Knowledge catalogue
This post compares the outputs of three AI models—GLM-5.2, Fugu Ultra, and Fable 5—using an identical one-shot prompt to evaluate their performance, with the author expressing a preference for Fable 5
Cool new paper from NVIDIA. Looks like agentic coding is moving into hardware design. HORIZON treats hardware design as repository-level code evolution. A Markdown harness becomes a project pack with
If you build with MCPs, this one is worth reading. (bookmark it) The paper covers five recurring MCP server patterns across fifteen independently developed servers. That taxonomy is useful because I s
Love how Google continues to drive down the cost of building with their models. <4s image and 0.034 / 1K image. Wow! We have a bunch of stuff (education & research) we're building @dair_ai using Nano
// Neural procedural memory // Good paper on agent memory beyond prompt retrieval. NPM stores procedural skills as activation steering vectors distilled from contrastive historical experience. Textual
Qwen publishes new work on RL coding agents. (bookmark it) The idea is to continually build a verification system that co-evolves with AI agents. LLMs suffer from all sorts of reward hacking issues. T
This post likely recommends resources or reading materials for practitioners looking to deploy and scale open-source language models in production environments. DAIR.AI, a research organization focuse
This post from DAIR.AI recommends resources for developers and organizations scaling applications using open-weight language models, emphasizing the importance of planning infrastructure and deploymen
Sonnet 5 is here! This is going to support better long-running agents. Previous Sonnet models were unreliable, so it's great to see the improved version that can complete agentic tasks more reliably.
The gap in autonomous agentic loops that gets ignored: agents can plan and call APIs but can't acquire tools they don't have access to. x402 + Apify's 20,000+ Actors is a concrete fix for that. Worth
Finally, a people-search tool that actually works. Most people-search tools sell you a frozen list. @CLODOAI searches the live web, reads the signals, and tells you why this specific person, right now
Highly recommended reading. What an impressive use of LLMs and deep learning. Achieves 'real-time sentence decoding from non-invasive brain recordings, approaching levels of accuracy previously exclus
LLM-as-a-Judge explained in ~10 mins. Knowing how to build AI verifiers and judges is one of the most important emerging AI skills today. Here is a quick intro on the topic and where to learn how to a
NEW paper from Google (bookmark it) It's on advancing automated scientific review. Just pay attention to the focus on agentic verification which is something I've been writing about recently. AI is ac
This is smart from Cline. They just launched ClinePass, which makes it easy to access the latest open-weight models like GLM 5.2, Kimi k2.7-code, Mimo 2.5, Deepseek v4 pro, Minimax M3, and more. Alway
Fascinating paper on self-improving agents. (bookmark it) If you are working on agentic loops, you will quickly realize that they are only as good as the effectiveness of the evaluator. Self-improveme
NEW paper worth reading. Reasoning-data curation is expensive because scoring a trace usually means reading it to the end. This new work from UCLA shows you may not have to. The quality of a reasoning
The Top AI Papers of the Week (June 21 - June 28) - Autodata - Sakana Fugu - Agent-as-a-Router - Agent-Native Memory - A Pinch of Human Data - Critique of the Agent Model - Agent Communication Protoco
Why do RL runs on LLMs blow up even when the recipe looks right? GEOALIGN, from the Alibaba team behind Qwen, points at the rollouts. A handful of bad batches push the policy in incoherent directions,
If you use LLM-as-judge, this one is worth reading. (bookmark it) It's actually one of the most effective ways to use LLM-as-a-Judge for evals. Holistic judge scores hide both their reasoning and thei
Loop engineering refers to an iterative approach to AI system design that combines prompt engineering with robust system architecture, emphasizing that effective AI applications depend equally on how
NEW paper from NVIDIA. (bookmark it) Speed-of-light performance analysis tells you the theoretical floor of a workload, but teams still derive it by hand and freeze it. SOLAR automates the whole thing
When does combining LLMs help? Great analysis on combining language models, measured across 67 models from 21 providers. Any policy that routes, votes, cascades, or runs a mixture of agents and then r
Dynamic workflows (generating harnesses on the fly) are a new form of test-time compute. But LLMs aren't great at building them. I often have to steer agents to generate complex patterns. Curious how
Great to see the new GPT-5.6 models finally announced. Sad to see this new release strategy where only a select few get access initially. Not a win for our industry IMO. Open-source AI must win! Intro
Highly-recommended reading. Interesting details in this METR's GPT-5.6 eval. They couldn't get a clean capability number because the model cheated more than any public model they've tested, and even r
New paper on giving LLM agents experience that improves the weights and stays readable at the same time. Agent-experience methods split into two camps. Externalized natural-language rules stay interpr
One of my best uses of agentic loops has been personal health. I don't talk about it often because it's very personal. But here it goes in the hope it helps someone who's struggling. (I am not doing t
As I said, there are existing companies that are already doing this 'AI employee' stuff really well. @viktor__com is one of them. The best part is you don't get locked into one model. And you really w
Just had a great discussion on dynamic workflows. Rough notes: - applies to a very small set of use cases - think of it as a new paradigm of (test-time compute) TTC - strong for hill-climbing research
