@GEH00073 I created something similar. I created a private MCP server for facial identification: it receives a photo and tells you who the person is, comparing against a local database of known faces using InsightFace's buffalo_l model. I haven't tested it yet; my StackChan is on its way.
What a time to be alive 🔥
• DeepSeek-V4-Flash-0731
• Qwen3.8-Max
• MiniMax-H3
• Kimi K3
And still coming:
• DeepSeek V4 Pro
• GLM-5.5
The open-weight era is accelerating insanely fast.
Boa! Sou dev, eu instalei o Hermes na minha VPS há uns meses pra testar, mas sinceramente acabou ficando parado lá. Na minha rotina, não achei nada que superasse o que já faço com outras IAs de CLI no dia a dia.
Mas já que você mergulhou fundo nisso, compartilha aí depois uns workflows seus! Quem sabe não me anima a dar uma segunda chance pra ele. 🤣
Tried the same technical question with Grok and Gemini yesterday.
Grok nailed it. Precise answer, current code examples, and even linked a solid repo.
Gemini’s reply felt generic and leaned on tools that are already a bit behind.
Still processing the fact that the company with the most data in the world is getting outpaced like this.
@iFood Abri o app e recebi a telinha que recebi um “presente” uma promoção do clube Ifood anual e quando fui pagar apareceu “essa oferta não está disponível pra você”. Tá de sacanagem em?
Awesome news! The model is super impressive. For Brazilian Portuguese, it’s not quite at 100% yet—there’s still a noticeable European Portuguese and American accent in the outputs. I tested a voice clone and it turned out well, but not perfect. With a bit more fine-tuning for BR-PT, I’m sure it’ll get there!
@DenisShepelin Congrats, this looks awesome! Mine is arriving soon. Took a quick peek at your code and noticed the latest commits are for Gemini Live… drop a video later so we can see it in action!