Fast LLM service tier

ElevenAgents now offers a fast service tier for supported LLMs. ElevenLabs says it provides higher token throughput and lower latency at a higher per-token price. Agents stay on standard unless you change the setting.

Use Update agent to set service_tier to fast in the agent's prompt configuration. Check available_service_tiers in List LLMs first, since not every model supports it. The LLM usage calculator also accepts service_tier so you can estimate the cost difference.