Tag

Step-Audio: The First Product-Level Open-Source Speech Interaction Model by StepStar

Step-Video-T2V: StepStar's Open-Source Video Generation Model

Zonos is an open-source text-to-speech (TTS) model that delivers high-quality, natural voice generation, supports multiple languages, and features real-time voice cloning capabilities.

Mistral Small 3 is a new open-source language model developed by the French startup Mistral AI, featuring 24 billion parameters.

Qwen2.5-VL is the latest flagship vision-language model launched by Alibaba’s Tongyi Qianwen team, featuring significant technological advancements and a wide range of application capabilities.

Qwen2.5-1M is an open-source large language model developed by Alibaba Cloud's Tongyi Qianwen team, released in January 2025. It is designed to handle up to 1 million tokens of context.

The MiniMax-01 series, launched by Hailuo AI, comprises open-source large language models and vision multimodal models.

MiniCPM-o is a new series of edge-based multimodal large models designed to handle various inputs such as images, videos, text, and audio, and generate high-quality text and speech outputs.

Moondream is an innovative open-source visual-language model designed to provide efficient image processing and understanding capabilities.

CogAgent is a multimodal Vision-Language Model (VLM) jointly developed by Tsinghua University and Zhipu AI, designed specifically for understanding and interacting with graphical user interfaces (GUIs).

DeepSeek V3 is the latest AI model released by DeepSeek, aimed at advancing natural language processing and multimodal understanding capabilities.