Model scorecard
Google gemini-3.1-pro
Composite 77.8/100 across quality, throughput, latency, price efficiency and usable context — measured through the gateway, not vendor-reported.
A77.8/100google/gemini-3.1-protext · vision · audio · video · 2000k context
Capability
eval compositeQuality (blended eval)92
Category: long
Usable context100
2000k advertised window
Serving
gateway measuredThroughput34
118 output tokens/s median
Time to first token70
180ms p50
Economics
at provider costPrice efficiency57
$7.00 per 1M output tokens
Quality per dollar39
llmcloud adds $0 to these token prices.
Suggested usage
- Long-context retrievalauto:long-context
Effective recall at depth, not the advertised window, is the deciding number.
- Vision & document parsingauto:vision
Chart and table extraction quality separates models far more than natural-image captioning.