Model launchesStartup General Intuition says robotics is close to a 'ChatGPT' momentGeneral Intuition, led by CEO Pim de Witte, believes robotics will shift from bespoke robot-specific models to general-purpose foundation models trained on high-quality datasets.3 min read
Model launchesOpenAI launches GPT‑Live‑1 models to enable more natural, interruptible voice conversationsOpenAI introduced GPT‑Live‑1 and GPT‑Live‑1 mini, full‑duplex conversational speech models designed to listen and speak simultaneously, enabling natural interruptions, live translation and longer hands‑free sessions.3 min read
Model launchesOpenAI to Release GPT-5.6 in Three Tiers After U.S. Security ReviewOpenAI will publicly release its most advanced model, GPT-5.6, after postponing the launch last month at the U.S.2 min read
Model launchesGPT-5.6 Sol, Terra and Luna models available in public preview from ThursdayAccording to the announcement, the GPT-5.6 Sol and the Terra and Luna models will become publicly available on Thursday; preview access is also being expanded globally.1 min read
Model launchesMeta's Muse Image can use public Instagram posts to generate images, raising misuse and privacy concernsMeta has launched Muse Image, an AI image generator that can draw on photos from public Instagram accounts to create new images.3 min read
Model launchesOpenAI launches GPT‑Live for continuous, real‑time voice interactions in ChatGPT VoiceOpenAI has introduced GPT‑Live, a full‑duplex voice model family that lets ChatGPT Voice listen and speak simultaneously for more natural, interruptible conversations.4 min read
Model launchesNVIDIA releases open Isaac GR00T 1.7 platform for end-to-end humanoid robot policy developmentNVIDIA announced the Isaac GR00T Development Platform and the open-source GR00T 1.7 vision-language-action model to unify humanoid robot development from simulation and data collection through post-training, evaluation and deployment.5 min read
Model launchesOpen models advance as Anthropic–US tensions and government controls reshape AI landscapeSummer developments in AI have seen open-weight models close the gap with proprietary systems while regulatory pressure around Anthropic and other firms prompts new access controls.5 min read
Model launchesMeta launches Muse Image and previews Muse Video—agentic models for media creationMeta Superintelligence Labs released Muse Image, an agentic image-generation model that uses search, code, and self-refinement to improve factual accuracy and edit precision, and previewed Muse Video, a high-fidelity text-to-video model.4 min read
Model launchesTencent unveils Hy3 MoE large language model with 256K context lengthTencent's Hy Team has released Hy3, a Mixture-of-Experts large language model with 295 billion parameters (21B active) and a 256K token context window.2 min read
Model launchesRacka: a Hungarian-optimized, reasoning-capable experimental LLMRacka is a Hungarian-language large language model (Racka-4B) developed by researchers at Eötvös Loránd University’s Department of Artificial Intelligence with the Digitális Örökség Nemzeti Laboratórium and in collaboration with Mynds.ai.5 min read
Model launchesAnthropic introduces Claude 3.7 Sonnet with optional “extended thinking” and visible thought tracesAnthropic has released Claude 3.7 Sonnet (announced Feb 24, 2025), which adds an opt‑in “extended thinking mode” that lets the model spend more computation and time on difficult problems, and optionally shows its intermediate thought process.6 min read
