claude dobrou, mas Codex tbm até dia 31

[Luiz.js @luizfeer]

@FelipeTechPM @BolhaDevs Oracle tem vps 32gb ram free melhor que isso po

[shadcn @shadcn]

We're doing it again.

[stevibe @stevibe]

Qwen3.6 35B-A3B dropped yesterday, so I ran it on 4 GPUs to see how it performs:

🟣 RTX 3090 — 49.78 tok/s, TTFT 852ms
🟡 RTX 4090 — 118.93 tok/s, TTFT 686ms
🟢 RTX 5090 — 160.37 tok/s, TTFT 409ms
🔵 DGX Spark — 59.98 tok/s, TTFT 228ms

I went with ollama as the backend because honestly, it's the easiest way for most people to get started. One command, model pulled, done.

I used Q4_K_M (24GB) across all four cards. The reason is the 3090 and 4090 don't support NVFP4 (only the 5090 and DGX Spark could use it). Keeping the same quant everywhere felt like the fairest way to compare.

And yes, you can absolutely squeeze more performance out of every card with vLLM, SGLang, or TensorRT-LLM. But that's not what this test is about. This is just the out-of-the-box experience for folks who own a GPU and want to try the new model tonight.