Bernini
Bernini 7+14B is a unified image and video generation and editing framework. The full pipeline combines a Qwen2.5-VL-7B-based multimodal semantic planner with a Wan2.2-T2V-A14B diffusion renderer. It…
MODEL DIRECTORY
New releases, in one place. Explore models by their publication date.
Bernini 7+14B is a unified image and video generation and editing framework. The full pipeline combines a Qwen2.5-VL-7B-based multimodal semantic planner with a Wan2.2-T2V-A14B diffusion renderer. It…
Qwen3.7-Max is the largest model in the Qwen3.7 series, positioned as a flagship for agent-oriented tasks. Alibaba Cloud lists both qwen3.7-max and the snapshot qwen3.7-max-2026-05-20 as released on…
Hy-MT2-30B-A3B is Tencent’s open-weight multilingual translation model for real-world translation tasks. It uses a mixture-of-experts architecture with 30 billion total parameters and approximately 3…
Hy-MT2-7B is the 7-billion-parameter member of Tencent’s Hy-MT2 family of fast-thinking multilingual translation models. Tencent released its weights on May 21, 2026. The model card lists the Apache 2…
Hy-MT2-1.8B is Tencent’s lightweight, 1.8-billion-parameter, fast-thinking multilingual translation model for complex real-world scenarios. Tencent announced the open-source release on May 21, 2026.…
Qwen3.7-Max-Preview is a preview of the largest Max model in the Qwen3.7 family, offered through Alibaba Cloud Model Studio. It is optimized for general conversation, including knowledge question…
Gemini Omni Flash is the first model in Google’s Gemini Omni family. It combines Gemini’s multimodal intelligence with generative media capabilities, initially focusing on video creation and…
Gemini 3.5 Flash is a fast, natively multimodal reasoning model in the Gemini 3 series, based on Gemini 3 Flash. Google positions it for coding and complex agentic workflows. Adjustable thinking…
Qwen3.5-LiveTranslate-Flash-Realtime is a real-time multilingual audio and video interpretation model built on Qwen3.5-Omni. Alibaba Cloud lists the stable model as currently equivalent to the…
Lance is a research-oriented, unified multimodal model with approximately 3 billion active parameters. It uses a dual-stream mixture-of-experts architecture to support image and video understanding,…
WorldStereo 2.0 is the approximately 17-billion-parameter world-expansion component of Tencent’s HY-World 2.0 framework. It uses camera-guided generation and geometry-aware memory to produce…
HY-Pano 2.0 is Tencent’s panorama-generation model and the first stage of the HY-World 2.0 world-generation pipeline. It converts a text prompt or a single perspective image into a 360°…
ERNIE 5.1 is a general-purpose model derived from ERNIE 5.0, positioned for text generation, reasoning, knowledge-intensive work and agentic tasks. Baidu officially announced it on May 9, 2026. (ernie…
GPT-Realtime-Whisper is a streaming speech-to-text model for low-latency, live transcription. OpenAI announced it on May 7, 2026, and made it available through the Realtime API. (openai.com)…
GPT-Realtime-Translate is a streaming speech-to-speech translation model for live multilingual audio experiences. OpenAI announced it on May 7, 2026, for use through the Realtime API. (openai.com)…
GPT-Realtime-2 is OpenAI’s real-time speech-to-speech model for voice agents. OpenAI describes it as its first voice model with GPT-5-class reasoning. It was released on May 7, 2026, and is available…
GPT-5.5-Cyber is an OpenAI cybersecurity model for specialized, authorized security work. It was announced on May 7, 2026, initially as a limited preview for defenders responsible for critical…
Gemini 3.1 Flash-Lite is a low-latency, cost-efficient multimodal model designed for high-volume, relatively lightweight tasks. Google announced its general availability on Gemini Enterprise Agent…
Cola DLM is a research-oriented, hierarchical continuous-latent diffusion language model for text generation. The released checkpoint corresponds to the 2000 EFLOPs training-compute point, not a…
Released on May 5, 2026, GPT-5.5 Instant is an everyday conversational model introduced as ChatGPT’s default, replacing GPT-5.3 Instant. OpenAI also made it available through the API as chat-latest at…