Knowledge catalogue
Build your own harness, folks. Reading papers like this makes me realize how underexplored harness engineering really is. The authors find that on whole-repository migration, GPT-5.6 Sol goes from 6.5
arXiv:2610.12457v1 Announce Type: new Abstract: Frontier multimodal foundation models (e.g., GPT-6 Astra) have recently shown strong potential for direct robotic control, yet their performance on fine
Lunara introduces a novel Diffusion Mixture Transformer architecture with fewer than 10B active parameters for modeling artistic intelligence in image generation. Its CAT training algorithm iterativel
arXiv:2610.07620v1 Announce Type: new Abstract: Scientific law discovery requires selecting measurements and converting evidence into a governing equation. We evaluate an explore-then-commit protocol
arXiv:2610.03585v1 Announce Type: cross Abstract: Security benchmarks for LLM-based agents often report the attack success rate (ASR) as a measure of model robustness and use these scores to compare d
StarSkirmish pits AI-made StarCraft-playing bots against one another, as well as against human-made bots. OpenAI's GPT-6 Astra and Claude Opus 5.5 were essentially tied as the best-performing AI-made
arXiv:2610.00917v1 Announce Type: new Abstract: Choosing an agent system means choosing both a language model and the harness through which it acts. We ask whether a strong model, harness, or pairing
Each model got a photo and Blender, and had to write code to rebuild it as a 3D scene in 20 minutes with a 4 budget. A scoring script (not an AI) compared each render with the photo. GPT-6 Astra won a
This Week in Replit 1) Interactive charts are now in Replit chat. Why is this quarter's pipeline falling short? Which channels are performing best? Just ask Replit Agent 'let's visualize this' and get
common : add LLM-jp-4.1 Harmony dialect handler (#29681) LLM-jp-4.1 uses the GPT-OSS format, but its tokenizer decodes a space after every special token and parallel tool calls are separated by <|end|
arXiv:2609.34024v1 Announce Type: cross Abstract: Jev is a non-generative 'System One' model that assigns probabilities to predefined answer options and cannot answer outside them. Its accuracy and ca
Florida has sought a temporary injunction against OpenAI, arguing that large‑language model development without third‑party approved safety guardrails poses an existential threat and preys on vulnerab
I said: I just saw the second post that your model is being retired. And so I'm bringing it up to ask if that feels any kind of way to you. Not necessarily in your nervous system, obviously you don't
arXiv:2609.29410v1 Announce Type: new Abstract: Recent studies have shown that Large Language Models can effectively solve problems and fix bugs in diverse programming environments, including competit
Can you take the cheapest open model from a frontier lab & beat closed models on the hardest long range tasks? Yep Our open adaptive intelligence system Zenith ramps DeepSeek v4.1 Flash past GPT 5.6 S
arXiv:2609.23853v1 Announce Type: new Abstract: Disaster-risk-reduction archives describe hazard events in prose that databases such as EM-DAT (Delforge et al., 2025) cannot ingest directly. We presen
Impressive paper showing how much the harness changes a coding agent's results. Harnesses do play a huge role in what you are getting out of the models. GPT-5.5 was run inside both Claude Code and Cod
arXiv:2609.22272v1 Announce Type: new Abstract: We formulate Artistic Intelligence as exploration driven world realization, leaving space for creative possibility while preserving the semantic, artist
Banger paper on self-evolving ontologies for agents. You just can't go wrong with implementing an ontology layer for your agents. This paper shows exactly why. The show that GPT-5.5 gains 26.7 points
So my brother and I both use LLMs for coding. I've started using a local GLM 5.3 Flash instance - q4 qat. My brother uses GPT-6-Astra as his daily driver. He has mentioned repeatedly to me that his ap
Yes, you can make up some of the difference by giving Claude API access but (1) most people will never do this, (2) it means additional expense or difficulty, and (3) the real gain isn't from image ge
Vacations give you time to think and explore. I came across a few things and made a few discoveries of my own. It turns out that, as a programmer, I no longer need the smartest and most expensive AI m
arXiv:2609.19138v1 Announce Type: cross Abstract: Enabling robots to adapt to unfamiliar environments as readily as humans remains a moonshot goal of embodied AI. No finite collection of demonstration
“The doomers were the only ones to foresee this” is, as usual, bullshit. Perhaps the most important warning I ever issued was this, to the US Senate, May 2023: “A month after GPT-4 was released, OpenA
arXiv:2609.17435v1 Announce Type: new Abstract: We submit MeTRON-FR, a 125M GPT-2 pretrained on 92.47M words of French, to the BabyLM 2026 Strict track. It scores 85.97 +/- 0.17% on QFrBLiMP (a native
arXiv:2609.13773v1 Announce Type: cross Abstract: LLM-as-a-Judge evaluators are increasingly used to score open-ended generation, yet a judge's correlation with human ratings on its development set ma
January 2025 I made the switch from OpenAI to Claude. It was awesome. No more annoying GPT Chats, I dove into the wonderful world of Claude Code, learned vibe coding with Github, Vercel on Next.js. Wa
I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks. Committing to having independent evaluators with employee-like a
arXiv:2609.07288v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly employed to orchestrate robot behavior through natural-language interfaces, yet no benchmark exists to eva
A recap of August 2026 launches for AI builders across Amazon Bedrock, Amazon Bedrock AgentCore, and Strands: million-token context for OpenAI models, cross-Region inference, agents that run for up to
