
LLM serving capacity planner: VRAM, KV cache, GPU topology, latency, and cost.
Streamable HTTP
https://tokcalc.vercel.app/api/mcpclaude mcp add --transport http tokcalc https://tokcalc.vercel.app/api/mcp@tokcalc/mcp-server · npm 0.3.4 · stdio
claude mcp add --transport stdio tokcalc -- npx -y @tokcalc/mcp-server@0.3.4