Credits are the unit of platform usage, metered against what costs us to run. Bring your own model keys to pay providers directly at provider pricing.
One unit. Predictable. Metered against actual platform cost.
Every run includes its full execution trace and append-only audit log, retained for your plan's window. Skills, knowledge bases, MCP tools, and webhook/schedule triggers are bundled by plan tier — never metered per run.
Bring your own model key. You pay the model provider directly at their public rates — Brahmalabs never marks up inference. We charge a flat LLM gateway fee of 0.1 credits per 1k tokens for routing, governance, failover, and cost tracking on every call.
Start from a workload close to yours, or itemise every line. The estimate updates live and tells you which plan fits — no sign-up, no sales call.
A busy testing team — browser test runs in 2 GB sandboxes, spec generation, screenshot artifacts.
In-app builder assistant — help authoring workflows & agents.
AI agent replies across WhatsApp · email · Instagram — billed as gateway tokens.
Inference is bring-your-own-key — you pay your model provider directly; Brahmalabs charges a 0.1 credit / 1k token gateway fee on tokens routed.
Skills, knowledge bases, MCP tools, and triggers are included with your plan — not metered per run, so they're not dials here.
Uses your LLM token volume above. Set your blended provider rate per 1k tokens.
Estimate only. Credits are metered against actual usage — your real cost depends on your real workload. Model spend (BYO) is paid to your provider at their rates.
Start on the Free tier today, or talk to us about volume pricing for high-throughput and regulated deployments.