Pricing
Free. MIT. Bring your own model.
Linda is an SDK, not a SaaS. You install it. You ship it. You pay only the LLM provider you pick — exactly what their tokens cost.
Linda
$0
forever, MIT licensed
- Full SDK — all packages
- All 5 providers (Anthropic, OpenAI, Groq, OpenRouter, Ollama)
- All 8 built-in skills
- All 19 lifecycle hooks
- MCP client + server
- Optional
@linda/serverproxy - Community support via GitHub
- No telemetry, no CLA, no rug-pull risk
Your LLM provider
Their rates
whatever the model costs per token
- Anthropic: $3/$15 per M tokens (Sonnet 4.6)
- OpenAI: $2.50/$10 per M (GPT-4o)
- Groq: $0.05/$0.08 per M (Llama)
- OpenRouter: aggregated, ~50 models
- Ollama: free (local hardware)
Need help shipping?
We do paid integration work for teams that want Linda wired into their stack with custom skills, custom transports, or compliance-specific guardrails. Get in touch and we'll see if we're a fit.
Frequently asked
Will Linda ever be paid?
The SDK is MIT-licensed in perpetuity. We may build a hosted-evaluation platform later for teams that want managed deployment + a dashboard, but the SDK itself stays free.
Can I use Linda commercially?
Yes. MIT lets you use, modify, and resell as long as you keep the copyright notice. There's no NoCommercial clause, no field-of-use limit.
Is there a CLA?
No. Contributions are accepted under the same MIT license as the rest of the codebase. We won't relicense your contribution out from under you.
No gates. Just install.
The whole thing is npm install away.