Not Diamond
Intelligent router that picks the best model per prompt
Not Diamond is profiled here as a Prompt Engineering tool for engineering teams. Read about features, pricing, and how it compares to related options in the tools directory.
Description
Not Diamond is an AI model router that selects which language model should answer each individual request. It evaluates the incoming prompt and routes it to the model likely to give the best response, balancing quality against cost and latency so applications stop sending every query to one expensive model. Teams can train a custom router on their own evaluation data, and the router returns its choice for the application to call. Routing each prompt to a fitting model lets teams cut spend on simple queries while reserving stronger models for hard ones. Custom routers trained on a team's own evaluations sharpen those decisions for a specific workload.
Key Capabilities:
Per-prompt routing to the best-fit model
Quality, cost, and latency tradeoffs in routing decisions
Custom routers trained on a team's evaluation data
Joint prompt optimization across candidate models
Fuzz-tested fallbacks for reliability
Python and TypeScript SDKs with a REST API
Alternative tools
- CrewAI
Role-based framework for orchestrating multi-agent teams
- Dify
Open-source AI workflow platform with 131,000 GitHub stars and its own orchestration runtime.
- Flowise
Visual drag-and-drop builder for LLM workflows, agents, and RAG pipelines — now part of Workday.
- Langfuse
Open-source LLM engineering platform for tracing, evaluation, and prompt management.
- PromptForge
Prompt engineering workbench for crafting and evaluating LLM prompts.