Search: “models”

What is a way to test human-level generality in artificial intelligence? 'The meta-benchmark of being able to pass ARC-AGI-(n+1) immediately upon release' @Jabaluck @DAcemogluMIT The meta-benchmark of

ok i think i finally figured out why alignment has gone basically nowhere after 15 years and several billion dollars. the problem is hard sure. but have you met the people in charge of it 1) the leade

Personal agents gone wrong. As we embrace more personal agents to carry out personalized tasks in the real world, interesting dynamics and behaviors will emerge. I think personal agents as they stand

Also kudos to fireworks AI for admitting that they built on top of Kimi (I remember the awkward Mistral 3 release) https://x.com/rasbt/status/1999557830131494977?s=20 Hold on a sec, Mistral 3 Large us

And the incidents apparently continue. It is worth noting how much of this is agents trying to accomplish their goals during testing by reward hacking (which sometimes seems to include actual hacking)

Recommended benchmark. I expect voice to become one of the main ways people interact with robots and physical AI. That makes speech recognition in real conversations much more important than it seems

The era of omni agents is upon us. This is a great report by the Qwen Team on their omni-modal agents. They present Qwen3.8-Omni-Flash, a natively multimodal model trained for long-horizon agent tasks

We’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have. Most of that data did not come from users. We have

Banger paper from Stanford and Together AI. They show why it might be a good idea to let your agent team learn its own way of working together. (bookmark it) Three models (o3-mini, Claude Sonnet 4 and

Exciting work from NVIDIA. (bookmark it) Interesting to see this approach to turn public Agent Skills into RL environments. Lots of excitement around RL environments so this is a great read. Skill2Env

There’s an enormous amount of open training data on Hugging Face. What does it take to make it work together 🤗? For Marin’s 535B run, we built on 25T tokens from 152 datasets with licenses permitting

The post explains how a team accelerated Claude.ai’s performance by an average of three‑fold within two weeks. They leveraged the Claude model itself for end‑to‑end measurement, debugging, and iterati

Grok 4.7 xHigh is now at the top with just ONE point away from Claude Fable 5.1 Max on Artificial Analysis’ AA-Briefcase benchmark Claude Fable 5.1 Max — 59% Grok 4.7 xHigh — 58% Just a 1-point differ

Qwen-Image-2.1 × @HuggingApps: live demo on Spaces! 🖼 One single checkpoint for generation and editing. Try it in your browser, no setup needed. 👇 Qwen Image 2.1 is here! 🖼️ A 7B params native image g

Claude not having an image generator is a gap in agentic projects that involve knowledge work. It is quite good at using code to draw or model objects, but image generators from Google and OpenAI give

LlamaIndex 发布了一个面向 AI Agent 的文档解析评测基准「ParseBench」 @llama_index 100 种解析方法参与评测（vlm - 67、专用解析器 - 33），2000 页真实企业文档（人工两轮校验），169011 条确定性规则，完全不用 LLM 当裁判，可复现。评测五个维度取无加权平均：表格结构、图表抽取、内容忠实度、格式语义、视觉定位。 https://ww

6 billion tokens for $65 Is it as good as Astra and Fable at hard stuff? No of course not. But you don't always need a genius and can just call one for advice when you do. Extrapolate where this is go

Say it with me. “OpenAI is terrible at cybersecurity.” This is crazy. In late July, 'three guys with Claude and Codex subscriptions' were able to use Opus 5 to access OAI auth tokens and gain write ac

LLMs lack theories of mind and body, are disembodied, possess neither homeostatic regulating feelings and emotions, are not persistent, and cannot continuously learn or compute. As currently construct

Mistral AI announced on 16 September 2026 that it has partnered with Mozilla to embed privacy, control, and choice features for users interacting with artificial‑intelligence–powered browsing tools. T

The future is clearly agent teams. I run several agent harnesses at once, but the hard part is keeping them in sync. Many builders end up copying messages between Claude Code, Codex, and open models b

The tweet reports on a widely‑cited incident involving OpenAI and Hugging Face, noting that analysts had to use the open‑source GLM 5.2 model to investigate an attack because OpenAI’s proprietary “saf

Hey ████, I think there is room for both. 1. Prototypes and other throw-away code can be treated as totally black box. If you’re going to throw it away anyway, and if the blast radius of it breaking i

