TimeTonic lets you use your own API key in your AI automations, as an alternative to the credits included in your plan. This option is not mandatory · it addresses specific needs that this guide will help you identify.
How to activate it ?
When should you use your own API key ?
What is the difference from TimeTonic AI credits ?
How does billing work ?
Impact on your TimeTonic credits
Best practices
Frequently asked questions
How to activate your own API key ?
The API key is activated when configuring an Ask an AI action in your automation scenario.
In your action, select the Public Cloud AI source.
Enable the "Use personal API key" toggle.
Select your provider:
ChatGPT: gives access to OpenAI models: GPT-5, GPT-4.1, GPT-4o, o3, o4, o1, and their variants.
The API key is specific to your configuration. It can be modified or removed at any time without affecting your other automations.
Switching between personal API key and TimeTonic credits
Once your API key is configured and your model selected, you can switch between the two modes at any time without losing your configuration.
How does it work ?
The Use a personal API key toggle acts as a switch in the configuration of your Ask an AI action:
Enabled 🔵
The action uses your personal API key and the selected provider (ChatGPT or Mistral AI). Your key is saved and the chosen model is retained.
Disabled ⚪
The action automatically falls back to TimeTonic AI credits (public cloud). TimeTonic selects the available model closest to the one you had configured.
💡 Good to know: disabling the toggle does not delete your API key - it remains saved. You can re-enable the option at any time and find your configuration intact.
⚠️ Security - re-entering your API key: if you click the pencil button to edit the API key field, the saved key is automatically cleared for security reasons. You will need to re-enter it in full before saving.
This toggle lets you quickly test both modes on the same action, or temporarily fall back to TimeTonic credits if your provider quota is reached.
Models available with your own API key (BYOK)
Models available via OpenAI API key
🔑 Models available via OpenAI API key (BYOK)0 TimeTonic credits · Billed directly by OpenAI
ModelContextInput /M tokensOutput /M tokensAttachmentsIdeal for
📎 Supports attachments (images, documents) · Pricing in USD per million
tokens · Prices are subject to change.
For up-to-date pricing and model details →
openai.com/api/pricing
The o1, o3, and o4 models: what is reasoning?
These models work differently from standard GPT models. Before responding,
they perform an internal thinking phase · they break the problem down into
steps, verify their logic, and self-correct. This process produces what
OpenAI calls reasoning tokens, which are billed in addition
to the standard input and output tokens.
They are relevant for complex tasks requiring multi-step logic: consistency
checks, in-depth legal or financial analysis, anomaly detection in structured
data. For simple, repetitive tasks (extraction, classification, reformulation),
a standard GPT model will deliver equivalent results at a significantly lower
cost.
256K$0.20$0.20📎Image analysis, structured extraction, multilingual, strong cost-to-performance ratio
📎 Supports attachments (images, documents) · Pricing in USD per million
tokens · Prices are subject to change.
For up-to-date pricing and model details:
mistral.ai/pricing
When should you use your own API key?
This option exists to address specific situations. It is not necessary for the majority of users · the credits included in your plan are sufficient for standard use.
You may consider using your own API key if:
You already have an active subscription with an AI provider
OpenAI or Mistral. If you already have an account and available credits
with one of these providers, you can use them directly in TimeTonic.
You want to centralize AI billing within your organization
Some organizations prefer to consolidate all their AI usage under a single
provider account for unified budget management.
You exceed the credits in your plan without wanting to upgrade
If your monthly quota is reached mid-period, you can fall back on one
of your active provider subscriptions to keep your automations running
without interruption.
You have internal budget governance requirements
Some organizations require AI spend to be tracked and approved separately,
with direct visibility at the provider level.
What is the difference from TimeTonic AI credits?
Both modes coexist in TimeTonic. Here is what concretely distinguishes them:
TimeTonic credits
Personal API key
Billing
TimeTonic
Directly with the AI provider
Consumption tracking
Visible in your TimeTonic administration space
To be monitored directly in your provider dashboard
Usage limit
Based on your TimeTonic plan
Based on your provider account
Usage alerts
Automatic email at
80%
and
100%
of quota
To be configured directly with the provider
Automatic suspension
At 100% of monthly quota
No TimeTonic suspension
Credit management
Two separate counters: Private AI and Public Cloud AI
Independent, managed directly with the provider
How does billing work?
When you use your own API key, AI calls are processed and billed directly by the provider (OpenAI, Mistral, Anthropic...). TimeTonic does not consume any credits and is not involved in billing.
The only limits that apply are those of your provider account: monthly quota, rate limits, etc.
If your provider account reaches its limit or your API key is invalid, the automation will fail. Make sure to monitor your consumption directly in the provider's interface.
Impact on your TimeTonic AI credits
When an automation uses your personal API key, no TimeTonic credits are consumed · your plan counter does not move.
Best practices
Monitor your consumption directly with the provider · TimeTonic will not alert you in case of overage on the provider side.
Set spending limits in your provider's interface to avoid unexpected charges.
Test with efficient models before deploying at scale · costs can vary significantly from one model to another.
Check your provider's GDPR compliance · certain processing may require hosting in France or Europe. In that case, prefer TimeTonic's Private & Secure AI.
Frequently asked questions
Can I mix TimeTonic credits and a personal API key ?
Yes. The choice is made action by action within your automation scenarios. Some actions can use TimeTonic credits, others your API key · both modes coexist without constraint.
Can I use a different API key for different automations ?
Yes. Each action can be configured independently with the key of your choice.