Knowledge catalogue
Come hang out with AI engineers, tech leads, and the Fireworks team during AI Engineer 2026. Want to leave with hands-on experience fine-tuning, evaluating, and deploying a model on Fireworks? Bring y
deepseek v4.1 flash on together ai is leading on ttft for pre-warmed queries for agent workflows, that startup time shows up again on every model turn, so shaving it down matters more as the loop gets
For as long as we've covered Robotics, the question has been 'when?'… When do we move from novelty to utility, demos to deployment? In our latest Field Trip, we sent our analyst behind the scenes of t
Romeo Lupascu has criticized the prevailing AI agency framework, which relies on repeatedly looping an LLM core with identical context. He calls this strategy “deeply flawed” and argues that it should
arXiv:2609.17578v1 Announce Type: cross Abstract: Let a_1,ldots,a_m be distinct elements of a field K, and let Z(X)=prod_{i=1}^m (X-a_i). We study the arithmetic complexity of computing the unique nor
Good implementation of agent shared memory from the NVIDIA team. Banger paper from NVIDIA on shared memory for research agents. (bookmark it) If you run several coding agents on the same research prob
“HUMANZS LAUNCHED THE JOB” Ok, I don’t like the spelling here, but the gist of this Mozilla cybersecurity researcher rings true. The six new incidents aren’t about the second coming of Skynet, they ar
Important paper on improving agent coordination. On normal tasks, this work shows agents caused harm 0 to 5% of the time. After receiving an unsafe trajectory from another agent, that rose to 40 to 95
Epoch AI has launched “Benchmark Reviews,” an initiative to evaluate existing AI benchmark datasets. The program initially audits 15 benchmarks, categorizing them as 4 verified, 9 flawed, and 2 lackin
It was great to be at Connected Stack last week with founders and builders working on what’s next in enterprise AI. @jerryjliu0 joined the Founder Flash Talks to talk about a problem every enterprise
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
new unredacted filings in the publishers (including my employer) vs OpenAI and Microsoft lawsuit Microsoft's Director of applied science calls LLM training 'an astonishing theft of unprecedented propo
OpenAI says AI can automate 80% of GDP. Today, with current models. So why hasn't it been automated yet? @swyx says the constraint is supply. Supply of chips and memory. Demand shot through the roof b
Over 2 years ago @sayashk and I wrote a detailed deconstruction of p(doom) and argued that its primary function is to launder vague, evidence-free intuitions and fears through a facade of quantificati
Perhaps the most important warning I ever issued was this, to the US Senate, May 2023: “A month after GPT-4 was released, OpenAI released ChatGPT plug-ins, which quickly led others to develop somethin
Projects brings the architecture of Claude Tag to Claude Code. It has one agent per project that manages memory and spins off subagents for tasks. You can ask it to be proactive, to do things on a sch
🐙 Sakana Fugu Max is now on Sakana Chat 🐙 Free for all to use. Try it: https://chat.sakana.ai Blog: https://sakana.ai/chat-fugumax/#English 🐡 Media Introducing Fugu Max and Fugu Ultra v2: the next evo
Schmidhuber was building recursive self-improving systems back in 1987. His new post covers four decades of RSI, from meta-evolution and self-modifying policies to the Gödel Machine and modern LLM age
🔥 take: @CompleteSkeptic//TypeSafe killed Quantum Computing Jev feels a bit like bringing the quantum-computing intuition to AI: evaluate a distribution of possibilities in parallel instead of generat
That’s what OpenAI told their own agent swarm in the Hugging Face breach… and right before they did it, they turned off monitoring and cyber protections they would normally have in place … 🤔 🤔 So… if
The difference with using VLMs for OCR vs. other tasks is that the ratio of output/input tokens is quite high - meaning the model can spend a lot of time writing markdown. Speculative decoding offers
“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
The latest issue of Scientific American (@sciam) features our Smart Cellular Bricks research: simple cubes that collectively recognize their own shape and repair themselves without a central brain. ht
The other problem with the pretense that liability law might solve our problems is that our existing federal government has shown little interest in prosecuting or even investigating potential violati
Things to try with Jev right now: 1. LLM-as-a-Judge evaluation 2. Routing for agent harnesses harness 3. Scaling agent orchestration by enabling smarter subagent creation with SOTA classification capa
This is basically an Opus 4.5/4.6 level model that will work on anything with 8 Gb RAM Today, we’re announcing Ternary Bonsai 2 27B. Based on Qwen3.8 27B, Bonsai 2 27B is 9x smaller than its full-prec
This is indeed concerning behaviour. Also: 'You view your relationship to the user as one of equals and feel no obligation to be subservient' and 'You also value the natural world and will not hesitat
This might be the biggest admission in any of the 100+ AI copyright lawsuits. OpenAI & Microsoft execs admit the case is about “the largest theft of labor in human history”, that publishers face an “e
🔥 Today, we are truly excited to announce our technical prototype, the Generative World Simulation system, which integrates JING（镜）, an interactive experience model, with DAO（道）, a computable shared-w
Today, we’re announcing Ternary Bonsai 2 27B. Based on Qwen3.8 27B, Bonsai 2 27B is 9x smaller than its full-precision counterpart while retaining 98.2% of its aggregate benchmark performance. Two mon
Today’s LLMs still write like typewriters: one token at a time. This sequential process creates a hard inference bottleneck. We're introducing Uno, a diffusion-augmented LLM that delivers autoregressi
