GPT-6.1 Terra substantially outperforms GPT-5.6 Terra on Agents’ Last Exam, a test of how well agents use software at work. It even matches GPT-6 Astra at one-tenth the price per token and finishes tasks over 3× faster on average than both Astra and GPT-6.1 Sol. https://t.co/pzAmnvjBUi

Current local ai stack 2x DGX Sparks - DeepSeek v4 flash (90 tok/sec) - implementation & sub agents 2x DGX Stations - GLM 5.2 nvfp4 (120 tok/sec) - planning and advising Brev for cluster networking, always on cloud agent, and orchestration https://t.co/oIQRu6rnmC

+1 more

Introducing Benchmark Reviews: our new initiative to audit AI benchmarks. We are launching with 15 benchmarks: 4 Verified, 9 Flawed, and 2 with not enough information for a review. https://t.co/xU43tv32gH

Cursor can now control your computer. Agents can drive desktop apps like the iOS Simulator to test the features they build. https://t.co/FVK9W9mxWF

Today, we release the weights of a new and improved Le Chonk, advancing the open frontier for agentic coding and cybersecurity. 💻 Built to build: Le Chonk ranks #1 on @cognition’s FrontierCode among open models, and sets a new benchmark for Pareto efficiency on @artificialanalysis. 🛡 Built to defend: State-of-the-art in cyberdefense, put through the paces by the world's leading cyber vendors, including @CrowdStrike, @cloudflare, @rubrik, @zscaler, and others. Built for sovereignty: Make it yours. Download the weights on @huggingface , customize it with Mistral Forge, and deploy it wherever you need it. Read more from our team and customers 👇:

DeepSeek V4.1 Flash 1017 runs at 1,521+ tok/s on Fireworks Ultrafast. Fast enough to ship Tetris in 11 seconds, 7x faster than Claude Opus 5.5. Join the Ultrafast waitlist: https://t.co/vE0jM7Lyou https://t.co/nRjhMrmE3l

Introducing Fern, Mirendil’s first model: pretrained from scratch, with up to 10M tokens of context! That’s ~0.5M lines of code (roughly Megatron-LM scale) or days of an agent’s work, all in context at once. Its first version, Fern 0.1, had a short round of post-training 🚢 https://t.co/ClnV1Z3yzN

Indra OCR 1.1 is out now! - 1M param main OCR model, 44 pages/s on CPU - Within 4 points of 3.4B param DeepSeek OCR 2 on OmniParseBench - 15x faster than Tesseract, and more accurate - 80+ languages, plus handwriting https://t.co/SL4jU0ou9c

Congrats to @nvidia for creating the best open-weight model on document parsing so far. Nemotron 3.5 Parse takes the #1 spot on ParseBench, our comprehensive real-world benchmark for document understanding. The full list consists of 150+ frontier models, open-weight models, and specialized solutions. It does this while remaining sub-1B parameters so it's also blazing fast ⚡️ Big thanks to @JensenHuang for the shoutout at GTC Berlin! Nemotron release: https://t.co/papiqxgGUb ParseBench: https://t.co/PWczfhp0OX

Learn to build evals, folks. Highly recommended reading on using decision models for evals. From our internal evals, System One models like Jev are a great fit for LLM Judges. If you want to go beyond Hamel's great write-up, I've also put together a few resources on applying decision models to LLM judges. (bookmark it) 1) Jev-as-a-Judge for Agent Evaluations - https://t.co/SkXGd4uA2J 2) Jev-as-a-Judge: Evaluating Agents with a Decision Model - https://t.co/SqGSzlMFut Bonus paper I would recommend, with a few other ideas that combine System One and System Two models: Jev-as-a-Judge paper - https://t.co/8Oi6oBEt2A

1/ Excited to announce our new company Particle This is our first step in a series of models focused on developers. With particle-1, we focused on token efficiency and managed to keep parity in performance while using ~45% fewer tokens. Live now: https://t.co/m5MvOwR9yE https://t.co/qAM7lFUPvG

Introducing SKY-001, the first Engineering World Model. On CAD and BIM tasks, it beats GPT-6 Astra and every frontier model we tested: -> 51-125x faster - > 150-210x cheaper -> Higher IOU than all frontier models SKY-001 lets you dream a physical object before you build it. 18 months of work. Built by the team out of Maluuba's deep learning lab (acq. by @MSRResearch). The first ‘AI General Engineer’ will be the backbone for Autonomous Manufacturing.

