Cerebras API Pricing
🌍 GlobalPayment Methods
Cerebras is an official AI API provider. ComputeUnion tracks 6 price records for it (6 auto-scraped from the official pricing page, Updated 5m ago). Browse the categories below to compare Cerebras official rates against relay providers.
Model Pricing on Cerebras
Prices per 1M tokens
LLM6 models
| Model | Input | Output | Context | Updated |
|---|---|---|---|---|
| Llama3.1 8B llama3.1-8b | $0.1000 | $0.1000 | 125K | Updated 5m ago |
| Gpt Oss 120B gpt-oss-120b | $0.3500 | $0.7500 | 125K | Updated 5m ago |
| Qwen 3 32B qwen-3-32b | $0.4000 | $0.8000 | 32K | Updated 5m ago |
| Llama 3.3 70B llama-3.3-70b | $0.5900 | $0.9900 | 125K | Updated 5m ago |
| Qwen 3 235B A22B Instruct 2507 qwen-3-235b-a22b-instruct-2507 | $0.6000 | $1.2000 | 125K | Updated 5m ago |
| Zai Glm 4.7 zai-glm-4.7 | $2.2500 | $2.7500 | 125K | Updated 5m ago |
FAQ
How does Cerebras API pricing work?
Cerebras is an official AI API provider with usage-based billing. ComputeUnion currently tracks 6 price records — 6 auto-scraped from the official pricing page (Updated 5m ago), 0 manually maintained. The table below lists current per-model rates.
How does Cerebras official pricing compare to relay platforms?
Relay platforms typically resell access at a discount, while the official API offers stronger latency, SLA, and feature guarantees. Use the category links below to compare Cerebras official rates against relay providers on ComputeUnion.
Is Cerebras accessible from China?
Cerebras is an international service — direct access from China mainland may be restricted.
Related pages