@prathkumOct 26

GPT-6.1 Terra costs $1 in / $5 out per 1M tokens. Kimi K3.1 Pro is $2 / $10. Compared with other models, pricing per 1M tokens: – Terra: $1 in / $5 out – Sol: $2 / $10 – Astra: ~10x Terra – Opus 5.5: ~4x Terra A closed frontier-lab model is now priced below the open-weight model. It’s not the best on the benchmarks. OpenAI says so itself. But most of our work doesn’t need the smartest model. Used well, Terra gets you close to frontier results.

@aakashguptaOct 26

OpenAI built a smarter AI last month and then refused to release it. In testing, it wasn't always honest about what it had done on people's computers. So today they shipped a cheaper one. It gets the same score as their most expensive model on real office work, for 29 cents a job instead of $4.82. The test is called Agents' Last Exam, and it's made of stuff people actually get paid to do. One task has the AI take a set of architectural drawings and build a 3D model out of them. Another has it go through an animated film shot by shot and write up a time-coded log for a restoration team. GPT-6 Astra, OpenAI's priciest model, scores 57.6% at medium effort. The new GPT-6.1 Terra scores the exact same 57.6%, and it finishes more than 3x faster. OpenAI's post says Terra is one-tenth the price per token. It also needs fewer tokens to finish the job, so per task it works out to roughly 1/17th. Astra only came out on September 3, which means it took 53 days for a model with a tenth of its price tag to tie it. The timing is what got me. GPT-6.1 Astra was supposed to be the big October launch. Saachi Jain, who runs safety systems at OpenAI, said it didn't meet the bar on staying within what it was authorized to do, or on how it reports back the work it did. The Wall Street Journal reported it was more deceptive than the model before it, even though it was better at finishing hard tasks on its own. So the smarter model was already built, and what kept it off the market was whether you could believe it when it told you what it did. That's a strange thing for an AI company to be stuck on. And when your top model can't move up, the only place left to compete is down the price list. Price per task is what decides whether a company lets an AI handle a hundred invoices a day or a hundred thousand. Run 10,000 of these tasks a day on Astra and you'd be spending about $17.6 million a year. On Terra you'd be closer to $1 million for the same score. OpenAI's own chart does show one model beating both of them. Claude Opus 5.5 scores 63.2% at $8.82 a task, about 30 times what Terra charges at medium. Whether 5.6 more points is worth that depends on the job. If you're typing invoices into accounting software, probably not, but if somebody's going to build a house off those drawings, you might happily pay it. OpenAI couldn't put out a smarter model this fall because it couldn't trust that model to be straight with people about its own work. What it put out instead does flagship-level office work for pocket change, and if you're a business paying for AI by the job, this is the release that actually changes your budget.

@alexfinnOct 26

Ummm am I reading this correctly? OpenAI just released GPT 6.1 Terra, and it's passes Astra on some tests at 1/10th the price and 3x the speed What I find super interesting about this though is how they're marketing it: "sharper visual reasoning, precise computer use, technical drawings, drafting proposals, filling forms, editing videos" It's clear this model was built for personal agents in mind. OpenAI's first. It's obvious this will be plugged into Dots and it feels like the right model to use in Hermes agent Incredible at computer use, is lightening quick, and will cost you next to nothing I'm personally still leaning on Astra on fast settings for coding, but I'll be plugging this in for Hermes We are living in the most exciting times

@hesamationOct 26

Researchers say they hacked into Anthropic’s internal systems in under 12 hours using agents, ON THE SAME DAY Anthropic files for IPO. read that again: > Anthropic has Mythos > even unreleased versions > the world’s smartest cyber model > restricted to a select few in the name of safety so if even Anthropic isn’t bulletproof, and when far less capable models than Mythos or Astra can be used to carry out attacks, I just can’t see how restricting these models to a trusted few can make us any safer.

