OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.
3M Pulls 5 Tags Updated 1 week ago
DeepSeek-V3.1-Terminus is a hybrid model that supports both thinking mode and non-thinking mode.
79.3K Pulls 8 Tags Updated 3 days ago
Alibaba's performant long context models for agentic and coding tasks.
438.9K Pulls 10 Tags Updated 6 days ago
A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.
1,688 Pulls 1 Tag Updated 4 days ago