Luma just launched a multimodal image model that beats Google and OpenAI on benchmarks. Uni-1 is built on a decoder-only transformer trained on images, video, audio, language AND spatial reasoning. #1 in human preference Elo for Style & Editing 10–30% cheaper than
LLMS
-
Mistral Launches Voxtral Transcribe 2: Fast, Cheap, Open-Source
By
–
Mistral just upgraded Voxtral – and it's fast, cheap, and open. Voxtral Transcribe 2 comes in two flavors: • Mini – batch transcription, speaker diarization, word-level timestamps, 13 languages • Realtime – sub-200ms latency, open-source (Apache 2.0) Priced at
-
9 New AI Tools This Week Including Mistral Anthropic Google
By
–
9 AI tools just dropped this week — and one massive shutdown you need to know about. Here's everything that happened in AI this week • Mistral Voxtral Transcribe • Anthropic Computer Use • Luma Uni-1 • Google Gemini 3.1 Flash Live • Google Gemini Advantage
-

OPUS: Intelligent Data Selection for LLM Pre-training
By
–
Is there a smarter way to pick data for training Large Language Models? Researchers from multiple institutions, led by Shaobo Wang, introduce OPUS. This novel method dynamically and intelligently selects the most impactful data for LLM pre-training in every single training
-
Extended GPT 5.2 Sessions and Benchmark Effectiveness Analysis
By
–
I very rarely see a benefit in making such a large block of work. I did had codex sessions going on over 24h back with gpt 5.2 but IMO wrong benchmark
-
The Agentic AI Bible: Complete Guide to LLM-Powered Agents
By
–
Check this out at https://t.co/yb8splH5S3
— Kirk Borne (@KirkDBorne) 30 mars 2026
"The Agentic AI Bible: The Complete and Up-to-Date Guide to Design, Build, and Scale Goal-Driven, LLM-Powered Agents that Think, Execute and Evolve" pic.twitter.com/KKnT16NzXFCheck this out at http://
amzn.to/4sUCgzc "The Agentic AI Bible: The Complete and Up-to-Date Guide to Design, Build, and Scale Goal-Driven, LLM-Powered Agents that Think, Execute and Evolve" -

Hundred-Page Language Models Book with PyTorch Implementation
By
–
The Hundred-Page Language Models Book — Hands-on with PyTorch: https://
amzn.to/4sJl7YC by @burkov -
Powerful Claude Code Features: /loop and /schedule for Automation
By
–
3/ Two of the most powerful features in Claude Code: /loop and /schedule Use these to schedule Claude to run automatically at a set interval, for up to a week at a time. I have a bunch of loops running locally: – /loop 5m /babysit, to auto-address code review, auto-rebase, and
-

Top AI Research Papers of the Week (March 23-29)
By
–
Here are the top AI Papers of the week (March 23 – 29) 1. Meta AI's V-JEPA 2.1
2. LeWorldModel
3. TurboQuant
4. ARC-AGI-3
5. Claudini More details on each paper below.

