Search: “gpt”

By end of next year. End of year after for 16gb vram. Do folk really need more for most things? Not really. But just call the cloud models when you do. What's a reasonable timeline for a GPT 5.4/Opus

embers GPT-5.5, not fully saturating the TikZ unicorn test yet but getting awfully close ... (yes this is actual TikZ code, I personally find it so unbelievable that I'm putting the code below for any

important (and very jakub-coded) jakub quote: OpenAI Unveils GPT-5.5. Company Says Expect a Faster Model Release Pace 👀 OpenAI: 'We see pretty significant improvements in the short term, extremely sig

I've been previewing this in Codex for a few weeks - it's very good! Had some great results from it having it run security reviews against code written using other models Introducing GPT-5.5 A new cla

wow the codex app is unrecognizable… almost like it shouldve been Atlas the whole time With GPT-5.5, Codex now gets more of the job done across the browser, files, docs, and your computer. We've expan

This post discusses concerns about AI model degradation through the inclusion of low-quality or irrelevant training data, using butternut squash image generation as a humorous example of how excessive

open-source community will work their magic! If any inference providers could serve kimi k2.6 at 300+ tps reliably, I would be happy to switch from gpt 5.4 xhigh to kimi for daily tasks. It's a really

To go deeper on our new Life Sciences model series, research lead @joyjiao12 and product lead Yunyun Wang joined @AndrewMayne on the OpenAI Podcast to discuss how we’re building models for biology, dr

Grok Grok 4.20 Reasoning just took the #1 spot on the BridgeBench reasoning benchmark. 🔥 Beating GPT-5.4, Claude Opus 4.6, Google Gemini and others. Week after week, Grok keeps climbing across benchma

Meh on the Lem Test as well https://x.com/emollick/status/2019478314767835634?s=20 Opus 4.6 saturates my Lem test, which I've done since GPT-3.5 SciFi author Stanislaw Lem wrote of two rival construct

Great paper on improving memory for AI agents. NEW paper: Memory Intelligence Agent (MIA) MIA boosts GPT-5.4 by up to 9% on LiveVQA. Quick summary: Most memory-augmented agents treat memory as a stati

GLM-5.1 by @Zai_org just launched in the Text Arena, and is now the #1 open model. It outperforms the next best open model, its predecessor, GLM-5, by +11 points and +15 over Kimi K2.5 Thinking. It sh

INCREDIBLE GLM-5.1 weights are now opensource > i’ve had early access to the weights for the past few days > and yeah… this one matters a lot benchmarks? > SWE-Bench Pro: 58.4 > beats Opus 4.6 (57.3)

New open source app built with Together AI! Gloss turns your saved articles, PDFs, and X bookmarks into a beautiful personal library, with lightning-fast AI answers powered by DeepSeek V4.1 Flash. Try

We were able to fine tune an open weight model (Qwen3.5-9B) on proprietary internal data at @coinbase, and it out performed frontier models, at much lower cost and speed, for a specialized task (fraud

AI solved open mathematical problems. But the biggest surprise came afterwards. A team at Google DeepMind has used AI to resolve 9 open Erdős problems, including questions that had remained unsolved f

Banger paper from Meta Superintelligence Labs on user simulators for agent training. Agent RL setups usually let an assistant LLM play the user, so the simulated user is too cooperative and too explic

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

Some early first-hand accounts of encountering a narrow superhuman intelligence as mathematicians grapple with the hundreds of big AI proofs released by OpenAI. Problems solved in inhuman ways that ma

Introducing GeoGuess Bench! A benchmark to measure how good AI models are at playing GeoGuessr. We gave each model 210 photos from around the world, asked it to predict where each photo was taken, and

Like Jev, I believe RL will unlock several more like this, slashing the cost of critical agent operations. flow-1 is competitive in performance, but a huge cost-saver for finding failures in agent tra

Codex Security Cloud is getting a major upgrade, with access to cyber-capable models through Daybreak Blue included by default. It scans entire GitHub repos, continuously reviews new commits, investig

OpenAI introduced “Dots,” an AI system built to run continuously and free users’ time so they can focus on higher‑level tasks. The announcement was posted on September 29 2026 by CEO Sam Altman via Tw

OpenAI just announced the 200 Pro plan is back but with half the total API spend included, offset by some of their smaller (NOT Astra!) getting a 50% price drop. Does this impact the average user? No.

