The Performance tier provides prioritized capacity for low, consistent latency. It includes a 99.9% availability SLA and a 99% latency guarantee aligned to your enterprise agreement.

Best Use Cases

• User-facing or production-critical paths where latency consistency matters most.
• Workloads that need low p99 time to first token.
• If you want to easily burst beyond your limits, you can pass service_tier=auto in order to use your performance tier limits if they are avaiable, and burst into on_demand. This is a great way to balance perforamcne and costs.

Packaging and pricing

• Performance is delivered as provisioned throughput: you purchase input and output capacity bundles and pay for that provisioned capacity rather than per-token usage. Reach out to inquire about pricing.
• Availability SLA: 99.9% uptime and 99% low latency guarantee. (Details specified in offline agreement)

Model availability

Requirements

• Context length (uncached) must be under 8,192 tokens

Note: If you want automatic context-length handling and tier mapping, use service_tier=auto and we'll route requests to appropriate tiers behind the scenes based on your limits and context size.