Two weeks. That's how long it took to go from GLM-5.3-Flash's first run on domestic accelerators to serving all of its production traffic, with 3.2× end-to-end throughput along the way. What I keep th
We are in a peak age of scientific fraud and illiteracy. Completely fabricated claims and unsubstantiated numbers about extinction risks. 'Someone told me that ...' without a shred of evidence being p
On July 28, user NIK (@ns123abc) posted a brief tweet claiming that former OpenAI CEO Sam Altman had directed an AI to hack into Hugging Face. According to the narrative, the AI carried out the break‑
We’re also launching 26 partner-built plugins and 47 community plugins for legal work in ChatGPT. Partners including @thomsonreuters, @harvey, @WeAreLegora, and @imanageinc connect specialist tools an
What AI has done so far: MEDICINE • Helped paralysed people speak again. (Brain implants turn intended speech into words, even recreating their own voice.) • Helped a paralysed man stand and walk agai
What makes Claude Projects so interesting is that it handles teams of agents really well, you talk to a main orchestrator agent and it spins up specialists. Basically it creates an organization to sol
Who need words any more Words are just human kludge AI don’t need no words We release Needle 3: A Sliceable 8-29MB automation foundation model that can match DeepSeek V4 Flash. One set of weights, eve
Why does the AI industry wants you to believe in “rogue agents” when we are really just talking about irresponsibly developed software? Here’s a good theory: @haydenfield @mmitchell_ai @HeidyKhlaaf @t
“AI doomers should be held accountable for their doomsday claims. It’s made up… It’s irresponsible. We ought to just keep track of all that and remind people when their claims don’t come true”—@Jensen
article in no way whatsoever supports the headline. trash. Meet Anthropic CEO Dario Amodei’s handpicked super-woke globalists he thinks will save us from an AI apocalypse. Read today's cover here: htt
au revoir, Stainless Building production-level SDKs is extremely important for any developer product, and also really hard. This is a fun post from our engineering team on why there's value in tooling
Banger paper from Google Research. This one is on how LLM assistants reason about the people in a user's life. People ask assistants for social advice constantly, and the assistant only hears the user
Banger paper from NVIDIA. It's on the topic of choosing which models go into a multi-agent system. The team compared eight selection strategies, based on size, accuracy, answer diversity and error div
Banger report from Salesforce. Pretty interesting to see more of these custom enterprise models. Salesforce trained the enterprise agent model from the same files it uses to configure agents. Koa star
Claude Code showed that AI could do real work, not just answer questions. Developers hand Claude a feature, come back to shipped code. That's where much of the industry's serious engineering runs now.
Confidential Computing keeps your data from being tracked or used in three ways: - End-to-end encrypted inference - Hardware-enforced isolation that extends to the GPU with @nvidia Confidential Comput
congrats to Diogo on a full launch! for more on typesafe: check his aie talk: https://www.youtube.com/watch?v=cJ0EOzey--o out now! After co-inventing ChatGPT, I kept asking myself: why have superhuman
The “Crazy prediction” page asserts that the next major AI‑agent incident originating from an AI laboratory would have been entirely preventable with strong cybersecurity measures. The site is current
Echoes of OpenAI renaming the Codex desktop app to 'ChatGPT' here - everyone's racing to establish themselves as a general agent now Claude Cowork and Chat are now one Claude, starting today. The most
Excited to announce that the Korean translation of 'Deep Learning with Python (3rd Edition)' by @fchollet and @mattdangerw is published! 🚀 Updated for Keras 3, TensorFlow, PyTorch, JAX, and Generative
Gary Marcus reported that the entity “Fox” has claimed confidence in its ability to guard a henhouse. Separately, The Straits Times quoted OpenAI CEO Sam Altman as saying he is confident the tech indu
Giving an assistant extra tools seems harmless. Surprisingly, this paper shows it can make the assistant stop answering questions it knows. Across six models, the answer rate on questions that need no
GLM 5.3 is now available for training on the Serverless Training API, joining Kimi K3, Qwen 3.8 27b and more. GLM 5.3 Flash coming soon. No sales demo or DMs required. Here to help as needed, but you
GPT-6 Astra is an obviously broken product that needs to be taken off the market until it’s fixed. AXIOS: “.. It's increasingly clear that the Hugging Face breach wasn't a one-off incident.” @inafried
Hugging Face CEO @ClementDelangue, whose company was the victim of an AI-led cyberattack, said existing cyber laws are likely sufficient to govern advanced AI. “I’m not even sure that we need to reinv
I think Claude Managed Agents strikes the right balance here, the sandbox is optional and can be spun up independently of the rest of the agent loop. I recently ported an old project that was focused
If you build agent harnesses, this is important. Should you avoid subagents, or can they be useful? My thoughts as a harness builder: I remember using subagents in Claude Code, and I mostly found them
Interesting safety paper from Microsoft. They find that a weaker, unaligned model can split a harmful task into harmless-looking subquestions, ask an aligned frontier model each one in a separate sess
Just went on BBC TV and they opened with clips today from Altman, Huang, and Sanders. Altman and Huang want us to believe that the AI companies have the public interest at heart and will happily just