Gemini 2.5 Pro is a new generation of AI model launched by Google. It has "thinking ability" and conducts multiple steps of reasoning before responding, thereby greatly improving performance and accuracy.

ReasonGraph is an open source platform that visualizes and analyzes the inference process of large language models (LLMs), and supports 50+ mainstream models such as OpenAI, Google, and Anthropic.

gpt-4o-mini-transcribe is a speech-to-text model launched by OpenAI, and is a streamlined version of gpt-4o-transcribe.

CosyVoice 2.0 is a leading multilingual voice generation model, using streaming modeling technology to achieve ultra-low latency (150ms), and the sound quality is natural and stable.

HaiLuo Artificial Intelligence (a product of MiniMax) recently officially released its latest text-to-speech (TTS) model T2A-01-HD .

Qwen2.5-14B-Instruct-GGUF is an optimized large-scale language generation model that combines advanced technology and powerful instruction tuning with efficient text generation and understanding capabilities.

Tiangong Model 4.0 is online, with dual upgrades of reasoning and voice assistant. It is free and open, bringing a new AI experience!

Amazon Nova Premier is Amazon's new multi-modal language model that supports the understanding and generation of text, images, and videos, helping developers build AI applications.

InfAlign is a new model released by Google that aims to solve the problem of information alignment in cross-modal learning.

DeepSeek V3 is an advanced open source AI model developed by Chinese AI company DeepSeek (part of the hedge fund High-Flyer).

EleutherAI is an open source artificial intelligence research organization dedicated to developing and releasing large-scale language models similar to OpenAI's GPT model.

BLOOM-3 is the third generation in the BLOOM model series. It inherits the multi-language capabilities of the previous two versions and has been optimized.

Generate high-quality images based on text descriptions provided by users, and have flexible control options, suitable for art creation, visual design, advertising production and other fields.

A unified framework model launched by Google

FacebookAI/roberta-base is a pre-trained language model based on the RoBERTa architecture developed by the Facebook AI Research Team (FAIR).