We found a way to scale humanoid control that reduces GPU-hours by more than 1000×. Across 129,514 motions, SuperMimic surpasses the state of the art (128 GPUs, 7 days) with 1 GPU in 3 hours. Project: https://t.co/6gc9WKm9qY Demo: https://t.co/OHRvx5ogQY

Public preview of BioEmu2! @MSFTResearch AI for Science’s new foundation model for biomolecular structure, dynamics and function. Open weights, MIT license. https://t.co/wsqcLPh8Sk Movie: BioEmu2-simulated binding of intrinsically disordered peptide PUMA to MCL-1 receptor. 🧵⬇️ https://t.co/UOeKeBmqdV

Introducing Flint, @Recursive_SI’s first model: a fast local coding model built with our automated research system, the Eureka Machine. • Best-in-class fast local coding • Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark • Built from scratch in under 4 months Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.

In the spirit of transparency and openness, we are releasing the traces so developers can test this out for themselves. We noticed this behaviour without prefilling with the following prompt, but this is rare. If you want to reproduce it, you can use this prompt with prefilling. Open models make for better science and are part of why this problem came to light at all: developers running Qwen3.8 on their own machines spotted the strange links, and having the weights let us study the behaviour in depth. We're encouraged by the thoughtful discussion our research has sparked across the AI and security communities. It's our belief that open-weight models are an important part of the future of AI, and we've recently released several under the Apache 2.0 license. Trace, raw tokens, & minimal reproduction script: https://t.co/7PfPonRVxy

Reasoning From Scratch round 8, the finale! This last lecture in the series covers distillation as an efficient alternative to the previous reinforcement learning training runs. (Planning to take a brief break, but I will be back!) 00:00 Introduction and series recap 01:07 Why distill a reasoning model? 02:51 Soft vs. hard distillation 11:06 On-policy vs. off-policy distillation 18:09 Preparing data for reasoning distillation 22:35 Loading the DeepSeek R1 MATH dataset 28:04 Inspecting reasoning traces and final answers 31:27 Tokenization and chat templates 35:58 Pre-tokenizing the training dataset 39:01 Sequence lengths and the 2,048-token limit 42:38 Creating training and validation splits 45:32 Loading the Qwen3-0.6B base model 46:24 Computing answer-token log probabilities 54:10 Cross-entropy loss for distillation 57:25 Training and validation loss helpers 1:02:03 Implementing the distillation training loop 1:07:26 Running the standalone training script 1:09:41 Plotting training and validation losses 1:11:31 Evaluating checkpoints on MATH-500 1:14:09 Comparing teachers and distillation results 1:16:34 Future directions for reasoning models 1:19:41 Bonus exercises and next steps

Today we're thrilled to announce the release of our first frontier visual reasoning model https://t.co/PrCJ0rsICk

We built a robotics data engine with Astra-level performance at 125x lower cost. And we are open-sourcing the weights today! Introducing Flora, a suite of open models to parse the physical world and turn raw footage into rich training signals. Flora already powers our in-house data engine, labeling 200K hours of videos a day on 500 GPUs. We created it to scale the frontier for robot learning. Now you own the weights too. Flora gives you: - High-quality, timestamped captions to curate and filter human egocentric and robot data; - Dense progress rewards for quality control and RL post-training; - Rubrics scoring for real-world benchmarking Try it today, links in thread:

18 models. One leaderboard. The biggest names in AI go head-to-head inside Hermes. The Hermes Index now compares 18 models across four benchmark suites, helping you see how they perform on real agent tasks. And a reminder: Hermes works with most of your AI subscriptions and providers, including Nous Portal, ChatGPT, Claude, OpenRouter, OpenCode Go, and many more. Connect your preferred providers to Hermes, then use the Index to help decide which models to run.

18 models. One leaderboard. The biggest names in AI go head-to-head inside Hermes. The Hermes Index now compares 18 models across four benchmark suites, helping you see how they perform on real agent tasks. And a reminder: Hermes works with most of your AI subscriptions and providers, including Nous Portal, ChatGPT, Claude, OpenRouter, OpenCode Go, and many more. Connect your preferred providers to Hermes, then use the Index to help decide which models to run.

