MiniMax M2.7 tokens per second

Live benchmark data for MiniMax M2.7 on opencode-go-anthropic. Speed, latency, and reliability — benchmarked every ~10 minutes.

Latest TPS
28.0 tok/s
24h avg TPS
38.6 tok/s
TTFT
1.7s
24h reliability
67%
Last tested
9m ago

Tokens per second over time

Time to first token over time

Speed is measured as output tokens ÷ generation time. Each benchmark fires one streaming chat completion with a fixed ~300-token prompt, every ~10 minutes. Full methodology →