Use “Together AI RAG relevance brief” in Recued. It provides a read-only Together AI RAG preparation workflow.

How it works 8 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Process (8 steps)

◆ embedding recued-core.togetherai.embedding.create

Create embedding on togetherai

◆ rerank recued-core.togetherai.rerank.create

Create rerank on togetherai

■ defaults defaults

Apply defaults

■ result_count count

Count items in results

■ embedding_count count

Count items in embedding rows

✦ summarize core.ai.summarize

AI generates a concise summary

■ rerank_table to_table

Format results as a data table

■ card to_summary

Format results as a summary card

▼

Output

summary card

table rerank_table

Settings 7 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

focus setting = Summarize reranked relevance, likely best source documents, gaps, ambiguity, and whether the document set is ready for RAG grounding.

query setting = [object Object]

top n setting = 5

documents setting = [object Object]

max length setting = 800

rerank model setting = Salesforce/Llama-Rank-v1

embedding model setting = togethercomputer/m2-bert-80M-8k-retrieval