ai-models

OpenAI's new full-duplex voice models listen and speak at once and hand complex queries to GPT-5.5, replacing Advanced Voice Mode in ChatGPT.

OpenAI moved its GPT-5.6 model family to general availability on July 9, after federal reviewers cleared the release that had restricted Sol, Terra, and Luna to trusted partners.

Mistral AI has unveiled Robostral Navigate, its first robotics model: an 8-billion-parameter navigation system that guides wheeled, legged, and flying robots with one RGB camera and language instructions, scoring 76.6% on the R2R-CE benchmark in unseen environments.

SpaceXAI's Grok 4.5, trained alongside Cursor, is priced at $2 per million input tokens and $6 per million output tokens; Musk calls it 'an Opus-class model, but faster, more token-efficient and lower cost.'

Anthropic released Claude Science, an AI workbench for scientists with more than 60 curated skills, in beta to paid Claude subscribers, and said it has started an internal program to develop treatments for neglected diseases.

OpenAI began a limited preview of its GPT-5.6 models — Sol, Terra, and Luna — restricting access to a small group of government-approved partners at the request of the US government.

Moonshot AI released Kimi K2.7-Code, an open-weight 1-trillion-parameter MoE coding model under a Modified MIT License, claiming roughly 30% lower thinking-token usage than K2.6 on self-run benchmarks.

Microsoft unveiled seven proprietary MAI models led by the 35B-active reasoning model MAI-Thinking-1, all trained from scratch without distillation as it reduces its dependence on OpenAI.

Google's new efficiency flagship outperforms Gemini 3.1 Pro on most evals while running 4x faster and costing 40% less.

The four-year commitment — described as the largest deal of its kind between an AI company and a global philanthropy — targets health services for 4.6 billion people in low-income countries.

Mistral released Medium 3.5, a dense 128B open-weight model with a 256k context window that consolidates Medium 3.1, Magistral, and Devstral 2 under a Modified MIT license, with a per-query reasoning toggle.

DeepSeek's V4-Pro and V4-Flash arrive with 1M-token context, three reasoning modes, and pricing that undercuts frontier rivals by up to 8x.