Memory, chatbots, and inference in one calm workspace with simple APIs and low-cost defaults.
What did we decide about the API rate limits?
Write endpoints stay at 600 RPM per org. Read endpoints can burst to 1200 RPM for 30 seconds when the workspace is healthy.
Start with the product you need today, then connect the rest when your AI workflow gets more ambitious.
Persistent agent memory with search, encryption, and a knowledge graph that grows with every conversation.
Train document chatbots on PDFs, URLs, and text sources, then embed them anywhere with a tiny script tag.
OpenAI-compatible Kimi K2.6 inference on Gonka with streaming, tool calling, and low-cost defaults.
Use the same primitives across memory, chatbots, and inference.
curl https://api.dogecat.ai/v1/chat/completions \
-H "Authorization: Bearer $DOGECAT_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"moonshotai/Kimi-K2.6","messages":[{"role":"user","content":"Hello"}]}'Free tier on all products. No credit card required, and every pathway has a real API behind it.