// Models
Deploy the open cloud models you already rely on — with vision, tool use, and reasoning.
GLM-5.2 is Z.ai's flagship model for the era of long-horizon tasks.
MiniMax M3: Coding & Agentic Frontier. 1M context window. Native multimodality.
Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.
NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.
DeepSeek V3.2 — strong general reasoning and coding with open weights and a large context window.
Qwen3 Max — Alibaba's largest open model for agentic, multilingual, and tool-use tasks.
Open-weight 120B reasoning model tuned for agentic workloads and tool calling.
Meta Llama 4 Maverick — a multimodal mixture-of-experts model for chat and vision.
Qwen3 Embedding — multilingual text embeddings for retrieval, search, and RAG.
EmbeddingGemma — a compact, efficient open embedding model from Google.