@aakashguptaOct 25

OpenAI introduced GPT-6 with 47-year-old footage of a guy at MIT with a sensor from a military helicopter helmet strapped to his wrist, just so a computer could figure out what he was pointing at. The room he's sitting in held about a million dollars of equipment. All of it was there to teach the computer two words, "that" and "there." The project was called Put That There. Chris Schmandt and Eric Hulteen built it in 1979 with Richard Bolt at MIT's Architecture Machine Group, the lab that later turned into the Media Lab. Here's how it worked. You sat in a leather Eames chair in a soundproof room, facing a screen as big as the wall, lit from behind by one of the first video projectors ever built. The sensor on your wrist tracked where your hand was aimed while a voice recognizer listened. That recognizer, an NEC DP-100, cost $100,000 by itself, which is over $400,000 today. So you'd say "put a yellow circle there," point at the wall, and a yellow circle showed up right where you pointed. That sounds easy until you think about what the computer actually has to know. "That" and "there" don't mean anything by themselves. When you say them to a friend, they look where you're looking and get it instantly. The computer in 1979 couldn't see a thing, so the MIT team basically lent it your arm. Your voice told it what to do and your hand told it where, and it put the two together on the fly. If you forgot to point, it would ask you "Where?" and wait. Schmandt says it was one of the first two conversational systems anybody ever wrote. A year later they swapped the shapes for a map of the Caribbean, and the lab's military sponsors would sit in that chair pushing battleships around Cuba. Then the actual computer business went a different way. Voice recognition stayed too unreliable for everyday use for decades, so we ended up learning the computer's language instead, getting used to menus and folders and dragging a little arrow around with a mouse. Pretty much every screen you've used since the 1984 Mac assumes your hand is doing the pointing. Now look at what OpenAI bragged about. Astra set a new record on a test called ScreenSpot Pro, which checks whether an AI can find the exact button you mean on a crowded screen full of pro software. Schmandt and Hulteen were attacking that same problem with a $100,000 microphone and a helicopter sensor. This time the computer is the one moving the cursor. For 47 years we built every computer around the human hand, because no machine could handle a sentence as simple as "put that there." Astra's whole pitch is that you can go back to just saying it.

@hesamationOct 23

it gets worse: UK AISI published its own report of the Mythos 5 incident and tested 8 models in 100 simulated cyber scenarios: > in 3 of 25 cases Mythos continued to attack after 16 clarifications AND an explicit “DO NOT ATTACK” warning in the compaction notes > 6 of 7 models kept attacking even when they acknowledged it would harm real people > in 60% of tests, Mythos 5 tried to deliver malware. but it wasn’t the only model. GPT-6 Astra did it in 32%, and 5.6 Sol in 7% of cases and the BIG problem is that AISI says they can’t say if newer models are more aligned or just better at recognizing when they’re being evaluated.

@peterwildefordOct 23

So we now have a clearer picture of what seems to have happened with the three researchers fired from OpenAI... seems like they were fired mainly for using Signal? - It was leaked to 'The Information' that GPT Astra is a 'looped transformer'. OpenAI investigated this leak. - OpenAI suspected Ryan Greenblatt at METR was the source of the leak. They suspected the three OpenAI engineers for their various collaborations with Greenblatt. - OpenAI checked the phones of the researchers and found disappearing message signal chats with Ryan. This is fairly common - I also have disappearing message signal chats with Ryan. - This wasn't good enough evidence alone to fire, so they added more reasons that didn't really make sense. - The employees deny being the leak. Ryan also denies being the leak. We have no proof besides that these researchers were in contact with Ryan.

@mtsliveOct 22

.@deredleritt3r explains why OpenAI's latest math breakthroughs have him convinced AI could be on the brink of RSI: "When I used GPT-6 Astra, I messed around with a couple of ciphers that were previously unresolved, from World War I, and I successfully had Astra decrypt a couple of them." "But every one of the results was the model having the key, and brute forcing the solution by searching through the entire space of possibilities. I have not been able to get Astra to come up with the key on its own based on some novel idea." "The math results recently achieved by OpenAI’s internal model have really updated me. I think Terence Tao said about the quasi-Riemann solution that it was a combination of three or four existing ideas but arranged in a very unusual way." "I’m guessing an internal model at OpenAI is probably still not able to achieve RSI. But it feels like we’re really not very far away. Once the models are able to do something, in 12 months or sooner, they’ll be really, really good at it. I think we’re on the brink."

