Skip to main content
Cerebras provides free API access to various open-source models optimized for their specialized AI hardware.

​
Overview

Cerebras offers free access to multiple open-source models running on their custom AI accelerators, providing extremely fast inference speeds.

​
Rate Limits

Each model has specific rate limits:
Model NameRequests/MinTokens/MinRequests/HourTokens/HourRequests/DayTokens/Day
gpt-oss-120b3060,0009001,000,00014,4001,000,000
Qwen 3 235B A22B Instruct3060,0009001,000,00014,4001,000,000
Llama 3.3 70B3064,0009001,000,00014,4001,000,000
Qwen 3 32B3064,0009001,000,00014,4001,000,000
Llama 3.1 8B3060,0009001,000,00014,4001,000,000
Z.ai GLM-4.61060,000100100,0001001,000,000

​
Available Models

gpt-oss-120b

120B parameter open-source model

Qwen 3 235B A22B

Qwen’s largest instruction-tuned model

Llama 3.3 70B

Meta’s latest 70B model

Qwen 3 32B

Efficient 32B parameter model

Llama 3.1 8B

Fast 8B parameter model

Z.ai GLM-4.6

GLM-4 generation model

​
API Usage

import openai

client = openai.OpenAI(
    base_url="https://api.cerebras.ai/v1",
    api_key="YOUR_CEREBRAS_API_KEY"
)

response = client.chat.completions.create(
    model="llama3.3-70b",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

​
Getting Started

1

Create Account

Sign up at cloud.cerebras.ai
2

Generate API Key

Create an API key from your dashboard
3

Start Building

Use the OpenAI-compatible API for inference

​
Key Features

  • Ultra-fast inference powered by Cerebras hardware
  • OpenAI-compatible API
  • Generous rate limits on free tier
  • Access to large models (up to 235B parameters)
  • High token throughput

​
Performance

Fast Inference

Specialized hardware for ultra-fast generation

Large Models

Support for models up to 235B parameters

High Throughput

Up to 64,000 tokens per minute

Consistent Speed

Low latency across all model sizes

​
Additional Resources

Cerebras Cloud

Access the platform

Documentation

API documentation

Build docs developers (and LLMs) love

Get started for freeTalk to us