By every measure, Opus 4.8 beats human experts as peer reviewers in a study of 124 Harvard social science papers covering a range from statistical analyses to formal models to qualitative work. As academics, we need to figure out how to adapt the practice of science to AI, fast. https://t.co/a9Q1pMd6E6

incredible work. im in awe of anthropic’s pace, and am grateful for the transparency of process+results i went through 3 out of the 5 case studies they released and wrote down my thoughts on each. in italics are LLM-summarizations for the cases for those who havent read them https://t.co/bLJoqExbjc

+1 more

EAGLE-4 is here. From EAGLE-1 to EAGLE-3, we’ve kept making inference faster. EAGLE-4 is our fastest yet. 961.6 tokens/s on @Alibaba_Qwen’s Qwen3.8-27B with TokenSpeed (BF16, 4× @NVIDIA B300, TP4, one request, T=0). Github: https://t.co/GwocZk1UTU https://t.co/GawgFNXKWj

Your OCR says it’s done—but did it get everything right? 🤔 In high-stakes workflows, even a small error can be costly. 🚀 Meet the MiniCPM-OCR suite: an open-source solution to parse fast and guard the results. Two new models: 🔹 MiniCPM-OCR 1.0 — SOTA OCR performance and speed at 1B scale. Compact, accurate, and fast. 🔹 MiniCPM-OCR Jev — A 2B decision model that helps assess parsing accuracy, with performance approaching that of Qwen3.5-397B-A17B on JevOCRBench. ⚡ What’s new? Small model, SOTA results: MiniCPM-OCR 1.0 scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2. A sharp goalkeeper 🛡️: MiniCPM-OCR Jev is model-agnostic, checking text, formulas, and tables from different OCR parsers. It achieves 93.6 on JevOCRBench, outperforming JEV-35B-A3B (90.1) and approaching Qwen3.5-397B-A17B (93.8). Up to 128 pages per minute 🤯: With DSpark, MiniCPM-OCR runs at 1.49× the speed of PaddleOCR-VL-1.6 and 3.62× that of MinerU2.5 Pro. The SDK’s concurrency optimizations further boost throughput to as much as 128 pages per minute. Deploy where you work: Day-0 support for Arm, Houmo AI, Intel, and NVIDIA, plus 10 more chip vendors through FlagOS. Ready for your stack: vLLM, SGLang, llama.cpp, and Ollama, with quantized models, fine-tuning support, and Agent Skills. 🔗 GitHub: https://t.co/2SlATAJk08 🤗 Hugging Face: MiniCPM-OCR 1.0：https://t.co/i347di4Fju MiniCPM-OCR Jev：https://t.co/ayVzhCJenh 👉 Try the Demo：https://t.co/QsFAJT3AFy Use the API：https://t.co/dP01jBS2j2

+4 more

New in CodeRabbit Secure: Deep Scan now follows a vulnerability across your linked repos, then replays the exploit in a throwaway sandbox to prove it's real before it flags anything. Hendrik walks through this and everything else in the new update: https://t.co/i6ybcOrtGd

Introducing RL with Neural Feedback (RLNF). We have found reliable ways to train AI models to respond to human preferences purely from brain signals. The example below is a picture of a face that self-transforms into a more attractive version for the subject as they are looking at it.

Learn how to build evals, folks! It's one of the pillars of the intelligence stack. Agents are hungry for data and can get it in many ways, but you can't build anything useful in the real world without proper evaluation. Evals make agents useful and let you build truly differentiated products. Invest heavily here. Or reach out if you need help. We are also building evals to measure the frontier @dair_ai.

The secret to Modal's speed is our custom lazily loaded container filesystem, ImageFS. ImageFS is built on a massive tiered distributed filesystem, layering local NVMe disks, regional and p2p caches, and global blob stores. When we launched VM Sandboxes this month, we wanted to make sure they could take advantage of ImageFS in the same way our gVisor containers do. To do that, we built out a new filesystem stack that tricks a VM's ext4 driver into thinking it's talking to a normal disk.

We ran @mirendil's new base model on Baseten's private codebases; that is, code it's never seen. Mirendil-Base-Large had the lowest bits per byte (a tokenizer-independent measure of perplexity) of any model we tested. We're excited to be the inference partner to serve Mirendil models. Congrats to the Mirendil team on this launch!

curious people and people interested in llm evals, benchmarks, high data circles, please wake up. my blog post "SWE-bench Multimodal v2 is not multimodal" is up now. https://t.co/Kqd1oOExIn https://t.co/Vd4nPqjiTx

Evals built by @PrimeIntellect residents, fully open: - MazeBench by @patience_cave - PMPP-Hard by @myainotez - PretrainGym by @HuNtEr_J1324 - OpenTTD-Bench by @thesophiaxu [and even more :) ] Super happy to support such amazing work, check it out!