Intern Large Models has released the Intern-S2-397B model, which is available on Hugging Face. The release includes day-0 support from vLLM for running the model.
The model is described as having multimodal, reasoning, coding, and scientific agent capabilities.
IBM and NASA have open-sourced the NASA-IBM Lunar Foundation Model, an AI system that consolidates decades of multi-modal lunar data from US and Japanese missions into a single representation. Built on IBM's TerraMind architecture, the model integrates observations at varying scales and viewing angles to address challenges like complex lighting and data resolution.
DeepSeek has released its new model, DeepSeek V4-1 Flash. It is a multimodal Mixture-of-Experts (MoE) model featuring a 552 billion parameter backbone.
inclusionAI has released the open-source Ling-3.0-flash-VL model, which extends the language and reasoning capabilities of its predecessor with native image and video understanding.
Researchers introduce NeoMME, a family of 260M and 800M multilingual multimodal encoders that process text and raw image patches using a single bidirectional Transformer. Unlike generative visual language models, NeoMME does not rely on separate pretrained vision towers or causal decoders, training the entire model from scratch with a masked discrete-diffusion objective.
DeepSeek has published the DeepSeek-V4-Flash-Vision-Exp model checkpoint on Hugging Face. The repository is now available for public access.
We use cookies to measure traffic and improve the site. You can accept or decline analytics cookies. Privacy policy