Glossary
Response synthesis
In short
Merging several model answers into one clean result, rather than showing them side by side and making you pick.
When several models answer the same prompt, you can either read all of them and reconcile by hand, or have a final step do it for you. Synthesis is the second option: one answer that folds in the strongest points and drops the weak ones.
Good synthesis leads with the answer, keeps what the models agree on, and resolves or flags where they disagree, instead of averaging everything into mush.
In LLMWeave
The synthesis step in a weave produces one standalone answer, with no model-by-model attribution or repeated points. It reads as a single response, not a digest of several.
Related terms
Model ensemble
Combining the outputs of several models into one result, so the group performs better than any single member.
LLM as a judge
Using one model to evaluate, rank, or merge the outputs of other models.
Consensus
The points that several models agree on, which tend to be more reliable than any single model’s claims.
Try multi-model on your task
One prompt, several models, one answer. Free to start, no card.
Get started