LLMap
Models /Gemma 3 12B /Where to run
Provider guide

Where to run Gemma 3 12B

4 live offers tracked.

CHEAPEST
Novita AI
$0.050 / $0.10
per 1M tokens in / out
FASTEST MEASURED
DeepInfra
45 tok/s
measured throughput · $0.050 input / $0.15 output per million tokens
Novita AICHEAPEST$0.050 in/1M$0.10 out/1M—131K—6h agoOpenRouter$0.050 in/1M$0.15 out/1M—131K—14m agoDeepInfra$0.050 in/1M$0.15 out/1M—131Kbfloat166h agoDeepInfrazero-retention$0.050 in/1M$0.15 out/1M45 tok/s131Kbf1613m ago
Full specs, hardware verdicts and benchmarks on theGemma 3 12B model page →