GH Repository · lemony-ai
cascadeflow
Cascading runtime for AI agents. Optimize cost, latency, quality, and policy decisions inside the agent loop.
- stars
- 3,950
- 30-day movement
- -6-2/day
- Related entries
- 61
- Connections
- 1
node/pnpmgoogle-adkmodel-cascadingPythoncost-transparencyvllmagent-apptogether-aiagentaipythonllmtypescriptanthropichuggingfaceapinodecost-optimizationbudgetsgptn8nclaudeollamaopenai
CascadeFlow is a cascading runtime for AI agents that sits inside the agent loop to optimize cost, latency, quality, and policy decisions. It ships as a Python package with a TypeScript/Node toolchain (pnpm) and references model-cascading and cost-transparency in its topics.
Reach for it when you want model cascading to route agent requests across providers with explicit budget and policy control.
Use it to
- Cascade requests across multiple LLM providers
- Enforce budgets and policy inside the agent loop
- Track cost transparency per agent run
- Trade off latency and quality per request
For Developers building cost-conscious LLM agent applications
- Role
- agent-app
- Language
- Python
- Licence
- MIT
- Forks
- 901
- Open issues
- 4
- Last push
- 2026-09-08
- Latest release
- v0.2.0 · 2025-11-05
- Skills shipped
- 1
topicsagentmodel-cascadingcost-optimizationllmbudgetspython