Models /Qwen2.5 72B Instruct /Where to run

Provider guide

Where to run Qwen2.5 72B Instruct

3 live listings tracked.

CHEAPEST

$0.36 / $0.40

per 1M tokens in / out

FASTEST MEASURED

20 tok/s

measured throughput · $0.36 input / $0.40 output per million tokens

ProviderIn / 1MOut / 1MThroughputContextQuantSynced

OpenRouterCHEAPEST$0.36 in/1M$0.40 out/1M—33K—30 min ago DeepInfrazero-retention through OpenRouterDirect and through OpenRouter$0.36 in/1M$0.40 out/1M20 tok/sthrough OpenRouter33Kfp86 hours ago direct28 min ago through OpenRouter Novita AIzero-retention through OpenRouterDirect and through OpenRouter$0.38 in/1M$0.40 out/1M20 tok/sthrough OpenRouter32Kbf166 hours ago direct28 min ago through OpenRouter

Full specs, hardware verdicts and benchmarks on the Qwen2.5 72B Instruct model page →