Llama 3.3 70B
Llama 3.3 70B served on Cerebras hardware, aimed at latency-sensitive workloads.
Free tierLow latencyOpen model
Context window131K
Max output33K
Free tierFree quota available
China accessProxy required
Rate limits
| Requests per minute (RPM) | 30 |
|---|---|
| Requests per day (RPD) | 900 |
| Tokens per minute (TPM) | 60,000 |
| Tokens per day (TPD) | — |
Capabilities
Tool callingVision inputJSON modeStreaming
Access requirements
Credit card: NoReal-name verification: NoPhone number: NoChina access: Proxy required
Source
- URL
- View official docs · https://inference-docs.cerebras.ai/support/rate-limits
- Retrieved
- 2026-10-19
- Confidence
- medium
- Verified
- 2026-10-19
Information comes from public provider documentation and may change at any time. Check the official page before integrating. This site does not proxy APIs.