Models

Open MoE flagship with million-token context for coding and long agent runs

Multimodal Kimi workhorse for agent loops, coding tasks, and visual context

Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work

Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution

Tencent Hy reasoning model for coding, instruction following, and agent tasks

Balanced Mistral model for enterprise assistants, multilingual work, and tools

Qwen vision-language model for visual reasoning, documents, and agent tasks

Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.

StepFun flash lane for quick multimodal reasoning and coding assistance

Tencent Hy reasoning model for coding, instruction following, and agent tasks

Qwen vision-language model for visual reasoning, documents, and agent tasks

Thinking Kimi model for slower research passes, planning, and hard technical questions

Agentic coding model from Poolside in the XS size class for local deployment

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy