AWS Adds India-Based Inference for Kimi K3 on Amazon Bedrock
AWS now supports India-based inference for Kimi K3 on Amazon Bedrock, enabling local workloads with data residency requirements.
735 articles
New AI models as they ship — large language models, multimodal systems, and open-weight releases. Capabilities, context windows, pricing, and availability, summarized so you can compare what is actually new.
AWS now supports India-based inference for Kimi K3 on Amazon Bedrock, enabling local workloads with data residency requirements.
OpenAI models are being used to support engineering tasks such as design, simulation, and manufacturing, with early results showing improved iteration speeds.
Google has expanded its Dreambeans AI app to India, offering personalized visual stories based on users' digital activity, with support for 13 local languages.
Google Cloud announced the availability of Gemini 4 Argon and expanded model options, including Claude and OpenAI models, to enhance enterprise AI capabilities.
Nextcloud has adopted Olmo 3, an open-source model, to enhance digital sovereignty and transparency for its users, particularly public organizations.
Hugging Face introduces kniv v7, a family of models that extract entities, relations, and facts from text using user-defined schemas, without requiring retraining or additional LLM calls.
Anthropic updated its Claude models to include a new rule preventing the recreation of specific existing games, though it may not curb the broader trend of AI-assisted game copying.
Google released Gemini 4 Argon, a reasoning model, and Gemini 4 Carbon, its strongest model for cybersecurity, in October 2026, alongside other AI updates.
AI agents like Muse and Dots are transforming how users interact with the internet, automating tasks such as booking appointments and managing finances.
NVIDIA has released an open-weight model for multilingual transcription, speaker attribution, and translation, along with updated data tools, on Hugging Face and ModelScope.
NVIDIA's new open-weight speech model enables transcription, speaker attribution, and translation with a single prompt, available on RTX Spark laptops.
Chinese AI labs are now primarily using OpenAI's GPT-6 Astra model for training, following Anthropic's enhanced security measures.