Fast LLM service tier
By Steven Van ·
ElevenAgents can switch supported LLMs to a fast tier for lower latency at a higher per-token price, and the usage calculator now estimates the cost.
ElevenAgents can now run supported LLMs on a fast service tier that trades a per-token price premium for higher token throughput and lower latency. ElevenLabs lists the default as unchanged: agents stay on the standard tier unless configured otherwise.
Turning it on
Set service_tier to fast in the agent's prompt configuration using the Update agent endpoint. Not every model offers the fast tier. The available_service_tiers field returned by List LLMs shows which tiers a given model supports.
Estimating the cost difference
The LLM usage calculator now accepts an optional service_tier parameter, with allowed values standard and fast, defaulting to standard. It only affects models that offer the tier. For other models, the calculator falls back to their standard price.
