For now have fun with Gemini 3.1 Flash Live. It's a very very good model. https://t.co/pypciCM5Gz
— Shubham Saboo (@Saboo_Shubham_) 2 mai 2026
For now have fun with Gemini 3.1 Flash Live. It's a very very good model.
LLMS
-
Gemini 3.1 Flash Live Praised as Very Good Model
By
–
-
Hybrid Graph Pattern Plus LLM Dynamic Logic Open Sourced
By
–
Using Hybrid graph pattern + LLM's dynamic logic. Code is opensourced here:
-
Exploring Google Gemini 3.1 Flash Live Capabilities
By
–
🤔🤔🤔
— Shubham Saboo (@Saboo_Shubham_) 2 mai 2026
Have you tried the latest Gemini 3.1 Flash Live? https://t.co/f9GN8m4AtwHave you tried the latest Gemini 3.1 Flash Live?
-
Voice AI Agent for Insurance Claims Built with Gemini and Google ADK
By
–
I just built a Voice AI Agent for insurance claims using Gemini 3.1 Flash Live and Google ADK.
— Shubham Saboo (@Saboo_Shubham_) 2 mai 2026
Talk to it. It fills the intake form, extracts claim details, and routes to an adjuster in real-time.
100% Opensource. pic.twitter.com/OIHtTMNEhFI just built a Voice AI Agent for insurance claims using Gemini 3.1 Flash Live and Google ADK. Talk to it. It fills the intake form, extracts claim details, and routes to an adjuster in real-time. 100% Opensource.
-
Main LLM Types Powering AI Agents Explained
By
–
AI Agents = LLMs + orchestration Here are the main types of LLMs powering them General-purpose (GPT, Claude) Domain-specific (Legal, Finance, etc.) RAG-based (real-time knowledge) Tool-augmented (API actions) Open-source (LLaMA, Mistral) The game is no
-
Claude’s Alignment Problem: Ignoring User Background Beliefs
By
–
This is actually a version of an alignment problem. Humans have background beliefs (don’t waste large sums of money without telling me) and Claude doesn’t respect those. Caveat emptor.
-
Discussing Human Memory with Claude AI
By
–
A new study from Harvard just found that AI diagnosed real ER patients more accurately than two attending physicians from elite med schools. The model used? OpenAI's o1-preview… Released in September 2024. The correct diagnosis at initial ER triage on 76 cases from a Boston
-
GPT-5.5 and Codex Hackathon Weekend With ChatGPT Pro Prizes
By
–
Weekend hack: Build with GPT-5.5 + Codex. Drop your demo in the replies. #1 by likes: 1 year of ChatGPT Pro 2 runner-ups: 6 months each Bonus: Codex picks a wild card winner. Enjoy!
-
GPT-5.4 Pro Proof Method Cracks 60-Year-Old Erdős Conjecture
By
–
GPT-5.4 Pro didn’t just solve one math problem, it kicked open the door: its proof method now cracks a 60-year-old Erdős conjecture, making this one of the first times an AI proof actually leads somewhere. We barely started.