Cerebras review

Cerebras is the speed specialist. Its docs list about 3,000 tokens per second for gpt-oss-120B and about 1,850 for Qwen 3.8 27B, which is the highest documented output speed in this list. The self-serve Developer tier is pay as you go, with gpt-oss-120B at $0.35 per 1M input tokens and $0.75 per 1M output tokens, and a one-time $5 promotional credit when you add a payment method. Rate limits are generous on the paid tier at 1K requests and 1M input tokens per minute, while the free trial is capped at 5 requests per minute and 1M tokens a day. The catch is scope: only two models are available self-serve, and Cerebras describes the Developer tier as meant for development and evaluation, with production capacity, priority and custom weights on the quote-based Enterprise tier. Per token it costs more than GPU-based rivals for the same model.

Who it is for

Apps that need the highest tokens per second on gpt-oss-120B

Price

Self-serve pay as you go with a one-time $5 promotional credit; gpt-oss-120B at $0.35 per 1M input tokens and $0.75 per 1M output tokens.

Pricing: $0.35 per million input tokens, $0.75 per million output tokens.
Quoted from the vendor pricing page on November 12, 2026.

Key features

• About 3,000 tokens per second on gpt-oss-120B and about 1,850 on Qwen 3.8 27B, per Cerebras docs
• OpenAI-compatible chat completions plus a Cerebras Python and Node.js SDK
• Batch, predicted outputs and service tiers documented, some as preview features
• Enterprise tier with production capacity, higher rate limits and fine-tuning by agreement
• Also available through AWS Marketplace, OpenRouter, Hugging Face and Vercel

Strengths and weaknesses

• Strength: Fastest documented output speed in this list on gpt-oss-120B
• Strength: One-time $5 promotional credit when you add a payment method, valid 30 days
• Strength: No automatic enrollment in paid usage when the credit runs out
• Weakness: Only two models on the self-serve Developer tier: GPT OSS 120B and Qwen 3.8 27B
• Weakness: Cerebras says the Developer tier is not intended for production use
• Weakness: gpt-oss-120B costs more than on Together AI, Fireworks AI, Groq or Baseten

Key facts

• gpt-oss-120B: $0.35 input, $0.75 output per 1M tokens, about 3,000 tokens per second
• Free trial limits: 5 requests per minute, 1M tokens per day
• Developer tier limits: 1K requests per minute, 1M input tokens per minute
• Enterprise: Contact Cerebras for a quote

Where it ranks

Cerebras is No. 6 in our Best AI Inference Platforms verdict.