Foundation Model Hub & Inference Catalog
Verified open-weight foundation models ready for metered cloud inference via our unified OpenAI-compatible endpoint or self-hosted deployment on your GPU clusters.
DeepSeek V4 Pro
DeepSeek
Frontier 1.6-trillion parameter sparse MoE model engineered for advanced multi-step algorithmic reasoning, system architecture, and formal verification.
DeepSeek V4.1 Flash
DeepSeek
Ultra-low-latency high-throughput model with native multimodal reasoning and sub-200ms TTFT for real-time autonomous agent loops.
Kimi K3 (Moonshot)
Moonshot AI
Moonshot AI flagship 2.8T MoE architecture featuring Kimi Delta Attention (KDA) with 1,000,000 token context window for full-repository analysis.
Qwen 2.5 Coder 32B
Alibaba
The premier open-weights coding model. SOTA on HumanEval, SWE-bench, and multi-file code editing, beating proprietary code models.
Llama 4 Maverick
Meta
Meta frontier open foundation model with native multimodal reasoning, instruction-tuning, and calibrated enterprise tool calling.