Provider guide
Where to run Gemini 2.5 Pro Preview 05-06
9 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.
FASTEST MEASURED
104 tok/s
measured throughput · $1.25 input / $10.00 output per million tokens
Google Vertex AICHEAPEST$0.63 in/1M$5.00 out/1M—1M—32h agoGoogle Vertex AI$1.25 in/1M$10.00 out/1M104 tok/s1M—14m agoGoogle AI Studio$1.25 in/1M$10.00 out/1M81 tok/s1M—14m agoOpenRouter$1.25 in/1M$10.00 out/1M—1M—15m agoGoogle AI$1.25 in/1M$10.00 out/1M—1M—6h agoGoogle Vertex AI$1.25 in/1M$10.00 out/1M99 tok/s1M—14m agoGoogle Vertex AIzero-retention$0.63 in/1M$5.00 out/1M85 tok/s1M—14m agoGoogle AI Studioflex tier$0.63 in/1M$5.00 out/1M29 tok/s1M—14m agoGoogle AI Studiopriority tier$2.25 in/1M$18.00 out/1M—1M—14m ago
Google AI Studio
OpenRouterFull specs, hardware verdicts and benchmarks on theGemini 2.5 Pro Preview 05-06 model page →