Seed2.0 Mini
Seed2.0 Mini is the smaller general-purpose agent model in the Seed2.0 series, intended for low-latency, high-concurrency and cost-sensitive workloads. The official series announcement is dated…
MODEL DIRECTORY
New releases, in one place. Explore models by their publication date.
Seed2.0 Mini is the smaller general-purpose agent model in the Seed2.0 series, intended for low-latency, high-concurrency and cost-sensitive workloads. The official series announcement is dated…
Seed2.0 Lite is a general-purpose reasoning and agent model positioned for high-frequency enterprise workloads, balancing performance and cost. ByteDance describes the Seed2.0 series as improving…
Seed2.0 Pro is ByteDance’s flagship general-purpose reasoning model for complex, long-horizon agent workflows. The specified Volcengine ModelArk version supports deep reasoning, text generation,…
GPT-5.3-Codex-Spark is a smaller version of GPT-5.3-Codex designed for low-latency, real-time coding in Codex. OpenAI released it as a research preview on February 12, 2026. (openai.com) Supports…
M2.5-Lightning is the faster serving version of MiniMax M2.5, announced on February 12, 2026. MiniMax states that the two versions have identical capabilities but different speeds: Lightning is…
MiniMax-M2.5 is an open-weight language model designed for coding, tool-using agents, search and office productivity. MiniMax announced it on February 12, 2026. (minimax.io) Plans and develops…
GLM-5 is an open-weight foundation model designed for complex software systems engineering and long-horizon agentic tasks. It was released on February 12, 2026. The model has 744 billion total…
GPT-5.3-Codex is an OpenAI model optimized for agentic coding in Codex and similar environments. OpenAI announced it on February 5, 2026; the Responses API release followed on February 24, 2026. At…
Claude Opus 4.6 is an Anthropic model for complex reasoning, agentic coding and long-running knowledge-work tasks. It was released on February 5, 2026. Anthropic now lists it as an active legacy model…
GLM-OCR is a lightweight, 0.9-billion-parameter multimodal model specialized in optical character recognition and document parsing. Its encoder–decoder design combines a CogViT visual encoder with a…
Kimi K2.5 is a native multimodal model designed for visual understanding, coding and agentic tasks. It was released on January 27, 2026. Moonshot AI built it on Kimi K2 through continued pretraining…
DeepSeek-OCR 2 is an open-weight vision-language model for optical character recognition and document understanding. DeepSeek announced it on January 27, 2026. It introduces DeepEncoder V2, which can…
MiniMax Speech 2.8 is a text-to-speech model family focused on natural, expressive synthesized speech. MiniMax announced it on January 23, 2026; the supplied release identifier is Speech-2.8. The…
Qwen3-Max-Thinking is Alibaba Qwen’s flagship reasoning model for text generation and complex problem-solving. Alibaba Cloud lists qwen3-max-2026-01-23 as a January 23, 2026 release and snapshot. The…
GLM-4.7-Flash is a lightweight, open-weight GLM-4.7-series model intended to balance performance and deployment efficiency. It uses a mixture-of-experts architecture with approximately 30 billion…
GLM-Image is an open-weight image-generation model designed for instruction-following and knowledge- or text-dense visuals. It combines a 9B-parameter autoregressive generator with a 7B-parameter…
Qwen-Image-Max is a Qwen text-to-image generation model. The specified model ID is the December 30, 2025 snapshot; Alibaba’s release list records it under December 30, 2025. (docs.modelstudio.console.…
MiniMax-M2.1 is an open-weight language model focused on complex real-world tasks, particularly multilingual software development and agentic workflows. MiniMax announced it on December 23, 2025.…
Gemini 3 Flash Preview is a multimodal reasoning model designed to combine Gemini 3's reasoning capabilities with the speed and cost efficiency of the Flash family. Google announced it on December 17,…
Positioning: A reference-to-video model in Alibaba’s Wan2.6 visual generation series. It combines reference media with a text prompt to generate new scenes featuring the referenced subjects. Alibaba…