Tokenade
Local proxy that compacts what a coding agent sends to the model
Tokenade is a CLI that sits between a coding agent and the model provider, rewriting requests to cut token usage before they are sent. It ships command-specific compactors and an MCP proxy wrapper, so it works with agents that speak MCP without changes to the agent itself.
It runs locally rather than as a hosted service, so prompts do not leave the machine on their way through it.
Billing is by tokens saved: free up to 10M per month, then a paid tier with per-million overage. Distributed on npm as @tokenade/cli. Early-stage and closed source despite the public repository, which holds the distribution wrapper rather than the implementation.
Pricing: Monthly subscriptions
Tokenade Alternatives
Explore 33 products in the Frameworks & Stacks category. View all Tokenade alternatives.
Mastra
TypeScript-first AI framework for building agents, RAG pipelines, and workflows
vLLM
High-throughput LLM inference engine with PagedAttention for efficient GPU memory usage
Ollama
Run large language models locally with a single command
Work on Tokenade? Feature it at the top of Frameworks & Stacks.
Is your product missing?