Wan3.0-Video
Wan3.0-Video is Alibaba’s all-in-one multimodal video-generation model. Alibaba Cloud Model Studio provides its inference service. It supports text-to-video, image-to-video with first-frame or…
MODEL DIRECTORY
New releases, in one place. Explore models by their publication date.
Wan3.0-Video is Alibaba’s all-in-one multimodal video-generation model. Alibaba Cloud Model Studio provides its inference service. It supports text-to-video, image-to-video with first-frame or…
SeedRealtime is a native audio-visual, full-duplex large language model for continuous real-time interaction. Its unified end-to-end architecture combines audio, video and text so that it can perceive…
Hunyuan3D-Buffalo 1.0 is Tencent Hunyuan3D’s unified multimodal research framework for 3D generation, understanding and editing. It combines a Hunyuan3D-VLM backbone with Hunyuan3D diffusion…
Qwen-Image-3.0 is the standard variant of Alibaba’s third-generation Qwen image model, balancing image quality and generation speed. It accepts text and image inputs and produces images, supporting…
Qwen3.8-Max is Alibaba’s flagship multimodal foundation model, announced on August 3, 2026. It uses a sparse mixture-of-experts architecture with 2.4 trillion total parameters and a hybrid attention…
Seed-Evolving is a deep-reasoning Seed-series model designed for coding and agent workloads. Its unified model ID automatically serves updated versions; it is not a fixed-version identifier.…
MiniMax H3 is a general-purpose multimodal generation model focused on video with synchronized native stereo audio. It understands combined text, image, video and audio context and supports…
Seedance 2.5 is ByteDance’s multimodal video-creation model, designed for longer narratives, reference-guided generation and video editing. The official announcement is dated 2026-07-31.…
DeepSeek-V4-Flash-0731 is the official DeepSeek-V4-Flash release, announced on July 31, 2026, when its API entered public beta. It is a text-generation model with substantially enhanced agent…
Qwen3.7-Text-Embedding is a multilingual text embedding model based on Qwen3.7. It converts text into numerical vectors for semantic search, retrieval, recommendations, clustering and classification.…
Gemini Robotics On-Device 2 is an efficient vision-language-action (VLA) model designed for general-purpose robotic manipulation and local inference on robotic devices. Its architecture builds on…
Gemini Robotics ER 2 Streaming Preview is a vision-language model endpoint for embodied reasoning in robotics, optimized for real-time text streaming through the Gemini Live API. Google released it in…
Gemini Robotics ER 2 Preview is a preview vision-language model for embodied reasoning in robotics, built on Gemini 3.5 Flash. It interprets multimodal information about physical environments and can…
Gemini Robotics 2 is Google DeepMind’s vision-language-action (VLA) model for robot control. It converts visual and language inputs into motor actions and can control humanoid bodies and dual-arm…
Qwen-Audio-3.0-ASR-Flash is a non-real-time automatic speech recognition model for converting short audio recordings into text. Alibaba Cloud lists its release on July 30, 2026. (alibabacloud.com)…
A non-real-time automatic speech recognition model for transcribing complete audio and video files into text. It is intended for uses such as video subtitles, customer-service quality inspection, and…
Qwen-Audio-3.0-ASR-Flash-Streaming is a real-time, streaming automatic speech recognition model designed for low-latency, high-concurrency interaction. It accepts audio and returns text. Alibaba lists…
Lyria 3.5 is Google’s music-generation model, optimized for full songs with coherent verses, choruses and bridges. It uses latent diffusion applied to audio representations. (aistudio.google.com)…
Gemini 3.5 Flash Cyber is a lightweight cybersecurity model built on Gemini 3.5 Flash and fine-tuned to find, validate, and patch software vulnerabilities. Google positions it as an efficient…
Gemini 3.5 Flash-Lite is a fast, cost-efficient, natively multimodal reasoning model based on Gemini 3.1 Flash-Lite. It is designed for high-volume, latency-sensitive tasks and agentic workflows.…