Cerebras
Inference provider
Pricing page ↗ Cerebras's own price list, link checked 2026-10-19. Check the prices below against it.
- Endpoints
- 7
- Models
- 7
Documented limits
Curated, source linked — read from the linked page on the date in the quote tooltip.
- Rate limits
- Free trial: 5 RPM, 30K uncached TPM, 1M TPD; Developer pay-as-you-go: no hourly/daily restrictions (gpt-oss-120b 1K RPM, 1M uncached TPM) source
- Free tier
- $5 credits expiring 30 days after grant, no permanent renewal source
Models served
7 shown
| Model | Creator | Type | In $/1M | Out $/1M | Context | Max output | Tools | Region |
|---|---|---|---|---|---|---|---|---|
| gemma4:31b gemma-4-31b | Chat | $0.996.43× median | $1.49 | 131K | 41K | Region not stated | ||
| GPT-OSS 120B gpt-oss-120b | OpenAI | Chat | $0.359.46× median | $0.75 | 131K | 41K | Region not stated | |
| Llama 3.1 70B Instruct llama3.1-70b | Meta | Chat | $0.60 | $0.60 | 128K | 128K | Region not stated | |
| Llama 3.1 8B Instruct llama3.1-8b | Meta | Chat | $0.10 | $0.10 | 128K | 128K | Region not stated | |
| Llama 3.3 70B llama-3.3-70b | Meta | Chat | $0.85 | $1.20 | 128K | 128K | Region not stated | |
| Qwen3 32B qwen-3-32b | Alibaba (Qwen) | Chat | $0.40 | $0.80 | 128K | 128K | Region not stated | |
| Qwen3.8-27B qwen-3.8-27b | Alibaba (Qwen) | Chat | $0.99 | $1.49 | 131K | 41K | Region not stated |
One row per model and variant. Dated pins appear on the model page, not as separate rows. Prices are indicative; verify with the provider.