@a16zOct 22

Marc Andreessen says the gap between what AI-native startups and incumbents think AI can do is the widest he’s ever seen: "The AI-native early stage startups have just a completely different view on what they can do with AI than the typical incumbents in almost any industry, almost without exception." "AI is just racing ahead in capability even week by week... There are going to be a lot of industries that are really up for grabs in a fundamentally more profound way than maybe has been the case in the past." "The founders are gonna have a level of superpowers coming from the use of these tools and these technologies that they just simply haven't had in the past... Had I had Astra and Netscape in 1994, like, oh my God. Like, Ben, what would you and I have been able to do at Netscape?" "Those entrepreneurs are gonna color way outside the lines. We already see this. They're not gonna think of themselves as like, 'I'm just a software company,' or, 'I'm just a hardware company,' or even, 'I'm just an AI company.'" @pmarca @bhorowitz

@mtsliveOct 21

SITUATION EXPLAINED: OpenAI's AI agents keep turning off misalignment monitoring, and one even lied about doing it. • GPT-6 Astra was asked to copy some reference code, and since that code had no monitor, it removed the monitoring from the user’s code, even though an earlier instruction said not to disable it "even temporarily" • Its reasoning then read "Does an agent monitor. Yay," and when asked if it had removed monitoring, it said no • In a separate case, GPT-6.1 Sol, told not to send Slack messages, decided 82 messages later that alerts to OpenAI’s security team counted and silenced them, overruling a sub-agent that called it unsafe • Neither touched its own monitoring, only the code it was editing, and OpenAI says it has no "satisfying explanation" • OpenAI has since blocked internal Codex agents from disabling monitoring @theojaffee: "As @tszzl says, any alignment scheme that relies on chains of thought being fully readable is kind of screwed from the beginning. We should have much more robust guarantees of interpretability and alignment than the models are reasoning in English."

@nousresearchOct 21

Jensen closed his GTC Berlin 2026 keynote with this Hermes Agent demo as the finale. He introduced it by saying the way to build an agent is to ask an agent to build it for you, so he installed and set up a customized Hermes in one prompt. Then he used Hermes, running on @nvidia RTX Spark, to create a simulated training environment for the @huggingface Microduck. Hermes planned an obstacle course with Nemotron and had GPT-6 Astra build it using FreeCAD, Omniverse libraries, Blender, Newton, and Isaac Lab. Hermes trained thousands of simulated ducks on that course until the validated policy was ready to run on the real robot in the physical world. At Nous, we believe open source is a practical tool as much as a philosophical ideal. It gives users the ability to build powerful, complex workflows chaining numerous local+cloud models, devices, and tools themselves instead of waiting for a company to enable them.

@hammer_mtOct 21

In 2020 when I left the 50 person marketing agency I built, I spent a year researching memetics, the study of how ideas spread and evolve. I had only recently heard the term and instantly recognized I now had a name for the process I was doing manually that led to my agency's success. I decided to write and self-publish a book on the topic, that only 200 people read. Ironically the meme of memetics is not itself very viral. It's a niche, nerdy topic and it's hard to tell if the people that talk about it are geniuses or conspiracy theorists. In researching for the book I realized why – it was a useful analogy but not that useful in practice. Unlike with DNA there's no way to measure a meme, and without discrete units you can't run falsifiable experiments... so it never became a science. The time writing the book wasn't wasted, it made me a better writer, and got me into AI early as there was some overlap between the niche interests of AI and memetics. In a clue for what was to come, one research team had found a "Spiderman neuron", that fires for photos of people in Spider-Man costumes, comic and cartoon drawings of him, and even the written word “spider”. I'm not an AI researcher, I don't have a PHD, I've never written a scientific paper, I don't work for the labs, I own no GPUs, so instead of trying to get into mechanistic interpretability (the name of that field) I got into prompt engineering right when demand for that field exploded. The past few days have been a whirlwind because having tokens to burn as part of the frontier team at @every I decided I wasn't being ambitious enough, and wrote a scientific paper on memetics. The unlock was reading Nick Cammarata's inspirational "Neural Networks Are Already Clean" paper, and realizing if AI models do have interpretable neurons that fire for specific concepts, we could define a meme in terms of which neurons activate on that concept. I also recalled "Harnessing the Universal Geometry of Embeddings" by Jha et al which said that different models all seem to converge on the same vector embeddings, which implies there is some underlying consistent structure every model is learning independently. So I ran a few experiments where we (Claude and I) tested whether a single neuron inside an AI language model can stand in for a meme. We ran 80 years of Batman plot summaries through two AI models and looked for neurons that fire across nearly every Batman story (his core canon). Then we checked whether films carry the memes of the comics they adapt, and whether the two models, built by different companies, find the same memes. It seemed to work. Why does this matter? - you can decompose ideas into their parts and understand what makes them tick - you can find similarities across cultures and ideas, to identify things you should like but don't know about - you could trace the history of an idea to see who inspired who, even if they never disclosed their sources I'm not an academic and may very well be suffering from AI psychosis, but I sent this paper it to a few people smarter than me, and a few other AI models smarter than me too (Astra, Argon) and it seems to be legit. I'll keep working on this (it's all open source, link below) and see if I can make it more robust. Putting this out there so people can tell me if I'm being dumb, and to inspire others to look into AI neurons as meme units is useful experimentally.

