Models
Every model is called through one endpoint, model.graphn.ai, with your API key.
custom:cm-mock-2Deployingmistralai/Mixtral-8x7B-…——0custom:cm-mock-1Readynvidia/NVIDIA-Nemo…321 ms$6.50/h3,789imported:im-mock-1OpenAIBilled by your provider319 ms3,857deepseek-v4.1-flash1M··$0.30gpt-oss-120b128K····$0.15Embedding and reranking models, called through the same endpoint.
Deployments
Models you deploy run on dedicated GPUs and scale to zero when idle.
mistralai/Mixtral-8x7B-…Deploying20 runningscales 1–20—Mon09:12 AMnvidia/NVIDIA-Nemo…Ready11 runningscales 0–2612321 msWed04:37 PMPlayground
Copy requestLogs
Every inference request in this workspace, kept 365 days. Prompts and completions are not stored.
193 requests between 16:00:00 and 17:00:00 · scroll for older
Home





