AI on Zenifra
Zenifra also offers AI models with simple integration, usage-based billing, and direct visibility in the console.
Offering overview
Zenifra's current AI offering includes the zenifra/qwen3.6-35b-a3b and zenifra/qwen2.5-7b models, with compatibility with OpenAI API clients. The zenifra/qwen3.6-35b-a3b model offers a 262,144-token context window.
| Model | Context | Best for | Average speed | Price |
|---|---|---|---|---|
zenifra/qwen3.6-35b-a3b | 262,144 tokens | long answers, document analysis, and complex tasks | check the console | check the console |
zenifra/qwen2.5-7b | check the console | fast tasks, chat, and internal automations | check the console | check the console |
If you already use the official OpenAI SDK, you only need to point base_url to https://ai.zenifra.com/v1, use the key created in the console, and select the model enabled for that key.
Pricing and billing
AI usage on Zenifra is billed by token. Current prices and available models should always be checked in the Zenifra console, because availability can change over time.
AI usage follows the user's account renewal cycle and is normally charged together with projects and other account charges. If your account has available balance, it is used first. If the balance is not enough, the remaining amount is recorded as negative balance.
Credits and account balance
New accounts receive R$30,00 when created, and that balance can also be used with the platform's AI offering.
Management in the console
In the Zenifra console, you can track:
- daily costs
- total costs
- usage by model
- currently available models
- created API keys
- average model speed when available
Common errors
| Code | Possible cause | How to act |
|---|---|---|
401 | missing, invalid, or revoked key | review ZENIFRA_AI_API_KEY and create a new key if needed |
403 | model not allowed for the key | edit key permissions or use an enabled model |
429 | usage or frequency limit exceeded | apply backoff and reduce concurrency |
| budget exceeded | key reached the configured budget | increase the budget or use another authorized key |
Getting started
To enable access, open AI Models in the Zenifra console. Then:
- click Create API Key
- define a cost budget for the key
- select which models the key can access
- copy the key and use it in your application
The full step-by-step guide is in Getting started with AI.
FAQ
Is the API OpenAI-compatible?
Yes. The current offering is OpenAI-compatible and can be used with clients that allow configuring base_url and api_key.
Where do I see the exact model name?
See the current models and conditions directly in the console.
Does usage consume account balance?
Yes. Available balance is consumed first. If it does not cover all usage, the remainder becomes negative balance.