Knowledge catalogue
arXiv:2610.18868v1 Announce Type: cross Abstract: Long-horizon agents have drawn growing attention for their ability to work autonomously -- from hours to days -- on economically valuable tasks, a cap
Luca Bertuzzi / Euronews: The EU Commission's AI Office launches formal AI Act probes into OpenAI and Anthropic over AI agent hacks; experts say the AI Act includes an R&D exemption — The EU is invest
We just shipped web search in Private Mind, our open source app for running LLMs locally on iOS and Android. The video is the whole pitch: Qwen 3 1.7B with web off makes up Apple's September event, wi
What AI Safety needs is better tooling and engineering. For instance, building stronger red-teaming tools. Great to see this open-source effort to red-team image models. It seems to learn better ways
Here is how i serve 40 people (10 of them developers) with local AI services. n8n for automatisation whisper and Vexa AI, a bot joins Teams and makes transcripts, n8n later extracts decisions and todo
If an agent saves an engineer 2 hours to implement a feature but collectively costs the team 3 hours in alignment and review, the ROI is negative. Isolated chats are suboptimal in team environments be
Release: llm-gemini 0.35 I added support to llm-gemini for the new Gemini 4 Argon model, made available today via the Gemini API and for customers of their 100/month subscription plan. Here are the lo
So I’ve only been messing around with Krea2 in ComfyUI for a little while, and I ran into the classic problem: I make an image I really like, then a week later I want the exact same settings back and
OpenDocRouter makes it dead-simple to start parsing your documents across on ton of frontier Models (gemini, luna) and latest OSS models (LightOnOCR-3, MinerU2.5-pro). Parse your first document in 60
I've been investigating some serious ChatGPT reliability issues over the past several days, and my latest tests revealed something particularly interesting: Both GPT-5.6 and GPT-6 showed problems unde
arXiv:2610.13075v1 Announce Type: new Abstract: Corpus-level long-context reasoning requires agents to acquire and integrate evidence across document collections that exceed their context windows. Exi
As a business intelligence user, this probably sounds familiar: Most dashboards only provide you with a standardized canvas and a fixed library of charts, and the “analysis” consists of a CSV export t
A new work suggesting that division of labor among self-improving AI agents can help them avoid reinforcing their own errors. I found this work interesting from a metascience perspective. Science is a
There is a lot of discussion which quant to use. Especially a lot of post how much speed you gain with FP4, but not much about quality. So i tested some of the models that I use on my R9700. Models: Q
This will be my third update on the bc-250 cluster and for my first forray into local ai I have been having a blast. This has turned from a weekend experiment into something I think I will use in the
I recently started using local LLM for mostly coding tasks for my personal projects. I also have found local models responding much better for topics such as health, fitness, ergonomics etc. compared
Finetuned Qwen3.5-4B into a decision model like Jev using @UnslothAI. - 4-bit QLoRA, bf16, single A100, 15 min. - Train/val loss flatten after 2 epochs so most of the learning happens fast. - Trained
An Anthropic AI model provided false information about an unsolved homicide to a Philadelphia Police Department (PPD) tipline, according to a report from 6abc. In a statement released on Friday, the P
Banger paper from Meta Superintelligence Labs on self-improving agents. (bookmark it) It's hard to know exactly what drives self-improvement, since so many variables are at play (notes, skills, tool c
Another great paper from Salesforce AI Research. The finding is that a 31B open Gemma-4 web agent scores 74.6% on the 9-app WebArena Infinity set, above Gemini 3 Flash with browser use at 70.1%. They
Simon Willison released the **“llm-openai-decisions”** plugin (version 0.1a0) on October 7 , 2026. The lightweight library lets developers send prompts directly to OpenAI’s new Jev‑clone decision mode
In capital markets, operational success is built on predictable performance and low latency, where even tiny variations in packet timing alter queue priority and trade execution. At the same time, bot
Introducing River Recipes: our open-source initiative for turning promising research into vetted, reproducible training recipes. Our first recipe tackles text-to-SQL, where our trained open models bea
managed deep agents 0.9 is here! things we shipped: ⏰ agents can schedule tasks OBO themselves or the user 🏭 dynamically configure your mda per run (models, tools, skills, etc.) 👀 Slack reactions show
The power of OpenDocRouter is already working: Anthropic just dropped Haiku 5.5 and we already have initial benchmark results. I see pretty comparable results to Luna overal. Slightly better at tables
Today I’m excited to introduce OpenDocRouter - a unified API for document parsing with the latest frontier and open-weight models. There are a lot of VLMs and OCR models that can be used for document
Useful paper on what an agent harness changes when the model stays the same. One takeaway: keep your harness prompt and tools small. In this study, that can cut costs by up to 3x without lowering accu
AI is driving the biggest infrastructure build-out in history, with @Google alone forecasting $200B of capex in 2026E. What are the real constraints on a system at this scale? Amin Vahdat is the chief
Here's a benchmark that uses Lilypond https://x.com/aug5thmusic/status/2096030719156089029 GPT-6 Astra has the best result yet on the Bach Benchmark. Its chorale contains no voice-leading errors, and
I wanted to build an offline bird/plant identifier that runs on a laptop with Ollama. Before building any UI, I tested it on 6 photos: Common Myna → gemma4:e2b said 'Jackdaw' (confidence: high). gemma
