Cerebras Pricing Calculator
Updated May 2026Usage-based per-token inference API (free, $10 self-serve Developer, and Enterprise tiers) plus fixed-price Cerebras Code coding subscriptions; hardware systems on custom enterprise contracts
Cerebras pricing: Cerebras pricing 2026: per-token inference (GPT-OSS-120B $0.35/$0.75, GLM-4.7 $2.25/$2.75), free + $10 Developer tiers, $50/$200 Cerebras Code plans. Use the free calculator below to enter your usage and get an instant all-in monthly estimate — including overages — and see which plan is cheapest for your needs.
How Cerebras prices: Usage-based per-token inference API (free, $10 self-serve Developer, and Enterprise tiers) plus fixed-price Cerebras Code coding subscriptions; hardware systems on custom enterprise contracts.
How Cerebras prices
Usage-based per-token inference API (free, $10 self-serve Developer, and Enterprise tiers) plus fixed-price Cerebras Code coding subscriptions; hardware systems on custom enterprise contracts
Best fit — auto-selected: Enterprise / CS-3
$0.00/mo— the cheapest plan that fits; adjust usage and it re-picks
Your usage
💡 Prompt + context tokens sent to the model. 300M ≈ a small production chatbot (~10M input tokens/day).
💡 Generated response tokens — the pricier side. Output is typically 20-40% of input volume.
Pick the same model in both rate selectors. ZAI-GLM-4.7 is a Preview model for evaluation only.
Output tokens cost more than input on both public rate-card models.
Compare Plans at Your Usage
Monthly Estimate
Enterprise / CS-3
$0.00
per month
Cost Breakdown
Annual
$0.00
Why Enterprise / CS-3 — $0.00/mo
- Enterprise / CS-3 at $0.00/mo.
- How we got there: 300 M tokens (Input tokens / mo) × 0 $/M tokens (Model — input rate) = 105 $.
- How we got there: 90 M tokens (Output tokens / mo) × 1 $/M tokens (Model — output rate) = 68 $.
Budget range
Plan for usage swings, not just today's estimate.
Conservative
$0.00
45 M tokens/mo
Expected
$0.00
90 M tokens/mo
Aggressive
$0.00
180 M tokens/mo
Need a calculator like this on your pricing page?
Embed interactive pricing calculators on your website to help customers understand costs and boost conversions.
Get StartedCerebras plans at a glance
| Plan | Price | Best for |
|---|---|---|
| Free | Free | Developers getting started — rate-limited, Discord support |
| Developer (pay-per-token) | Usage-based | Power users and production — self-serve, 10x higher rate limits |
| Cerebras Code Pro | $50/mo | Indie devs, agentic coding — fixed monthly (sold out) |
| Cerebras Code Max | $200/mo | Full-time dev, IDE + multi-agent — fixed monthly (sold out) |
| Enterprise / CS-3 | Usage-based | Custom weights, guaranteed uptime, hardware — contact sales |
Prices shown are list rates as of May 2026. Enter your usage above for an all-in estimate including overages and add-ons.
About this Cerebras calculator
This calculator estimates your Cerebras cost from publicly available pricing. Actual costs may vary with your specific agreement, volume discounts, and usage patterns — always verify on the provider's official pricing page for the most current rates.
Cerebras pricing — frequently asked questions
How much does Cerebras Inference cost per million tokens? ▼
The public per-token rate card lists two models: GPT-OSS-120B at $0.35 input/$0.75 output per million tokens (production) and ZAI-GLM-4.7 at $2.25 input/$2.75 output per million tokens (a Preview model, intended for evaluation only, not production). Other models such as Llama 3.3 70B and Qwen3-32B are available via Dedicated Endpoints on custom pricing rather than the public rate card.
Does Cerebras offer a free tier for the inference API? ▼
Yes. Cerebras provides a free tier with rate-limited access to all public models and Discord support. To raise limits, the self-serve Developer tier starts at just $10 and offers 10x higher rate limits and higher-priority processing; the Enterprise tier (contact sales) adds custom weights, dedicated queue priority, and guaranteed uptime.
How fast is Cerebras inference compared to GPU-based providers? ▼
Cerebras Inference delivers 1,000–2,100 tokens per second on Llama 3.1 70B-class models, compared to 40–80 tokens/second on GPU-based providers like Together AI or Fireworks AI. The speed advantage comes from the on-chip SRAM of the WSE eliminating GPU memory bandwidth bottlenecks.
What models are available on Cerebras Inference? ▼
As of mid-2026, the public per-token rate card lists GPT-OSS-120B (production) and ZAI-GLM-4.7 (a Preview/evaluation model). A much wider catalog — including Llama 3.3 70B, Llama 4 Maverick and Scout, Qwen3-32B and Qwen3-235B, Qwen3-Coder, Mistral, DeepSeek, and Kimi K2.x — is available through Dedicated Endpoints on reserved-capacity custom pricing. The catalog focuses on open-source models that benefit most from Cerebras's speed advantage.
Is this Cerebras pricing calculator free and accurate? ▼
Yes — it's completely free with no signup. It uses the latest publicly available Cerebras pricing (verified May 2026). Actual costs may vary with volume discounts or enterprise terms — always confirm on Cerebras's official pricing page.