Rafael Lobitsky

Building — an AI born in Brazil| Turning ideas into intelligent automation | Exploring the next generation of ERP + AI integration.

[ViggleAI @ViggleAI]

Introducing our new Text-to-Motion model, built in-house at Viggle, now live in @Viggle_PINOC!

Type a motion, get the animation. High-energy, expressive action is where it shines.

Block shots, prototype gameplay, fill your world with NPCs that actually move. Save your keyframes for the hero shots.

Try it for free on PINOC👇

[fal @fal]

LTX-2.3 Reframe is now live on fal, from @ltx_io.
Convert any video to a new aspect ratio

Landscape to vertical, square, and back: 16:9, 9:16, 1:1, 4:5, 5:4
Newly exposed areas are generated to match the original scene
Up to 60 seconds of input, 720p and 1080p outputs

[Rafael Lobitsky @RLobitsky_]

@arena @sseraphini @Kimi_Moonshot Is it?

[Rafael Lobitsky @RLobitsky_]

@Kimi_Moonshot Is it?

[Kimi.ai @Kimi_Moonshot]

🤯

[Superman @thesupermanmx]

SAM ALTMAN HAS LOST IT.

Google just shrunk 31GB of AI memory down to 4GB.

They open-sourced a vector index that fits 10 million documents in 4GB of RAM and searches them faster than FAISS.

→ 16x compression per vector
→ 10-19% faster than FAISS
→ Zero training. Zero rebuilds.

No GPU cluster. No cloud bill. No compromise.

Runs 100% locally.

[智东西China AI News @Chinazhidx]

🚀JUST-IN: Alibaba releases Qwen-Audio-3.0-Realtime(Plus/Flash)

Millisecond-level response, proactive tool use, adaptive emotion, and natural turn-taking

Hits SOTA on VStyle, while its preview topped Artificial Analysis speech reasoning & dialogue fluency rankings (97.6%/97.8%)

[MOSI @MosiAI_Official]

🤗 MOSS-VL-Realtime is now open source on @huggingface .

Built for real-time visual understanding over continuous video streams:
🧠 11B vision-language model
📜 Apache-2.0 license
💬 Ask questions at any point in a video stream
👀 Keeps watching while generating a response
🔄 Revises or interrupts its response as the scene changes
🤫 Can remain silent when more evidence is needed
🧩 256K-token context window
🌐 Chinese and English multimodal understanding
📦 Base, Instruct, and Realtime models are all open source

From “watch first, answer later” to “keep watching while answering.” 👀 @Open_MOSS

Thank you @sgl_project @lmsysorg for day-0 support! 🚀

[Wan @Alibaba_Wan]

Wan-Dancer-14B is now open-source!

You can now generate long-duration rhythmic dance videos locally.

Grab the model here👉
https://t.co/cQcaazK6qH

[Leonard Rodman @RodmanAi]

6 GitHub repos every AI engineer needs.

1. AI Website Cloner Template
Clone and reverse-engineer almost any website using reusable AI workflows.
https://t.co/Hntwa3ZZ5e

2. Recordly
An open-source alternative to Screen Studio with auto-zoom, smooth cursor movement, and polished screen recordings.
https://t.co/uBgnzyaKtN

3. Agent-Reach
Give your AI agent access to X, Reddit, YouTube, and GitHub—without paying for APIs.
https://t.co/oyMKmAo0kW

4. SkillClaw
Your AI automatically learns from every session and keeps your skill library organized and improving over time.
https://t.co/sXY5zmpYbU

5. Open Generative AI
Automate image, video, and media generation using AI coding agents.
https://t.co/YEuWCB4lhG

6. Firecrawl
Convert any website into clean Markdown and structured data that AI agents can actually use.
https://t.co/xsjMmveGm2

⭐ Which repo are you starring first?

[Google AI Developers @googleaidevs]

Rather than translating text step-by-step, Gemini 3.5 Live Translate processes raw audio in real time. This eliminates lag and ensures every nuance of the speaker's natural voice is preserved.

See how developers are building impressive multilingual apps with it ⬇️

[How To Prompt @HowToPrompt__]

Google has changed the game.

They’ve open-sourced an always-on AI agent that runs 24/7 and never forgets anything you feed it.

drop ANY file into a folder, pdf, image, audio, video, and it remembers it forever, links it to every other memory, and writes new insights on its own while idle.

no rag. no vector db. no forgetting.

100% open source.

[roman @Nozelcode]

HAN CLONADO CLAUDE DESIGN Y PUEDES USARLO GRATIS Y SIN LÍMITES

Se llama Open Design, un proyecto open source que te deja usar Claude para workflows de diseño sin pagar.

Sin suscripciones.
Sin límites (como la versión oficial).
Acceso total.

Esto es lo que puedes hacer:

→ Generar diseños UI/UX con Claude
→ Convertir prompts en diseños reales
→ Sustituir herramientas de diseño caras en muchos casos
→ Personalizarlo completamente (es open source)

Está hecho para devs, indie hackers y creadores que no quieren quemar dinero en herramientas.

De esos repos que pasan desapercibidos hasta que de repente todo el mundo los usa.

Si usas AI + diseño, tienes que probarlo

Enlace abajo👇

(guárdalo antes de que explote)

[Browser Use @browser_use]

Browser Use v4 is now available via API. ⚡️

Automate real work with a single request.

A full agent, working for you in Browser Use Cloud:
> Browsers with logged-in profiles, proxies, live view
> Environments reusable across runs via workspaces
> No scripts, no selectors, no hassle

Send a task. Watch it get done.
Try it now ↓

[Miguel Ángel @Miguel07Code]

Skill made to make the best hyperframes motion graphics, try it out!

[Vitor Darela @vitordarela]

Seu concorrente aparece no Google e no ChatGPT. Você não.
Não é sorte. Ele rankeia para palavras-chave que você nem sabe que existem.
O RankLayer mostra quais são, monitora o concorrente e gera artigos pra você competir por esse tráfego.
Enquanto você posta, ele captura.

[مالك بن حمود المالك @cnemalek]

🚨 عاجل: ChatGPT صار يقدر يساعدك في صناعة المحتوى باحتراف.

تقدر تستخدمه لإنشاء محتوى بجودة صانع محتوى محترف… مجانًا.

وهذي 7 برومبتات لازم تجربها إذا تبي تطور محتواك. 👇

[Alex Rivas | IA @alexconia]

🚨 Deja de decirle a Claude: "haz esto."

Deja de pedirle: "escribe este código."

Deja de decir: "corrige este error."

Estás tratando a una IA de nivel senior como si fuera un becario.

Aquí tienes 18 prompts para copiar y pegar directamente:

[IA João @MartinsJoa_]

🚨 BREAKING: O Claude agora constrói um canal do YouTube do zero e chega à monetização em apenas 30 dias.

100% gratuito.

7 prompts para fazer isso acontecer:

[Models Lab @ModelsLabAI]

Still using Suno AI for music in your app?

There's a developer-first API built for exactly this.

→ Generate up to 8 minutes of music from a text prompt
→ Any genre — cinematic, lo-fi, EDM, tropical house, vocal
→ Control BPM, key, and mood via API params

Start Building Now