QwenLive
Multimodal
Qwen3.6 27B
Qwen/Qwen3.6-27BA multimodal 27B model for language, vision, coding, and agent workflows.
Model catalogOpen models,
Open models,
carefully served.
One production route today. A focused expansion across language, multimodal, and image models is underway.
1 live10 coming soon
All models11
Moonshot AIComing soon
Language
Kimi K3
moonshotai/Kimi-K3The next generation of the Kimi model family for reasoning, coding, and agentic workloads.
Planned routeModel card ↗
Z.aiComing soon
Language
GLM-5.2
zai-org/GLM-5.2A flagship long-horizon language model with million-token context.
Planned routeModel card ↗
Moonshot AIComing soon
Multimodal
Kimi K2.7 Code
moonshotai/Kimi-K2.7-CodeA large multimodal coding model with native INT4 weights.
Planned routeModel card ↗
Moonshot AIComing soon
Multimodal
Kimi K2.6
moonshotai/Kimi-K2.6Visual understanding, reasoning, coding, and tool-driven workflows.
Planned routeModel card ↗
Moonshot AIComing soon
Multimodal
Kimi K2.5
moonshotai/Kimi-K2.5A multimodal agent model for text, vision, coding, and tool use.
Planned routeModel card ↗
DeepSeekComing soon
Language
DeepSeek V3.2
deepseek-ai/DeepSeek-V3.2Efficient long-context reasoning for agentic and general workloads.
Planned routeModel card ↗
MiniMaxComing soon
Multimodal
MiniMax M3
MiniMaxAI/MiniMax-M3Native text, image, and video understanding for long-context work.
Planned routeModel card ↗
OpenAIComing soon
Language
gpt-oss 120B
openai/gpt-oss-120bAn open-weight reasoning model for agentic and developer workloads.
Planned routeModel card ↗
QwenComing soon
Image generation
Qwen Image 2512
Qwen/Qwen-Image-2512Text-to-image generation focused on realism, natural detail, and text rendering.
Planned routeModel card ↗
QwenComing soon
Image editing
Qwen Image Edit 2511
Qwen/Qwen-Image-Edit-2511Instruction-guided image editing with improved subject and scene consistency.
Planned routeModel card ↗
Coming-soon models are planned additions, not currently available API routes.