SadBeaver (@sadbeaversol) 's Twitter Profile
SadBeaver

@sadbeaversol

I dun wanna be horny anymore - bonk

ID: 1685087864239841281

calendar_today29-07-2023 00:40:31

1,1K Tweet

1,1K Followers

838 Following

Sumit Doriya (@sumitdoriya21) 's Twitter Profile Photo

99% of the AI agent tutorials on YouTube are garbage. I’ve built 47 agents with n8n and Claude. Here are the 3 prompts that actually work (and make agent-building simple). Bookmark this post 🔖 Bonus: comment "Agent: and I’ll DM you AI agent system prompt + full guide ↓

99% of the AI agent tutorials on YouTube are garbage.

I’ve built 47 agents with n8n and Claude.

Here are the 3 prompts that actually work (and make agent-building simple).

Bookmark this post
 🔖 
Bonus: comment "Agent: and I’ll DM you AI agent system prompt + full guide ↓
Rohan Paul (@rohanpaul_ai) 's Twitter Profile Photo

This paper tweaks a looped Transformer so a small model gets better at multi-step puzzle reasoning without getting bigger. It reports 53.8% correct on its 1st try on ARC-AGI1 and 16.0% on ARC-AGI2, both pattern puzzles. The authors argue most prior gains came from recurrence,

This paper tweaks a looped Transformer so a small model gets better at multi-step puzzle reasoning without getting bigger.

It reports 53.8% correct on its 1st try on ARC-AGI1 and 16.0% on ARC-AGI2, both pattern puzzles.

The authors argue most prior gains came from recurrence,
Zheng Yuan (@ganjinzero) 's Twitter Profile Photo

Excited to announce Seed-Prover 1.5 which is trained via large-scale agentic RL with Lean. It proved 580/660 Putnam problems and proved 11/12 in Putnam 2025 within 9 hours. Check details at github.com/ByteDance-Seed…. We will work on autoformalize towards contributing to real math!

Excited to announce Seed-Prover 1.5 which is trained via large-scale agentic RL with Lean. It proved 580/660 Putnam problems and proved 11/12 in Putnam 2025 within 9 hours. Check details at github.com/ByteDance-Seed…. We will work on autoformalize towards contributing to real math!
Lovis Odin (@odinlovis) 's Twitter Profile Photo

For Official Wan 2.6 release I was invited to create the showcase . I wanted also to show people how to combine different api in fal for their own project as well. So, I Built an automatic pipeline with fal : → Voice synthesis (MiniMax) → Character generation

⚡AI Search⚡ (@aisearchio) 's Twitter Profile Photo

Z AI drops a realtime video generator for talking characters. Free & open source. Simply add a script, upload a voice & character, and it renders the video almost instantly. z.ai/blog/realvideo Btw, Z.ai is also behind one of the best open models, GLM 🔥🔥🔥

howie.serious (@howie_serious) 's Twitter Profile Photo

gemini 网页版已经打通 notebooklm 了。 在对话里选用,直接把 notebooklm 作为数据源。 对话和 gem 都可用。例如可以针对你的 notebook 调用“信息图大师”gem 来制作信息图。 加上今天 notebooklm 底层模型升级到 gemini 3,notebooklm 的价值越来越大。

Kshitij Mishra (@kshitijmis37062) 's Twitter Profile Photo

120 Mind-Blowing AI Tools; 1. Ideas - YOU - Claude - ChatGPT - Perplexity - Bing Chat 2. Presentation - Prezi - Pitch - PopAi - Slides AI - Slidebean 3. Website - Dora - Wegic - 10Web - Framer - Durable 4. Writing - Rytr - Jasper - Copy AI - Textblaze -

120 Mind-Blowing AI Tools;

1. Ideas

- YOU
- Claude
- ChatGPT
- Perplexity
- Bing Chat

2. Presentation

- Prezi 
- Pitch 
- PopAi 
- Slides AI 
- Slidebean 

3. Website

- Dora 
- Wegic 
- 10Web 
- Framer 
- Durable 

4. Writing 

- Rytr 
- Jasper 
- Copy AI 
- Textblaze 
-
BLACKBOX AI (@aiblckbx) 's Twitter Profile Photo

BLACKBOX CLI: Agents compete, judge picks the best! Claude, Blackbox, Codex, Gemini battle it out on each task. With Github repo or without github repo required. Superior results for coding tasks, ML, datascience, and more.

Avi Chawla (@_avichawla) 's Twitter Profile Photo

Deploy and run LLMs directly on your phone! Unsloth now lets you fine-tune LLMs and deploy them 100% locally on iOS/Android devices. The video shows this in action, where I ran Qwen3 on an iPhone 17 Pro at ~25 tokens/s. I have shared a guide in the replies.

Benjamin De Kraker (@benjamindekr) 's Twitter Profile Photo

Claude Code CLI running in 3 different terminals -One is actively building a grid-based futuristic city game -Two is improving an Android app (and taking screenshots of the app on my phone, over USB) -Three is building a web dashboard and using Playwright headless to auto-fix UI

Claude Code CLI running in 3 different terminals

-One is actively building a grid-based futuristic city game
-Two is improving an Android app (and taking screenshots of the app on my phone, over USB)
-Three is building a web dashboard and using Playwright headless to auto-fix UI
David (@david_eficaz) 's Twitter Profile Photo

🧑‍🎓 Te enseño la mejor manera de dominar Nano Banana Pro. Sube cualquier imagen a Gemini y pídele que la convierta en un mensaje JSON, incluyendo el tamaño y los detalles. A continuación, se mostrará un mensaje para reproducir la imagen, de modo que puedas modificar solo las

🧑‍🎓 Te enseño la mejor manera de dominar Nano Banana Pro.

Sube cualquier imagen a Gemini y pídele que la convierta en un mensaje JSON, incluyendo el tamaño y los detalles.

A continuación, se mostrará un mensaje para reproducir la imagen, de modo que puedas modificar solo las
Andrew Bolis (@andrewbolis) 's Twitter Profile Photo

OpenAI just released 11 free prompt courses. Beginner to advanced levels (I included all links): Use these courses to master ChatGPT prompts. [ bookmark 🔖 this thread for later ]

Prompt Mühendisi (@promptmuhendisi) 's Twitter Profile Photo

steal my secret viral slideshow prompt. I made this with ImagineArt: { "subject": { "description": "Young woman in car passenger seat at night, cozy tired energy, that 'drive me home' vibe", "age": "early 20s", "expression": { "eyes": "soft, slightly

steal my secret viral slideshow prompt. I made this with <a href="/ImagineArt_X/">ImagineArt</a>:

{
  "subject": {
    "description": "Young woman in car passenger seat at night, cozy tired energy, that 'drive me home' vibe",
    "age": "early 20s",
    
    "expression": {
      "eyes": "soft, slightly
fal (@fal) 's Twitter Profile Photo

🚨 Flux 2 Flash & Flux 2 Turbo are now live on fal! ⚡ Timestep-distilled Flux models - fastest in the world 🎨 Quality that matches or beats the base version 🚀 Sub-1 second generation

🚨 Flux 2 Flash &amp; Flux 2 Turbo are now live on fal!
⚡ Timestep-distilled Flux models - fastest in the world
🎨 Quality that matches or beats the base version
🚀 Sub-1 second generation
jtydhr88 (@jtydhr88) 's Twitter Profile Photo

ComfyUI native 3d node supports 3dgs and ply format. Integrated spark , developed by World Labs. Next we have ability to connect World labs Marble, or Tencent HunyuanWorld, or any other world model in ComfyUI. #ComfyUI

Teslaconomics (@teslaconomics) 's Twitter Profile Photo

During the SF power outage, Waymos shut down, but Tesla Robotaxis kept driving. When traffic lights went dark, Waymo vehicles relied on infrastructure and signals that were no longer there, so they all stopped for safety. Tesla Robotaxis did NOT. Tesla’s autonomy is built to

Robert Scoble (@scobleizer) 's Twitter Profile Photo

From Grok. I gave it a lengthy post teaching it about my lists and asked it to find the latest news. Does this provide any value to you? If it does I will run it a couple times a day. +++++ Top-Level xAI List Report: Key Insights from December 22, 2025 Overview: Across the

JumboBlockchain | Mainnet is Live! (@jumboblockchain) 's Twitter Profile Photo

5️⃣ AI-Contract Copilots Auto-generate smart contracts from natural language 💬 “Create a contract that unlocks salary on 30th if KPI met” → Done