Free LLM chatbot stack

A chat assistant with memory: a model, a place to run it, and a store for conversations.

3 roles3 with no end date1 of 3 need no card

Role	Pick	What you get
LLM API The model behind the replies.	LLM API Gemini API free tier Google · No end date · no card Flash models, free of charge	Flash models, free of charge
Runtime Streams replies to the browser.	Runtime Vercel Hobby Vercel · No end date · card not stated 100 GB transfer, 2 projects	100 GB transfer, 2 projects
Conversation store History and per-user memory.	Conversation store Upstash Redis Free Upstash · No end date · card not stated 500K commands a month	500K commands a month

Limits side by side

LLM API

Google Gemini API free tier

Price: Free of charge input and output (Flash) · Rate limits: RPM, TPM and RPD per project · Reset: Daily quota resets at midnight Pacific · Data use: Free-tier content used to improve Google products · Batch, Flex: Not available on free tier

Runtime

Vercel Hobby

Edge requests: 1M / month · Data transfer: 100 GB / month · Projects: 2 included projects · Blob storage: 1 GB / month

Conversation store

Upstash Redis Free

Data size: 250 MB · Commands: 500K / month · Bandwidth: 10 GB / month · Databases: 1