How much does a chatbot cost to run?
A chatbot is mostly one line: the LLM call. Scale is requests times token size.
100,000 requests / month · 700/350 in/out
LLM API
$1.36 /mogpt-oss-20bvia PrimeIntellect
Pick the cheapest model that clears your quality bar. Long system prompts? Cached-input pricing pays off fast.
next: Qwen3.6 35B A3B $3.92compare all ↗
◈ or self-host: 1× L40S at ~55% GPU use $314/mo· pays off above ~23.0M requests/mo
Observability
$0 /moArize Phoenix
Add tracing from day one - it's free until you're big.
next: Helicone $0compare all ↗
Total, your workload≈ $1.36 /mo
Tune it for your numbers ↗ Live prices, updated 2026-11-03. A model, not a quote - each layer is priced at its cheapest provider; embeddings amortise the initial index over 12 months. Confirm on each provider's site.