OpenAI API Key (ChatGPT)

Agencies who are subscribed to Stammer.ai's Agency, Full SaaS Mode or Enterprise plans are able to use their own OpenAI API key for generating the responses.

When an OpenAI API key has been enabled, the agency will pay OpenAI directly for usage instead of Stammer.ai.

When you have your API key enabled, your cost per AI message sent will vary greatly depending on the length of the query, length of the message response, size of the prompt, data in knowledge base and conversation history.

Benefits of using your own API key

1. Added functionality (MasterChat)

2. Cheaper cost per message

3. When using your API key, increased context from Prompt & KB

Why would I want to use my own API key?

Using your own OpenAI API key will reduce your overall usage cost since you are paying OpenAI a variable cost directly instead of paying Stammer.ai a static cost per message (ex: GPT-4o = $0.02 /message).

How to Access Your OpenAI API Key

1. Open an OpenAI account

1. Go to https://platform.openai.com.

2. Click Sign up.

3. Use Google, Microsoft, or any email you like.

4. Check your inbox for the verification email and click the link.

Tip: Use a work email tied to your agency so the billing receipts land in the right place.

---

2. Set up billing (first‑time users only)

1. While logged in, hit Billing in the left menu.

2. Click Add payment method.

3. Enter a credit card that can handle pay‑as‑you‑go charges.

4. Save. Done.

Why bother? Without a card on file your key will hit a hard cap after $5 of usage. That hits faster than you think.

---

3. Make your secret key

1. In the sidebar click API Keys.

2. Press + Create new secret key.

3. Name it something clear like stammer‑agency‑main.

4. Click Create.

5. Copy the long string that pops up. You will not see it again.

Safety check: Paste the key into the Stammer.ai Vault right away or store it in a password vault. Never email it or drop it in Slack.

---

4. Plug the key into Stammer.ai Vault

1. Log in to your Stammer.ai dashboard.

2. Go to Agency → Vault → OpenAI API Key.

3. Paste your OpenAI key.

4. Click Verify & Save.

Run a quick test chat. If the reply comes back, you are set.

---

5. Organization Verification

Why verify? It lets you try OpenAI’s newest models like o1, o3, o3-mini, o4-mini.

1. Head to https://platform.openai.com/settings/organization/general.

2. Click Verify Organization and follow the prompts.

You’ll need:

• A valid government‑issued ID from a supported country.

• One org per ID every 90 days.

• Learn more here

If verification isn’t offered yet, no stress — keep using the API and check back later. After you finish, give it up to 30 minutes for keys to refresh, or just make a new key for instant access.

---

6. Need more juice?

• Higher rate limits: File a ticket inside the OpenAI dashboard if you start getting “rate limit” errors.

• Set spend caps: In Billing → Usage limits you can set hard or soft limits so there are no bill shocks.

---

Support links

• OpenAI docs: https://platform.openai.com/docs

• OpenAI status page: https://status.openai.com

• Stammer.ai support: [e-mail]

That’s it. Your agents now have a direct connection with OpenAI for increased context and functionality.

---

FAQ

How can I estimate my cost per message using my own keys?

Inside the analytics dashboard you can see the average input tokens and output tokens based on current usage across your agents.

You can manually calculate the cost per message by using the pricing from OpenAI or by using Stammer.AI's token calculator.

How are Input/ Output tokens calculated?

Input tokens = base system prompt + user's query + knowledge base search results + past conversation history.

Output tokens = AI agent's generated response.

What is the average number of input/ output tokens?

The average input tokens (data from over 17k AI agents) is between 1,500-3,500 tokens.

For output tokens, the average is between 100-200 tokens.

---

For reference: 100 tokens ≈ 75 words ≈ 450 characters

Will using my own keys change how Stammer works?

Nope. Functionality stays exactly the same.

• You’ll still create, manage, and deploy agents through Stammer.

• You’ll be billed directly by OpenAI for token usage.

• You can still apply a markup (default 10%) on top of usage if you choose to re-bill clients.

• You gain more control over cost, usage, and scaling while being charged less for usage.

What happens to rate limits if I use my own API keys?

When using your own API keys, OpenAI’s default rate limits apply. You can manage this by:

• Verifying your OpenAI organization (unlocks higher limits)

• Setting a spend cap in OpenAI to prevent runaway costs

• Using Stammer’s built-in rate limit controls for each agent

What are the benefits of using my own API keys?

• Cheaper messages costs

• Unlimited size of base system prompt

• Added context from knowledge base search results

• Increased profits through re-billing

• More control over your costs