
botanu
The complete cost engine for AI builders
3 followers
The complete cost engine for AI builders
3 followers
LLM tools show cost per API call. But if you're pricing per resolution or per conversation, you need cost per outcome, across your full stack. botanu maps models, agents, RAG, vector DBs, and infra to business outcomes: cost per resolution by feature, margin bands, vendor-to-feature breakdown, multi-agent costs, and customer-level economics. See live demo: venturedemo.botanu.ai




Check out the botanu-sdk: https://github.com/botanu-ai/botanu-sdk-python
Installation: pip install botanu
How the SDK works -
• An event represents business intent
• Each attempt is a run, with its own run_id
• All runs are linked via a shared event_id
• A single outcome (success / failure / partial) is emitted per event
• Cost per outcome = cost of all runs and attempts, including failures
• Context propagates across services via standard W3C Baggage
We're building toward outcome-based pricing for AI systems and a clearer understanding of true cost per outcome.