What is a way to test human-level generality in artificial intelligence? 'The meta-benchmark of being able to pass ARC-AGI-(n+1) immediately upon release' @Jabaluck @DAcemogluMIT The meta-benchmark of

People on X: “CLAUDE IS BACK, baby!” Listen, Opus 5.5 is incredible and probably my favorite overall model right now, but it is still ridiculously verbose and over-justifies and over-qualifies everyth

DON’T LET YOUR CODING AGENT BE THE ONLY REVIEWER OF ITS OWN CODE. I’m seeing this in my own Claude / Codex sessions, and customers are sending me similar screenshots: | Qodo review is on the money. |

MiMo-V2.6-Pro is live on Wally. Day 1. @XiaomiMiMo's new open-weights flagship, top of the @ArtificialAnlys Intelligence Index for open-weight models. 320 tok/s, 2.4x the model maker's own API, in <24

On 22 September 2026, Allie K. Miller posted on Twitter that the Opus 5.5 model led her to curse at Claude less frequently. She notes that while reducing profanity toward another AI is an undesirable

Grok 4.7 xHigh is now at the top with just ONE point away from Claude Fable 5.1 Max on Artificial Analysis’ AA-Briefcase benchmark Claude Fable 5.1 Max — 59% Grok 4.7 xHigh — 58% Just a 1-point differ

Banger paper from MIT and Sakana AI. They show that self-improving coding agents work. The best part is that their approach, Self-Improvement via Fast Tree-search (SIFT), runs at a tenth of the CPU ho

One of the craziest use cases I’ve found for Jev: verifiers. I am so excited about this that I at least wanted to share the high-level idea. I used Jev to build a custom verifier for the /goal feature

Is there a middle ground between total bans & zero oversight? Marcus is right that we rushed into agents without fixing basic sandboxing first. Perhaps the most important warning I ever issued was thi

🔥: “While AI comes with danger, we should be sceptical of any narrative that conveys inevitability around its trajectory. AI systems do not build their own data centres. They do not manufacture their

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

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

Does anyone have any idea how this could vaguely be possible given the nature of and various routes to RSI? I am not being facetious I genuinely don’t know how this could be constructed or implemented

DeepSeek V4.1 Flash is now available on Together AI’s model hub (https://www.together.ai/models/deepseek-v4-1-flash). The release includes the DeepSeek V4.1 Flash API with updated pricing, performance

Wild 24 hours for AI and lots of different proposals have been made. TLDR; the only *tangible* new fact is that OpenAI and Anthropic are going to have embedded 3rd party evaluators from unknown organi

This model is insane at landing pages. I asked DeepSeek V4.1 Flash & Claude Fable 5 to build me a landing page for a movie theater. Fable cost $1.21 while V4.1 Flash cost 2.6 cents, making it more tha

Astra is awesome, but if you're using it specifically to parse and extract a massive volume of documents, you will break the bank (10c+ per page on parsing and extraction!) We've added a native LlamaP

Really strong benchmark paper on coding agents. Claude Opus 5 running under Claude Code passes 23.9% of the evaluations. An expert human reference scores 82.2%. Here is what the task actually is. The

“The expected value of a fully LLM-generated, non-human-reviewed patch is a net-negative by a considerable margin.” 🦔AI can only fix security vulnerabilities 26% of the time. Researchers at 1Password

Yet another win for Neurosymbolic AI! Fermat! Anthropic says its first Fermat runs failed not on math, but because dozens of Claude agents lost track of shared project state. They only succeeded after

Does Astra have human-like continual learning capability over long periods? If not, it's not AGI. Even a system that could pass @demishassabis's new 'Turing test' where it has to independently discove

Very important paper accelerating research around AI discovery. https://x.com/omarsar0/status/2095858839534862664?s=20 New benchmark from @Apodex_AI worth digging into. TRACES scores AI systems on dis