Models /Llama 4 Scout /Where to run
Provider guide

Where to run Llama 4 Scout

7 live offers tracked — output prices vary 2.0× between the cheapest and most expensive host, so the provider choice matters as much as the model choice.

CHEAPEST
$0.10 / $0.30
per 1M tokens in / out
FASTEST MEASURED
134 tok/s
measured throughput · $0.11 input / $0.34 output per million tokens
DeepInfra$0.10 in/1M$0.30 out/1M328Kfp810h agoOpenRouterCHEAPEST$0.10 in/1M$0.30 out/1M1.3M4h agoNovita AI$0.18 in/1M$0.59 out/1M131K10h agoDeepInfrazero-retention$0.10 in/1M$0.30 out/1M36 tok/s328Kfp84h agoGroqzero-retention$0.11 in/1M$0.34 out/1M134 tok/s131K4h agoNovita AIzero-retention$0.18 in/1M$0.59 out/1M8 tok/s131Kbf164h agoGoogle Vertex AIzero-retention$0.25 in/1M$0.70 out/1M53 tok/s1.3M4h ago
Full specs, hardware verdicts and benchmarks on theLlama 4 Scout model page →