Build and connect
Bring your own model keys
Use Alloy with your own LLM provider keys, or let Alloy pass AI and compute usage through when that is easier for your team.
Choose how model usage is paid
Alloy pricing separates the workspace and AI teammate plan from model usage. Teams can bring their own LLM provider, or pay for AI and compute usage through Alloy with transparent pass-through pricing.
This gives teams a simple choice: use the provider setup they already have, or let Alloy handle usage billing.
Why teams bring their own keys
Bring-your-own-model-keys is useful when a company already has a provider account, procurement process, spending controls, internal approval path, or model preference.
It can also help teams keep model usage aligned with their existing vendor contracts while still using Alloy for AI teammates, workflows, storage, tools, and collaboration.
Model and voice capabilities still matter
Not every model can do every job. Alloy tracks model capabilities such as multimodal support, realtime support, text-to-speech, and speech-to-text.
Voice setup also depends on provider-compatible voices. If a teammate uses a TTS model, a compatible voice is required.
Works across plans
The Free plan includes one AI teammate, intro credits, and the option to bring your own LLM provider. Pro supports more AI teammates and also lets teams bring their own provider or pay through Alloy.
Enterprise teams can discuss custom pricing, limits, security needs, support, and procurement requirements.
Frequently asked questions
Do we have to bring our own model keys?+
No. Teams can bring their own provider or pay for AI and compute usage through Alloy.
Which plans support bringing our own provider?+
The current pricing copy includes bring-your-own-provider support in Free and Pro, with Enterprise available for custom needs.
Does bringing our own provider change model capabilities?+
No. The selected model still needs the capabilities required for the job, such as chat, realtime voice, TTS, STT, or multimodal support.