Models /Kimi K2 Thinking /Where to run

Provider guide

Where to run Kimi K2 Thinking

3 live listings tracked.

CHEAPEST

Google Vertex AI

$0.60 / $2.50

per 1M tokens in / out

FASTEST MEASURED

Google Vertex AI

106 tok/s

measured throughput · $0.60 input / $2.50 output per million tokens

ProviderIn / 1MOut / 1MThroughputContextQuantSynced

OpenRouter$0.60 in/1M$2.50 out/1M—262K—2 hours ago Novita AIzero-retention through OpenRouterDirect and through OpenRouter$0.60 in/1M$2.50 out/1M47 tok/sthrough OpenRouter262Kbf162 hours ago Google Vertex AIzero-retentionCHEAPEST$0.60 in/1M$2.50 out/1M106 tok/s262K—2 hours ago

Full specs, hardware verdicts and benchmarks on the Kimi K2 Thinking model page →