One Interface for all AI models
Access every model through a single, OpenAI-compatible API. Self-host it, or use the managed service.
Featured Models
A curated selection from the shared model catalog
Nemotron 3.5 Lightning 30B A3B
NVIDIA
Fast 30B A3B MoE model optimized for specialized agentic tasks.
Muse Glimmer 30B
NVIDIA
Multimodal reasoning model for text and image inputs with tool-calling support, served through NVIDIA's OpenAI-compatible endpoint.
Riva Translate 4B Instruct v2
NVIDIA
Multilingual instruction-tuned translation model served through NVIDIA NIM.
Ising Calibration 1.5 31B
NVIDIA
Multimodal vision-language model for quantum-computing calibration experiment analysis.
MiniMax M3
MiniMax
MiniMax M3 Preview is a multimodal MoE vision-language model with strong reasoning, coding, and tool-calling capabilities.
DiffusionGemma 26B
Diffusion-based 26B parameter LLM enabling parallel token generation for real-time text apps.
Your control plane, your rules
1Router is fully open-source and self-hostable. Deploy the entire routing control plane on your own infrastructure — or use our managed cloud. Your choice.
Full Control
Run the entire control plane on your own infrastructure. No data leaves your network.
Privacy First
Your API keys, your usage data, your rules. Complete data sovereignty at every layer.
Open Source
Fully open-source codebase. Fork it, extend it, contribute back — no vendor lock-in.
Cloud or Self-Hosted
Use our managed cloud for zero setup, or deploy on your own K8s cluster with one command.
App Showcase
Public applications that opted in to share verified 1Router usage.