GPU Benchmark Comparison

— vs — — Pre-fill & Token Generation Performance

Model: —

Concurrency:

Pre-Fill Performance (tokens/sec)

Token Generation Performance (tokens/sec)

Detailed Comparison Table

prompt processing
tokens/sec
prompt processing energy efficiency
tokens/sec per watt
token generation
tokens/sec
token generation efficiency
tokens/sec per watt
Ctx

Efficiency and Price

Power Draw

Ratio

System Cost

Ratio

Avg Cost per 1M Tokens @ 0.10 $/kWh

Customize

Electricity Cost ($/kWh)