Grok 자율 작업, 에이전트 평가, 오픈 모델 확산과 Hugging Face 매각 가능성

Elon Musk

@elonmusk

6일 전

Grok @Bot works 24/7

Very interesting new paper from NVIDIA. (bookmark it) It takes a closer look at evaluating agent skills. Enterprise teams are starting to leverage shared skill libraries, and the review gate is typically a scanner that checks structure, style, and security. NVIDIA measured whether that gate predicts anything. Across 145 real skills from internal and public catalogs, structural scan scores correlate with LLM-judge quality at a Spearman rho of 0.14. ACES proposes Skill Lift instead. In other words, run the same task twice under the same model, sandbox, workspace, and scorer, once with the skill loaded and once without. Then you measure the difference in what the agent completed. They scored 947 paired cases from 58 production skills across four harnesses, normalizing trajectories into a shared Agent Trajectory Interchange Format, so results compare across harnesses. They fins that the largest process-metric gains appear in skill execution, behavior check, and skill efficiency. Paper: https:// arxiv.org/abs/2608.20614 Track more trending AI papers in our academy: https:// academy.dair.ai

OpenCode

@opencode

6일 전

LongCat-2.0 now available in Go 1.6T/48B · 1M context · fully open source

Wondering how to get started with Shieldstral? Check out our cookbook to get the basics on using this open model for policy-based moderation https:// mistr.al/4xQmxTk

Wan 3.0 is now live on @RunningHub_ai Native 30-second generation, multimodal references, and stronger temporal consistency. Available on RunningHub via Web UI and API. More API access： • Model Studio: https:// click.alibabacloud.com/m/20000002017/ • Qwen Cloud: https:// click.qwencloud.com/m/20000002025/

Wan 3.0 is now on @LeonardoAi ! Sharper. More controllable. More capable. Now available on Leonardo. More API access： • Model Studio: https:// click.alibabacloud.com/m/20000002017/ • Qwen Cloud: https:// click.qwencloud.com/m/20000002025/

You can't pretend you care about security if you are still sharing API tokens in .env files or sharing SSH keys. This might be good enough to provide simple access to a resource, but it's a horrible way to do security when it really matters. Look into "Identity-Based Access". The state of the art today is Identity-Based Access. With traditional security, you have to present a key to get access to the resource (an API key, a password, or an SSH key). These are long-lived keys you need to keep secret somehow. If somebody steals your key, they become you and get the same access you had. With Identity-Based Access, you need to prove who you are instead. If you are a human, you can prove this with a passkey, SSO with Google or GitHub, or Multi-Factor Authentication. If you are a computer, you can use AWS IAM, Kubernetes, a cloud service account, or any other trusted environment. Once the system verifies your identity, it checks your permissions and gives you a short-lived credential. Prove identity → Permissions → Short-lived credential → Access The difference here is huge, because those credentials are: • Short-lived • Tied to a specific identity • Scoped to specific permissions • Automatically expired • Much easier to audit In practice, implementing this is as simple as integrating with @goteleport , which will handle everything for you. Here is a full explanation of how Teleport works and how you can start using it: https:// fandf.co/4u1Bm3R Thanks to the Teleport team for partnering with me on this post.

@Observe_Inc removed the LLM middleman from its MCP server. GA is rolling out now, with a CLI at full parity. Your agent queries the context graph, picks the dataset, and writes the OPAL itself, against the same REST APIs that power the Observe UI. Less latency, no Observe-side inference cost. https:// bit.ly/4gKuC6b

Hugging Face has reportedly been fielding acquisition offers that would value the company at around $13B. But with the founders' feeling of responsibility to community, doubts arise as to whether a sale will happen.

Hugging Face is exploring a sale that could value the AI model-hosting platform at $13 billion or more, per Business Insider, nearly triple its $4.5 billion valuation from a 2023 round backed by Google, Amazon, and NVIDIA. The company is working with a bank to gauge bidder interest, and no buyer has been named.

New blog post: continuous diffusion for language is back! This research direction receded into the background for a while, but as of this year, it is once again a hot topic. I wrote down a historical perspective and some thoughts on the recent revival. https:// sander.ai/2026/08/24/con tinuous-dlms.html?v=1 …