AI Provider Setup · BYOK

Connect OpenAI

GPT-5, GPT-4o, and o4-mini direct from OpenAI.

OpenAI is the vendor behind the GPT family and what most teams have an existing account with. Good pick when you've already budgeted on OpenAI, need GPT-5's very large 400K context, or want tight integration with the OpenAI tool ecosystem.

Best used for

Key details

Where to generate your key
https://platform.openai.com/api-keys
Expected key format
Starts with sk- or sk-proj-
Environment variable name (if self-hosting)
OPENAI_API_KEY
Provider pricing
https://openai.com/api/pricing
Official docs
OpenAI API docs

Step-by-step setup

  1. 1 Create an OpenAI platform account

    Sign up at platform.openai.com. The platform account is separate from your ChatGPT account — you can reuse the same email.

  2. 2 Add billing

    Open Billing and add a payment method with prepaid credit. OpenAI will not accept API calls without funded credit — this is a common first-time gotcha.

  3. 3 Open the API Keys page

    Navigate to platform.openai.com/api-keys. If your org uses projects, select the right project first.

  4. 4 Create a new secret key

    Click Create new secret key, name it Admaxxer, and set permissions (All is fine; Restricted works if you want to scope it). Project keys start with sk-proj-, user keys with sk-.

  5. 5 Copy the key immediately

    OpenAI shows the full secret exactly once. Paste it into Admaxxer straight away.

  6. 6 Paste into Admaxxer

    Open Settings then AI Providers, find the OpenAI row, paste the key, and click Connect.

  7. 7 Pick a default model

    gpt-5 is our recommended pick — 400K context and best reasoning. Use gpt-4o for a balanced workhorse, o4-mini for cheap reasoning.

  8. 8 If the connection test fails

    Confirm the key starts with sk- or sk-proj-, billing is funded (not just a card on file — you need credit too), and the project has model access.

Test the connection inside Admaxxer

Once you've pasted your key into /settings/ai-providers, hit Test. Admaxxer makes a single no-cost call against the provider's /v1/models (or equivalent) endpoint and reports back:

Available models

These are the pinned model IDs Admaxxer tests against. The model picker in chat will show these plus any live-catalog entries the provider returns.

Model ID Display name Context Tools Best for
gpt-5 GPT-5 (recommended) 400K Yes Flagship — 400K context and best GPT reasoning. Default pick.
gpt-4o GPT-4o 128K Yes Balanced workhorse — cheaper than GPT-5, strong on tools and multimodal.
o4-mini o4-mini 200K Yes Cheap reasoning-style model. Good for chain-of-thought on a budget.

Common errors and fixes

401

When it happens: On Connect if the key is wrong or was revoked.

Fix: Create a new key at platform.openai.com/api-keys. Keys cannot be recovered — always rotate.

429

When it happens: Under sustained load, especially on lower-tier accounts.

Fix: OpenAI has tiered per-minute limits. Request a tier bump at platform.openai.com/account/limits.

402 insufficient_quota

When it happens: When your prepaid credit drops to zero.

Fix: Top up at platform.openai.com/account/billing. OpenAI does not auto-recharge unless you enabled it.

model_not_found

When it happens: If the project does not have the selected model enabled.

Fix: Go to platform.openai.com/account/limits and request model access for the project.

invalid_organization

When it happens: If your key was created in a different org than expected.

Fix: Create the key while viewing the correct org in the top-left org switcher.