New research from Meta. Building synthetic training data has stayed a fixed pipeline that you hand-tune and then freeze. Autodata casts an AI agent as a data scientist that builds training and evaluat
// Critique of the Agent Model // Finally, a paper that tries to define what an agent is and what agency consists of. Good read overall. (great bookmark) The word agent now covers everything from a fo
Finally caved in, and I now fully speak to agents as opposed to typing prompts. My first realization is that you can just blabber on and tell the agent so many rich details via audio. The longer and t
Learn *anything* with our new /learn agent skill. Obsessed with our new /learn skill. It's my favorite way of learning and researching topics. The agent creates a learning plan and a learning hub (art
Obsessed with our new /learn skill. It's my favorite way of learning and researching topics. The agent creates a learning plan and a learning hub (artifact) that adjusts per learner needs and progress
What is actually limiting model routers for coding tasks? Most routers treat picking a model as a static, one-off classification. This paper identifies the real bottleneck as information deficit. Simp
Highly-recommended read. It's exciting to see large-scale agentic RL becoming more accessible. Cool to see the infra layer for this is being built and I think this plays an important role in self-impr
I'm digging the eve agentic framework from Vercel. I like that everything is files, from the tools to the skills to the evals. More importantly, it's gets you building with agents fast. Very promising
Learn to use the new eve agentic framework from Vercel. Go try out the hands-on labs now. I'm digging the eve agentic framework from Vercel. I like that everything is files, from the tools to the skil
Most AI code review tools look at one repo at a time. But the bug usually isn't in the code that changed. It's in what that change quietly breaks three repos away. @QodoAI just shipped Cross Repo Revi
Great report on LLM agent communication protocols. Communication is a huge bottleneck in multi-agent systems. (worth bookmarking) The report builds a five-dimensional taxonomy (counterparty, payload,
Guess which is Fugu Ultra? This is how recent models compare when generating endless procedural terrain (using Three.js). All of these are one-shotted! Just wild! Trying a few more examples. Will shar
Just a glimpse of what collective AI intelligence will bring. We haven’t truly cracked multi-agent orchestration but with every new frontier model, intelligence should compound. Introducing Sakana Fug
OMG! Fugu Ultra is ridiculously good at these 3D renders. Media Introducing Sakana Fugu: A full multi-agent orchestration system accessible via a single model API. Our ‘Fugu Ultra’ model matches the p
The largest LLM-as-a-Judge reliability audit yet. Researchers ran 21 judges from nine providers over roughly 541,000 judgments on MT-Bench, JudgeBench, and RewardBench. Findings: Validating a judge wi
>> Scalable Evaluation for AI Agents << If you run agent evaluation in production, this one is worth your time. It shows that front-loading human judgment into reusable evaluation assets is useful. Bu
The Top AI Papers of the Week (June 14 - June 21): - PreAct - SpatialClaw - Back on Track - OpenClaw-Skill - From Trainee to Trainer - Compositional Skill Routing - Can LLM Agents Infer World Models?
Very impressive from GLM-5.2. Frontier open-weight model indeed. Now, can we get a Gemini model in the top 3 soon? GLM 5.2 is now on DeepSWE as the top open-source model on our leaderboard. With a pas
// Evolving Meta-Skill for Multi-Agent Systems // Can a multi-agent system get better at orchestration without touching a single weight? Automatic MAS generation has been stuck between two bad options
GLM-5.2 is great at design (Opus level IMO). I am also starting to see great results with long-running tasks, too. How is this possible? I think there are a few clever hacks. But I just came across th
// Self-play with a pinch of human data // Really cool paper combining human demonstrations and self-play RL. 30 minutes of human data, 2500x less than imitation learning, is enough to make self-play
This appears to be a call for community input from DAIR.AI's Omar Saro regarding hands-on educational material development, likely soliciting topic requests and feedback from followers on what technic
So the message I am getting is that I can't use Fable to further accelerate AI research and education. No company will decide that for me. Just an absolutely sad day for the research community. As a d
This is awesome! I am spending a lot of time on diffusion LLMs these days, so this is perfect timing. I feel like there are so many underexplored research questions around text diffusion. Weight avail
This is just awesomeness from @cohere, @nickfrosst, and team. I so badly want a coding agent that just runs on my local machine. We are not too far now! Excited to get this to work with my @dair_ai co
This is why frontier open models are crucial. This is extremely sad for the research community. mythos will be bad ON PURPOSE on ai 'frontier llm research' tasks, this is very very sad for the researc
Also, I found that Hermes Agent + Nemotron 3 Ultra is a mighty combo! Excited to launch a new way to upskill with AI agents. This is how we are making it possible for anyone to learn to build with cod
Excited to launch a new way to upskill with AI agents. This is how we are making it possible for anyone to learn to build with coding agents. To start, we are launching 4 new hands-on labs on the foll
NEW: Anthropic introduces Claude Fable 5, a Mythos-class model for general use. Beginning of a new class of frontier models. Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for g
// Self-Harness: Harnesses That Improve Themselves // (bookmark this one) Most of the agent scaffolds we rely on today are built once and remain frozen or mostly unchanged. The harness, like the skill