Launch

Cerebras Code subscriptions launch; rate card narrows

Cerebras pricing

Cerebras introduces fixed-price Cerebras Code coding subscriptions alongside a tightened public inference rate card.

From Cerebras's pricing timeline
Public rate card narrows; Cerebras Code subscriptions launch

By mid-2026 the public per-token rate card lists just two models — GPT-OSS-120B ($0.35/$0.75, production) and ZAI-GLM-4.7 ($2.25/$2.75, labeled a Preview/evaluation model). Llama and Qwen3 families moved to Dedicated Endpoints on reserved-capacity custom pricing. Access is now tiered (Free, a self-serve Developer tier from $10, and Enterprise), and Cerebras introduced fixed-price Cerebras Code coding plans — Pro at $50/month (24M tokens/day) and Max at $200/month (120M tokens/day), both sold out at launch.

About Cerebras
cloud.cerebras.ai ↗

Cerebras operates a per-token inference API (Cerebras Inference) powered by its proprietary Wafer Scale Engine (WSE) chips — the only major LLM inference platform that runs entirely on non-Nvidia hardware at inference scale.

Pricing model pure usagesubscriptioncommitment
Billing units tokensapi callsgpu hours
Sales motion self serveplgsales led
Free tier
Yes
Commits
Available
Transparency
public
Full Cerebras breakdown Plans, meters, and history Estimate your Cerebras cost Model usage and overages

Cerebras pricing history

  1. May 2026
    Public rate card narrows; Cerebras Code subscriptions launch
  2. Jul 2025
    Qwen-3-32B and ZAI-GLM-4.x Models Added
  3. May 2025
    GPT-OSS-120B Added — Fastest Open Reasoning Model
  4. Nov 2024
    IPO Blocked by CFIUS National-Security Review
  5. Aug 2024
    Cerebras Inference Launched in Public Beta — 2,100 Tokens/Second
Full Cerebras timeline
All pricing activity