{"id":15288,"date":"2026-10-01T03:37:19","date_gmt":"2026-10-01T03:37:19","guid":{"rendered":"https:\/\/playgama.com\/blog\/?p=15288"},"modified":"2026-10-01T03:37:19","modified_gmt":"2026-10-01T03:37:19","slug":"how-to-architect-llm-powered-npc-dialogue-in-a-browser-game","status":"publish","type":"post","link":"https:\/\/playgama.com\/blog\/business-faqs\/how-to-architect-llm-powered-npc-dialogue-in-a-browser-game\/","title":{"rendered":"How to architect LLM powered NPC dialogue in a browser game?"},"content":{"rendered":"

Never call the LLM API directly from client code – proxy it through your own backend to hide the API key, cap and summarize conversation history per NPC to control token cost, and keep generation calls async so they don’t block Canvas\/WebGL rendering.<\/strong><\/p>\n

At a glance<\/h2>\n\n\n\n\n\n\n\n
Fact<\/th>\nValue<\/th>\nSource<\/th>\n<\/tr>\n<\/thead>\n
API key must never sit in client JS<\/td>\nserver proxy required<\/td>\n<\/td>\n<\/tr>\n
Conversation memory fades without full history<\/td>\nresend or summarize context<\/td>\nar5iv.labs.arxiv.org<\/a><\/td>\n<\/tr>\n
In-browser inference needs WebGPU support<\/td>\nWebLLM + WebGPU browser<\/td>\nllm.mlc.ai<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

Put the LLM call behind your own backend, never in the browser bundle. A browser game’s JS is fully visible to players, so any API key shipped client-side will be extracted and abused within hours. Your server holds the key, forwards the player’s message plus the NPC’s context to the model, and returns only the text (or audio) the client needs to render. This also lets you rate-limit, cache repeated prompts, and swap providers without a client update.<\/p>\n

If you build or iterate on your browser game using an AI coding agent like Codex, Claude Code, Cursor, or VS Code, the Playgama MCP server connects your agent directly to the Playgama developer cabinet to upload builds and publish sandboxes.<\/p>\n

What decides the architecture for a browser game specifically<\/h2>\n