perhaps we should compare calculators to humans? also, if you actually follow my work (eg October 2025 NYT oped) you will know I am a fan of domain specific AI (like Waymo) and skeptical of domain-general AI (like chatbots), so I appreciate your making my case for me.
GENERATIVE AI
-
Converting 30k arXiv papers to Markdown using SOTA OCR
By
–
New blog post: converting 30k @arxiv papers to Markdown using SOTA OCR models to enable chat with paper functionality
— Niels Rogge (@NielsRogge) 7 avril 2026
Includes:
> leveraging an open OCR model (Chandra 2 by @datalabto)
> running on GPU infra – @huggingface Jobs
> using Codex with a SKILL.md pic.twitter.com/jrpin9oq5uNew blog post: converting 30k @arxiv papers to Markdown using SOTA OCR models to enable chat with paper functionality Includes: > leveraging an open OCR model (Chandra 2 by @datalabto) > running on GPU infra – @huggingface Jobs > using Codex with a SKILL.md
-
Hugging Face Storage: Feedback on New Cloud Storage Solution
By
–
did you try https://
huggingface.co/storage? would love to hear your feedback -
Two New Marble Model Updates Released for Scale and Quality
By
–
We're excited to be rolling out two model updates today!
— World Labs (@theworldlabs) 7 avril 2026
Marble 1.1: Improves lighting and contrast, with a major reduction in visual artifacts.
Marble 1.1-Plus: Our new model built for scale. Create larger, more complex environments than ever before. pic.twitter.com/pslqVXNqFaWe're excited to be rolling out two model updates today! Marble 1.1: Improves lighting and contrast, with a major reduction in visual artifacts. Marble 1.1-Plus: Our new model built for scale. Create larger, more complex environments than ever before.
→ View original post on X — @scobleizer, 2026-04-07 16:32 UTC
-
GLM-5.1: Revolutionary Open-Source Agentic Coding Model Released
By
–
Another big release: GLM-5.1! China is on fire! significant increase in evals compared to GLM-5.0 tl;dr GLM-5.1 is the new open-source agentic coding model that significantly outperforms its predecessor by sustaining long-horizon problem-solving over hundreds of iterations, continuously improving results instead of plateauing, achieving state-of-the-art performance on complex software engineering benchmarks. Z.ai (@Zai_org) Introducing GLM-5.1: The Next Level of Open Source – Top-Tier Performance: #1 in open source and #3 globally across SWE-Bench Pro, Terminal-Bench, and NL2Repo. – Built for Long-Horizon Tasks: Runs autonomously for 8 hours, refining strategies through thousands of iterations. Blog: z.ai/blog/glm-5.1 Weights: huggingface.co/zai-org/GLM-5… API: docs.z.ai/guides/llm/glm-5.1 Coding Plan: z.ai/subscribe Coming to chat.z.ai in the next few days. — https://nitter.net/Zai_org/status/2041550153354519022#m
→ View original post on X — @kimmonismus, 2026-04-07 16:27 UTC
-
Embeddings: The Unsung Hero Driving Model Accuracy Forward
By
–
Embedding is an unsung hero in model accuracy and today is a big leap forward. It's at the heart of grounding; it's the layer that does the hard work of searching, retrieving, organizing, and connecting information across sources for a holistic response.
-
Harrier Upgrades Bing Web Grounding for Agentic AI Era
By
–
Bing's web grounding already powers almost every major AI chatbot today. With Harrier, it just got a big upgrade for the agentic era. Better embeddings lead to better retrieval, often more accurate answers, and better multilingual performance in the 100+ languages Harrier
-
Microsoft Open-Sources Industry-Leading Embedding Model Harrier
By
–
Big kudos to @JordiRib1 and the @bing team – awesome to see the speed and quality shipping across @MicrosoftAI
. More on Harrier in today's blog: https://
blogs.bing.com/search/April-2
026/Microsoft-Open-Sources-Industry-Leading-Embedding-Model
… -
Rocket 1.0 Eliminates Research Phase by Connecting Strategy to Development
By
–
If the system performs as demonstrated, it could eliminate the weeks typically spent on research and strategic planning before development even begins.
— Chubby♨️ (@kimmonismus) 7 avril 2026
In practice, this preparatory phase is where the majority of time is currently invested. Really nice! https://t.co/xF80r9iI1DIf the system performs as demonstrated, it could eliminate the weeks typically spent on research and strategic planning before development even begins. In practice, this preparatory phase is where the majority of time is currently invested. Really nice! Vishal Virani (@Vishalvirani91) Rocket 1.0 is live. This is our first major step toward Vibe Solutioning. Vibe coding solved how to build. It never solved what to build, or why. That's the harder problem and the one where most products actually fail. @rocketdotnew connects the thinking and the building in one platform. Solve your hardest business question. Build from what you solved. Watch your competition while you work. Everything shares one context. Nothing resets between sessions. The video and blog explain it better than I can here. — https://nitter.net/Vishalvirani91/status/2041546557342855363#m
→ View original post on X — @kimmonismus, 2026-04-07 16:19 UTC
-
GLM-5.1: Open-Source AI Tops Coding Benchmarks
By
–
BREAKING : Z AI released GLM-5.1, an open-source model with top tier coding performance! “Number 1 in open source and number 3 globally across SWE-Bench Pro, Terminal-Bench, and NL2Repo.” “Runs autonomously for 8 hours, refining strategies through thousands of iterations.”