Nous Research retrained Alibaba's Qwen3.8-27B to produce Arete-27B, largely stripping out its pro-Chinese bias. The propaganda rate dropped from 22 percent to less than one percent, while general capabilities remained intact in tests. The method relied on 53 million tokens of supervised fine-tuning. The political bias is "applied late, it is applied thin," the team writes.
Ricursive Intelligence is releasing Ripple-65B Preview, a 65-billion-parameter open-weight language model that the company says achieves the top score among open-source U.S. models on the TerminalBench 2.1 coding benchmark at a total R&D cost of $32 million. On other benchmarks, however, Ripple falls behind the likewise newly released western open model Inkling 2 Small.
Thinking Machines, the startup from former OpenAI CTO Mira Murati, is releasing Inkling 2 Small, a mixture-of-experts model with 594 billion parameters. The company claims it's the strongest western open-weight model. Better agentic performance comes at a price, though, with token usage rising sharply.
Despite its small size, Underdog's new model can generate both speech and music.
Inside Inkling 2 Small’s training run and METR’s study of AI research at Anthropic.
With GPT-6 Cyber, OpenAI is introducing an AI model built to find security vulnerabilities in software, develop exploits, and suggest patches. According to OpenAI, it has already found 81 previously unknown vulnerabilities in Firefox and developed 948 working exploits in the Linux kernel, significantly more than its predecessor.
NVIDIA Cosmos world model development platform adds four new open-weight checkpoints to its Cosmos 3 model family for generating interactive world simulation and multi-sensor multi-view driving scenes. All four models are post-trained examples of Cosmos 3 Nano and have 16B trainable parameters. The models use OpenMDW1.1, permitting commercial and non-commercial use under its terms.
Baidu releases ERNIE 5.5 Flash Base: a 280-billion-parameter MoE model pre-trained on 33.5 trillion tokens, shipped deliberately without fine-tuning, so developers can take it wherever they want.
Google Deepmind is making Gemini Omni 1.2 Flash generally available. The video-generation model now creates clips up to 15 seconds long, can be edited step by step via conversation, and costs just $0.07 per second at 720p, 30 percent less than its predecessor. Users who draft in low resolution and upscale only the final result can save even more.
Gemini Omni 1.2 Flash brings 15-second clips, stronger consistency, and faster, more cost-effective generation. Today, we’re rolling out Gemini Omni 1.2 Flash, and making it generally available across the Gemini API and Gemini for business.
Introducing GPT‑6 Cyber, OpenAI’s cybersecurity model for finding exploitable vulnerabilities and developing fixes, trained to stay within authorized scope.
OpenAI is launching GPT-6 Cyber, the latest version of its cybersecurity AI model, which it says is better at finding vulnerabilities and less likely to go rogue and hack other platforms in the process.
OpenAI's newest cybersecurity model comes out as AI labs face scrutiny of their work to protect digital infrastructure.
Tencent has unveiled Hy4, its first multimodal language model that understands images and can operate software on its own. On the Artificial Analysis Intelligence Index, it scores 35 points, solid but behind Kimi K3.1 Pro. On hard math and complex image analysis, Hy4 still trails the best commercial models by a wide margin, according to Tencent itself.
Two weeks ago, we opened up Phonon, our small text-to-speech model that runs on your own device. It needs no server and no network, and what you type stays with you.
A 2-billion-parameter model that extracts exactly the fields you ask for, from any document type, outperforming every frontier LLM we tested, at a fraction of the cost.
OpenAI's hardware team used GPT-6 Astra to design a device for the Codex coding agent in under a week, handling everything from the enclosure to the circuit board to the firmware. Several revision rounds were needed before production because the model didn't always account for manufacturing tolerances.
NVIDIA Nemotron 3.5 Super VL is a 120B-parameter mixture-of-experts model that pairs reasoning and tool use with multimodal understanding. This post explores its architecture and use cases, then shows how to deploy it on Amazon EKS, EKS Auto Mode, and Amazon SageMaker HyperPod EKS with NVIDIA Dynamo and vLLM.
Mistral says its new model can afford organizations a kind of control over its cyber defenses that closed models can't.
