The fastest AionLabs model is Aion-2.0 at 26 output tokens per second. Aion-RP 1.0 (8B) (14 t/s) and Aion-1.0-Mini (1 t/s) round out the top three.
AI models ranked by output speed (tokens per second, p50). The fastest large language models for low-latency and high-throughput applications.
The fastest AionLabs model is Aion-2.0 at 26 output tokens per second. Aion-RP 1.0 (8B) (14 t/s) and Aion-1.0-Mini (1 t/s) round out the top three.
Aion-RP 1.0 (8B) (14 t/s) is the closest alternative on this metric, followed by Aion-1.0-Mini (1 t/s). See the full ranking above for the tradeoffs.
modelgrep tracks 4 AionLabs models with live benchmarks, speed, latency and per-provider pricing. 3 of them qualify for this ranking.