As recorded at collection: Cerebras Inference API offers 1 million tokens free per day
As recorded at collection: The Cerebras Inference API, a service for calling AI models from programs, offers 1 million free tokens per day. It can be used without registering a credit card or joining a waitlist.
- What you get
- 1 million free tokens per day
- Price
- Free
- Valid until
- To be confirmed
- Requirements
- No credit card registration required · No waitlist
Details
The free allowance for Cerebras Inference API is 1 million tokens per day. Depending on the model, it processes about 1,000 to 3,000 tokens per second and supports models in the Llama, Qwen, GPT-OSS, and Z.ai GLM families.
You can get started on the service website. Because it is compatible with OpenAI APIs, you can use existing OpenAI SDK (development tool) code by changing the base URL to https://api.cerebras.ai/v1. No separate sign-up process has been specified.
No credit card is required, and there is no waitlist. It has been available to everyone since June 2025, and no expiration date has been specified.
Sources and verification history
- Collection channel
- aicredits
- Collected at
- September 29, 2026, 03:46 KST
- First published
- Updated
- Official evidence
- inference-docs.cerebras.ai/support/rate-limits