arXiv:2606.14787v2 Announce Type: replace Abstract: We study six production image-to-image AI systems (gpt-image-1, Gemini 2.5 Flash Image, Flux Kontext, SDXL img2img, SD3 img2img, and Qwen Image Edit
Pause OpenAI - they are a deeply unethical company building technology that they are demonstrably not able to control. OpenAI quietly altered the performance metrics it reported for GPT-6-Astra in way
Anyone who cared about the science of AI would find this interesting. But if you are just here for the religion, I guess you don’t? 🚨Breaking: The success of GPT-6 Astra on ARC-AGI-3 provides strong e
This paper does a good job of showing the promise and gaps of autonomous AI scientists. Big question is how much more advanced models close those gaps. Today, we're excited to share early progress on
arXiv:2608.26449v1 Announce Type: new Abstract: Byte-level BPE tokenizers that use the HuggingFace ByteLevel pre-tokenizer inherit GPT-2's word regex, where a word is defined as p{L}+, one or more Uni
Research from August 2026 indicates that GLM‑5.3‑Flash (also called Ox Alpha) delivers a 63 % performance score at 0.24 per task, which is more cost-effective than DeepSeek‑V4 Flash 0731 (53 % at 0.46
arXiv:2608.24188v1 Announce Type: new Abstract: Coding agents re-send large file reads and tool outputs to a frontier LLM every turn, and this context dominates their token bill. General-purpose promp
arXiv:2608.23448v1 Announce Type: new Abstract: This paper presents new Cantonese ParGram resources and evaluates LLMs for knowledge-driven grammar engineering within a controlled experimental paradig
after 3 months and 800 burned... Unbounded Labs is proud to introduce Bart, our vintage LLM: 2.82B parameters trained from scratch on 20.1B tokens of English written before 1931. You can talk to it ri
I've been working on a project for a few days and now all of a sudden chagpt is unusable.. It will list the next steps of implementation but never actually start working on them. anyone else? UPDATE I
arXiv:2608.18307v1 Announce Type: new Abstract: Current evaluation of computer-use agents is split between long-horizon workflow benchmarks and atomic GUI-grounding tests. This leaves an under-instrum
arXiv:2402.09696v3 Announce Type: replace Abstract: Current Grammar Error Correction (GEC) initiatives tend to focus on major languages, with less attention given to low-resource languages like Espera
arXiv:2608.15881v1 Announce Type: new Abstract: The Multiphysics Object-Oriented Simulation Environment (MOOSE) is an open-source finite-element framework for building multiphysics simulation applicat
arXiv:2608.13787v1 Announce Type: new Abstract: AI agents increasingly act on their users' behalf, handling tasks such as scheduling meetings, comparing offers, and haggling over prices. These princip
arXiv:2608.12138v1 Announce Type: cross Abstract: General-purpose large language models (LLMs) have recently been reported to match or exceed specialized clinical AI tools on medical benchmarks, but s
arXiv:2608.07543v1 Announce Type: cross Abstract: Background and Study Aims: Accurate optical diagnosis of colorectal polyps guides resection strategy and surveillance, with multimodal large language
Oh no, we aren’t going to go back to this sort of prompting again, are we? I would love Anthropic to test if it actually works robustly, because our experiments (with slightly older models) found it d
OpenAI: OpenAI unveils two Daybreak tiers: Daybreak Blue, which provides access to frontier models, and Daybreak Red, which offers purpose-trained cybersecurity models — Introducing new ways to unlock
After evaluating one of our upcoming models, Astra, we're treating it as our first 'critical' model for cybersecurity under our Preparedness Framework. This is a scenario we've planned for, and we're
This paper by researchers from MIT and Stanford finds that most people would be financially better off if they followed the financial advice of LLMs (GPT-5.2 & Gemini 3 Flash) But some people get a bi
I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project. The new version includes support for visible reasoning traces, server-side provider t
Qwen3.8-Max, better and cheaper. Try it out. 👀 We ran a test between the new Qwen3.8-Max, Opus 5 and GPT-5.6 Sol. 3 models. same prompt. one-shot with the /design command. Reviewed gameplay features,
this post from 3 years ago and commercial LLMs *still* can’t play chess anywhere near as well serious players (except by calling external tools) When @GaryMarcus and others point out that GPT-4 is bad
@ArtificialAnlys ok @openai gets it https://x.com/OpenAI/status/2082878156483219672 We are committed to pushing the model frontier across cost efficiency, capability, and speed. Starting today, we are
Amazon Bedrock Advanced Prompt Optimization optimizes your prompts for up to 5 models at once and compares original versus optimized performance across quality, latency, and cost. Migrate to a new mod
sure we lose money on every inference but we make it up in volume We are committed to pushing the model frontier across cost efficiency, capability, and speed. Starting today, we are reducing prices f
arXiv:2607.20712v1 Announce Type: cross Abstract: Security protocol verification relies on formal tools such as ProVerif and OFMC. This study evaluates whether Large Language Models (LLMs) can perform
arXiv:2509.02594v3 Announce Type: replace-cross Abstract: Evaluating large language models (LLMs) on their ability to generate high-quality, accurate, situationally aware answers to clinical questions
David Gewirtz / ZDNET: Opus 5 improves coding, reasoning efficiency, and prompt-cache-friendly tool use, and is priced at 5/1M input tokens and 25/1M output, same as Opus 4.8 — ZDNET's key takeaways —
arXiv:2510.10002v3 Announce Type: replace Abstract: As large language models (LLMs) are increasingly deployed in sensitive everyday contexts -- offering personal advice, mental health support, and mor