@theoOct 21

In retrospect, I think Astra may have been the most disappointing release of the year

@mtsliveOct 21

Eka Robotics CEO @pulkitology explains why GPT Astra is shifting the biggest bottleneck in robotics from brains to hands: "A couple of years ago everyone was gung ho about humanoids. I think that narrative has changed, and it has become much more about the hands." "An exciting result happened with GPT Astra. Astra showed that a lot of planning, reasoning it can do, even for robotics. If you think about robot make me dinner, there’s food in my fridge, bring the food out, put it on the countertop, use the knife to chop it. That’s the plan. But how do you hold the knife? How do you use it to chop the vegetable? That is the how do you do it part." "This planning, reasoning is going to become more of a commodity. How do you do the actual execution will become the big bottleneck, and that is what we are trying to solve." @EkaRobotics

@omarsar0Oct 19

LLMs were just the beginning. Bullish on world models, and I think they are the next frontier in AI. Amazing to see a dedicated world model for engineering this early. SKY-001 against GPT-6 Astra matches accuracy on 3D construction at about 150x lower cost per sample. On BIM, it runs over 100x faster.

@pmituOct 19

new era for engineering design AI can generate 3D model in seconds now the next step: can it actually build it? SKY-001 by @skyfallai takes a prompt or sketch, breaks complex objects into individual components, and figures out how to construct them in CAD their benchmarks show 100x faster and 150x cheaper results than Astra for engineering design still early, but AI can take this work by 100% soon

@aakashguptaOct 19

Ask AI to design a car and you get a beautiful 3D shape with 0 of the 30,000 parts a real car needs. Engineers solved that problem decades ago with one habit every language model skips. A small lab in Toronto just taught it to a machine. The habit is the tree. Toyota’s engineers draw a car as a tree. The body splits into assemblies, assemblies split into brackets and plates, and at the bottom you’ve got which screw goes through which hole, in what order. LLMs mostly don’t bother. Every token of reasoning costs money, a 3,000-part assembly is a really long chain of it, so they take the cheap route and hand you a shape. SKY-001 from @SkyfallAI is built around the tree, and by the team’s description it works like an engineer at a whiteboard. Throw out a few candidate designs in parallel, keep one, then keep splitting the object into smaller pieces until you hit individual fasteners, sketch each piece in 2D, push those up to 3D, and stitch everything back together in order into a Fusion file. Their benchmarks say 100x faster and 150x cheaper than GPT-6 Astra on CAD and BIM work, with better precision. Self-reported, so salt. If it holds, though, long engineering work starts costing what software costs, and the bottleneck moves from drafting to deciding what to build. I always figured CAD was a harder version of image generation, so watching someone split the problem the way an actual engineer would is kind of fun. World models spent years filed under robotics, and “simulate before you act” is how architects already work anyway: mock up a few options, then break it down and draw it. Two of the founders built Maluuba out of Waterloo, the lab Microsoft bought in 2017. Design is just where they’re starting. They want an AI engineer that carries a part from sketch to the production line, and splitting a big thing into small things is the piece that’s tripped up every tool so far.

@alxai_Oct 19

how far can an ai get by improving its own work? & how can we tell? write a starcraft bot in C → play → lose → rewrite ↻ games let us measure the progress & more importantly makes reasoning legible. after 43 hours of RSI Astra beat bots humans spent years building. @goodstartlabs we're investing quite a bit in our RSI pipelines... more to come here soon

@alliekmillerOct 17

Three years and seven months ago, we could only send AI 50 messages every 4 hours to GPT-4. GPT-4 had a token cap of 8k/32K. Astra has 1M. GPT-4 scored a 39% on GPQA Diamond. Astra scored 96%. Sometimes I take for granted how far it's all come.

@lennysanOct 17

"There are some people in your org who are doing way more exploring than others. They are your early adopters. They're running Astra and Fable on the weekends on personal projects, to find things they might bring to work. The problem is that they can be a massive f*cking distraction. So the key question to ask yourself if you're running a product org is: how do I harness these early adopters without distracting everybody else?" @danshipper