X plans to publish the code behind its recommendation and advertising ranking systems, promising regular updates and increased transparency.
Qira is a cross-device personal AI super agent unveiled by Lenovo and Motorola at CES 2026, built to unify workflows across PCs, phones, tablets, and wearables.
OpenAI is intensifying its audio AI push with a new model and audio-first devices planned for 2026, aiming to make voice the primary AI interface.
OpenAI introduced GPT-5.2-Codex in 2025 as an agentic coding model for enterprise refactors, terminal-based workflows, and defensive cybersecurity research.
Gemini 3 Flash now defaults in Gemini and Search AI Mode, delivering fast curated answers with links, while classic Search remains best for source verification.
Google reimagined Gemini Deep Research on Gemini 3 Pro and opened it to developers via the Interactions API, while Gemini Advanced adds a supervised web-research workflow with Google Docs export.
ChatGPT 5.2 launched on Dec 11, 2025 with stronger multi-step reasoning, near-perfect long-context recall to 256k tokens, and improved reliability metrics.
Google enables Deep Think mode for Gemini 3, delivering improved reasoning benchmarks, parallel hypothesis processing, and exclusive access for AI Ultra subscribers.
Chinese startup DeepSeek has launched 685B-parameter models V3.2 and V3.2-Speciale, matching or surpassing GPT-5 and Gemini-3.0-Pro on major math and coding benchmarks. A 2025 technical report on Hugging Face details the DeepSeek Sparse Attention architecture, which cuts 128K-token inference costs by about 70% compared with the prior V3.1-Terminus model.