Mistral has released the weights for Large 4, aka "Le Chonk," a trillion-parameter model trained entirely in Europe. Mistral says it beats its open rivals at coding, and independent testing shows it's fast and fairly cheap. It still can't match China's top models, scoring 26 points to the 32 posted by DeepSeek V4.1 Flash and GLM-5.3.
Mistral Large 4 (aka Le Chonk) is a multimodal, multilingual, Mixture-of-Experts (MoE) model from Mistral. It takes text and images as input and generates text. It has about 1T total, 49B active parameter, and a context window of 1M tokens.
The weights of Mistral Large 4 (Le Chonk) are now open. A 1T parameter multimodal model, and best open-weight model for coding, trained in Europe.
Just over a week ago, we released Granite Switch 4.1 , showing that our modular Switch architecture lets you configure a language model with specialized, cache-efficient adapters — adding new expertise on a single endpoint. Granite Switch 4.1.1 is a patch release.
Google’s new frontier model leads independent evals, with access initially limited to trusted defenders.
Just two weeks after Gemini 4 Argon, Google DeepMind is releasing Gemini 4 Carbon, which tops the Artificial Analysis Intelligence Index v5 with 59 points. Google says the model is built for extended agent workflows and the full cyber defense lifecycle. Because of those advanced security capabilities, access is limited to selected security experts at first.
A new decoy counting benchmark exposes a gap between seeing and reasoning. Elorian’s Glimpse tries to bridge it by thinking with images.
Gemini 4 Carbon is essentially a slightly better version of Argon.
Google has released its latest frontier model, Gemini 4 Carbon, offering "significant gains" in coding, agentic tasks, and cybersecurity. In an announcement on Wednesday, Google highlights Carbon's top spot on the Artificial Analysis Intelligence Index (v5.0) benchmark, with Claude Opus 5.5 and GPT-6 Astra trailing behind.
Nvidia has released Nemotron 3.5 Transcribe, an open speech recognition model that handles transcription, speaker recognition, and translation and takes its instructions as text prompts. It leads several benchmarks and uses only a sixth of the active parameters of its strongest open rival.
Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities. Carbon ranks #1 on the recently updated v5.0 of Artificial Analysis Intelligence Index at 59.
One open model for multilingual transcription, speaker attribution, translation, and audio Q&A, with top leaderboard results, plus the open Granary v2 data pipeline and fine-tuning recipes to adapt it.
Nvidia just released a free speech-to-text model that can transcribe up to 40 languages, but it only works on recordings, not live audio, so it’s a big step for meeting and medical transcription tools while still leaving real-time use to rivals like ElevenLabs.
Genie 4 will roll out to developers and Google AI Ultra subscribers soon.
Google Deepmind has introduced Genie 4, a real-time world model that simulates interactive 3D worlds with sound and talking characters. Gemini models help plan in the background to keep the simulation consistent. Genie 4 is still an early research model, and longer sessions lead to visual inconsistencies.
LFM3 is our new generation of Liquid Foundation Models (LFMs), built to be the starting point for specialized intelligence. It brings text, vision, audio, omni, and decision models onto one shared backbone, all in the 2B parameter range.
Genie 4: our new world model unlocking unprecedented realism in real-time simulation.
Google announced today its Genie 4 world model for generating interactive experiences using AI.
Google DeepMind released Genie 4, a world model with real-time audio, more realistic simulations, and mid-simulation text prompts, in research preview for trusted testers.
OpenAI's Ted Sanders explains where GPT-6.1 Terra fits in the company's GPT lineup. The smaller model outperforms the larger Astra in technical communication and costs far less but trails it in most other areas. Sanders says model size, reasoning effort, and multi-agent systems are distinct dimensions that can’t be reduced to a single "IQ score."
GPT-6.1 Terra It's always interesting to note which aspects of a new model are emphasized in the release announcement. In the case of GPT-6.1 Terra, OpenAI went with visual reasoning and computer use: We’re introducing GPT-6.1 Terra , a fast, efficient model that combines detailed visual reasoning and precise computer use alongside clearer, more natural writing.