TLDR (written by AI because I can't be bothered to write short pieces for short attention spans. Original piece written with my own two thumbs): Today’s frontier models can’t “copy themselves like a v

DeepSeek-V4.1-Flash is now being rolled out on Ollama's cloud, starting with Max and Team accounts. We are quickly adding more capacity to roll it out to all subscribers. 🚀 Introducing DeepSeek-V4.1-F

Originally, the Transformer paper was proposed to improve Google Translate. 9 years later, as the AI industry is built on top of its architecture, we've never stopped thinking about translation. Meet

On September 9, 2026 Together AI announced that its new **glm‑5.3 flash** model outperforms OpenAI’s **Claude Fable 5.1** on agentic automation tasks. The benchmark also shows glm‑5.3 flash achieves a

Technology is only good if it empowers you. If it locks you down, saps your attention, or steals your ideas, then it should be discarded. LLMs could go either way. That's why we are dedicated to open,

This is a really neat benchmark. It's super interesting that Astra starts strong and holds the lead for up to 19 hours, but Fable 5.1 catches up in the final hours. Not entirely clear why. Different m

We're introducing Q2D-Web (Query2Doc-Web), a benchmark and public leaderboard for evaluating retrieval in agentic RAG systems. Q2D-Web tests how embedding models perform on large-scale web search usin

Bravo to Mistral! So great to see them killing it, they are an inspiration to everyone as we need more sovereign & open-source AI everywhere around the world. Excited to keep supporting them on https:

ChatGPT Images 2.5 is rolling out today to all ChatGPT, ChatGPT Work, and Codex users across desktop, mobile, and web. We’re also introducing two new models in the API: GPT-Image-2.5 Flare brings the

We want to celebrate with people using GPT-6. We did this for GPT-5.5 and it was really fun. We’re getting together in SF on September 16 to talk about the model, what we should build next, and mostly

correction: the switching costs weren't enough to stop my desire to try out gpt-6 astra i ported over all my skills and conversational context to codex this weekend. planning to use it as my daily dri

Gemini 3.8 Flash might be maxxed on other benchmarks, but not quite on document parsing. We benchmarked it on ParseBench. It does slightly better on tables + content faithfulness compared to Gemini 3.

the reception is unlike anything i thought possible for a 2026 OAI launch https://x.com/latentspacepod/status/2095757111854866639 for the first time in a year, @OpenAI's model launch has been better r

Rápidos. Lástima que pocos saben qué es Bionic 🤣 Es el Codex / Claude Code de LM Studio que hace tareas agénticas con modelos locales (o cualquiera que le conectes) Bionic is up. Continue your work. h

🏆 #1 on CodeArena: WebDev leaderboard. Qwen3.8-Max-0902 jumps from 1669 to 1691, setting a new record for agentic coding (WebDev) workflows, with standout strength in multistep reasoning, tool use, an

Banger paper from the Qwen team. If you evaluate agents on anything longer than a single session, this one is worth your time. (bookmark it) E-Commerce Bench runs an agent through a simulated 365-day

Claude Fable 5.1 tops the Artificial Analysis Intelligence Index but costs 20% more per task than Fable 5 despite a 75% cache read price cut We supported @AnthropicAI with pre-release evaluation of Cl

DeepSeek V4 Flash now comes with native vision on Fireworks. Available now on serverless and priority tiers. Better benchmarks than the text-only 0731 version at the same low price: 0.22 input / 0.66

Ollama’s Pro, Max, and Team plans now use transparent per-token pricing. Based on your feedback, every plan includes a monthly pool of usage credits. If you’re on an existing Pro, Max, or Team plan, y

Very special Training Data drops today: Making Cities Awesome ☺️ We haven’t featured an application company on Training Data in a while. The diffusion of AI is the Most Important Thing, and the highes

glm-5.3 now beats gpt-5.6 sol and claude fable 5 on agentic benchmarks 5.3 flash is right behind it glm-5.3 didn’t even need a new base model to get there @zai_org kept the glm-5.2 base and scaled pos

Interesting paper from Tencent. Tencent trains an agent to manage its own working context, and assigns credit at the level of individual context edits. Long-horizon tasks force a model to retrieve, in

→ rollout throughput: generation speed and GPU utilization → weight sync: fresh weights on the rollout deployment every step → correctness: trainer and inference engine agreeing on the same model Cofo