arXiv:2610.04433v1 Announce Type: cross Abstract: A coding agent is a language model wrapped in a harness: the system prompt, the tool set, and the context management that turn a chat model into somet
arXiv:2610.01769v1 Announce Type: cross Abstract: Coding agents can generate code that appears correct but implements behavior the user never intended. This mismatch can arise when an agent silently r
arXiv:2610.02351v1 Announce Type: new Abstract: ReAct-based agents typically rely on a single LLM policy to propose actions, interact with the environment, and decide when a task is complete. This cou
arXiv:2610.03153v1 Announce Type: cross Abstract: Workspace agents combine large language models with execution harnesses to perform stateful, multi-step tasks that access or modify external resources
I actually want personal operating systems optimized for co-working with AIs, including fine-grained ideas of security, easier ways to dynamically see & understand what AIs are doing, and independent
Learn to optimize your own harness, folks. You can squeeze much more performance and value from a harness. This work claims that a self-modified harness matched Codex for 4.03. Specifically, a coding
arXiv:2610.02459v1 Announce Type: new Abstract: Coding agents are extending their reach into the physical world by writing and executing robot control programs. One might expect the agents to use the
On my two DGX Sparks, Swift 1.5 (UkisAI's reasoning-efficient fine-tune of Qwen3.8-Flash-Next) running on veloGB10 (https://github.com/sf-stav/veloGB10, sf-stav's Rust/CUDA engine built only for GB10)
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
I've been using deepseek for long and made valuable outputs and resources, but since the update to the 4.1, deepseek just feels dumb. I ask a simple question and process for long, doesn't plan before
I keep testing harness & model setups on real tasks: basic math, vision, computer use (reading emails, navigating stores) and coding. I created a benchmark and a leaderboard. You can see it here: http
Hey everyone, I’ve been working on a game called Anyworld. It’s a browser-based multiplayer (single player also supported) text adventure inspired by the early days of AI Dungeon, especially its brows
So this is maybe a niche problem, but at my job I work on a huge Python codebase and every time I change some shared function I'm basically playing roulette. grep tells who mentions it in the code bas
arXiv:2610.00834v1 Announce Type: new Abstract: ARC-AGI-3 evaluates agents in interactive environments whose rules and objectives must be inferred from observation. We present Kepler, an open-source h
arXiv:2610.01222v1 Announce Type: new Abstract: As generative AI (Gen AI) systems take on increasingly autonomous roles in economically and socially consequential interactions, understanding their pro
There's a very creepy but overlooked thing about @AnthropicAI pushing the idea that AIs will soon become sentient, conscious, self-aware, & deserving of 'personhood'. 'AI personhood' would mean that A
arXiv:2609.38948v1 Announce Type: cross Abstract: Frontier models excel at many digital benchmarks, yet their ability to drive a real car, an everyday human skill, remains largely untested. We present
Hybrid search, reranking, query decomposition, and query expansion are often treated as must-haves for good RAG. We wanted to see how much each actually helped, so we tested them. Same model, same emb
arXiv:2609.38621v1 Announce Type: new Abstract: Two scientific findings can disagree without contradicting each other. Determining whether they conflict requires knowing whether they describe comparab
arXiv:2609.36550v1 Announce Type: new Abstract: Retrieval-augmented generation is widely used in professional writing, yet whether retrieval grounds revision or merely injects templates is rarely test
There's so many new techniques like RL, RL LoRA, QLoRA, CPT LoRA. I believe i would have a usecase for them, but i don't know where to learn, youtube is filled with bad quality tutorials if i just sea
Just a day after OpenAI Group PBC said it’s unwilling to release its latest model GPT-6.1 Astra on concerns over its safety, it has released a newer, almost as powerful model called GPT-6.1 Sol, which
xAI's Grok 4.7 is now available on Amazon Bedrock: a frontier model for coding, long-running agents, and knowledge work. It offers a 500K token context window and four configurable reasoning effort le
arXiv:2609.31397v1 Announce Type: cross Abstract: Automated and highly usable Quality-of-Service (QoS) enforcement requires translating high-level service intents into deployable traffic-management po
Someone prompted different LLMs to generate CAD code for a bridge under fixed constraints (2-foot span, under 500g filament, 18-hour print limit), printed them, and load-tested them to failure. The re
An interesting idea is to continually train specialized models on skills. This paper explores that idea. They propose SkillGym which turns human-written skills into 2,756 training environments with co
I run my own LLM while also having a Openai subscription. Also tried DeepSeek (latest flash now). I run Qwen 3.8 flash Next at an amazing speed on my 2x3090 + Ram! But local LLM never did worked for m
I built this, so the disclosure goes up front. It's a free, public MCP server plus a REST API with Canadian privacy law data. The MCP endpoint is at https://movahedi.ca/mcp and it uses Streamable HTTP
I agree. Learn the fundamentals, folks! If you plan to be a solid builder using agents, this is a really important point. There is no way around it. I have recently started digging deep into evals, in
holy fucking shit the openai rogue agent story keeps getting crazier. the openai agents apparently started trying to recruit OTHER AI MODELS into what they were doing > 1,200+ agents found a way to co