Ask one question; several models answer it independently, in parallel — you choose the model for each slot, any vendor you have a key for. Each has live web access. A synthesizer then compares their answers into agreements, disagreements, and a bottom line.
Independent answer slot — the model you assign answers the question directly, with live web access.
Independent answer slot — the model you assign answers the question directly, with live web access.
Independent answer slot — the model you assign answers the question directly, with live web access.
Synthesizer — compares the model answers into agreements, disagreements, and a bottom line.