ERNIE 5.5 Flash Base
ERNIE 5.5 Flash Base is a text-only, sparse mixture-of-experts base language model released on November 5, 2026. It is the checkpoint at the end of pre-training, before mid-training or post-training.…
MODEL DIRECTORY
New releases, in one place. Explore models by their publication date.
ERNIE 5.5 Flash Base is a text-only, sparse mixture-of-experts base language model released on November 5, 2026. It is the checkpoint at the end of pre-training, before mid-training or post-training.…
GPT-6 Cyber is a cybersecurity-specialized variant of GPT-6 Astra designed for authorized security work. OpenAI announced it on November 5, 2026. Its post-training targets both cybersecurity…
HyPDE is a 323-million-parameter, fused-history transformer foundation model for forecasting two-dimensional partial differential equation (PDE) simulations. Tencent says it was developed with Hy4 and…
Tencent Hy4 is Tencent’s general-release large language model for reasoning, agentic workflows and real-world productivity. Tencent announced it on November 4, 2026. It uses a mixture-of-experts…
GPT-Image-2.6 Flare is OpenAI’s smaller, speed-oriented image generation and editing model, intended for interactive workflows and rapid visual iteration. OpenAI announced Images 2.6 on November 4, 20…
GPT-Image-2.6 Sunburst is OpenAI’s image generation and editing model for detailed compositions and precise visual work. It accepts text and image inputs and produces images. OpenAI positions Sunburst…
Apollo World is an agent-driven, multimodal world model for physical AI, focused on simulating environments for autonomous-driving training and validation. Baidu first disclosed it on November 3, 2026…
HyCell is a 16.1-million-parameter single-cell transcriptome embedding model released on November 2, 2026, with weights and code as part of Tencent’s Hyra research artifacts. It was pretrained without…
GPT-6 Rosalind Research applies GPT-6 reasoning to multistep life-sciences research, with safeguards designed for biological work. OpenAI describes specialized biological reasoning and scientific tool…
GPT-6 Luna (October 29) is an update to the GPT-6 Luna model released in ChatGPT on October 7, 2026. OpenAI published the update’s system-card addendum on October 29, 2026. The update is intended to…
Gemini 4 Carbon is a frontier reasoning model based on Gemini 4 Argon, designed for complex, long-running agentic workflows, software engineering, scientific reasoning, and cybersecurity. Google…
Genie 4 is Google DeepMind’s natively multimodal world model for generating and exploring interactive simulations in real time. It was announced on October 27, 2026, as a research preview for trusted…
Qwen-Audio-3.1-Realtime-Flash is a real-time duplex speech conversation model. It accepts text and audio and produces text and audio. It supports function calling, web search and voice cloning. The…
Baidu released and open-sourced 文心 Unlimited OCR 2 on October 27, 2026. It is a multimodal document-parsing model designed for long, multi-page documents. Baidu’s Qianfan documentation describes it as…
GPT-6.1 Terra is a fast, efficient model for visual reasoning, computer use and natural writing. It accepts text and image inputs and produces text outputs. It supports configurable reasoning effort,…
Hy-SimulMT-30B-A3B-Interleave is Tencent Hunyuan’s open-weight model for low-latency simultaneous translation. It has approximately 30 billion parameters and is distributed under the Apache 2.0…
Tencent Hunyuan’s open-weight model for real-time, bidirectional Chinese–English simultaneous speech translation. It receives streaming audio and produces translated text while the speaker is talking.…
Kimi K3.1 Pro Highspeed is the faster-serving version of Kimi K3.1 Pro, a 3-trillion-parameter-class, open-weight model developed through additional post-training of Kimi K3. Moonshot AI says…
Kimi K3.1 Pro is the first model in the K3.1 series and an open-weight, multimodal model built by post-training Kimi K3. Moonshot positions it for complex coding, visual understanding, research, and…
WeVisOne-VL-4B is an open-weight vision-language model from Tencent’s WeChat Vision team, designed to unify object-level and pixel-level visual perception. It builds on Qwen3-VL-4B-Instruct and uses a…