---
type: "post"
id: "introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5"
slug: "introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5"
title: "Introducing Exa Agent Ultra – Lists without Limits"
description: "Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search."
retrieval_nugget: "Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search."
published_at: "2026-09-24"
updated_at: "2026-09-24"
record_date: "2026-09-24"
date_kind: "published_at"
topics: []
source_urls: ["https://exa.ai/blog/exa-agent-ultra"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:46c3a08463ea83cfdba32704fe08824a6ba6946fa77aff85fd4647377e977ab0","reviewed_at":"2026-09-24T18:00:45.811Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":2,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search.","observed_facts":[{"text":"Exa announced Agent Ultra as the highest-effort level for Exa Agent, aimed at deep research, list-building, and entity enrichment.","source_urls":["https://exa.ai/blog/exa-agent-ultra"]},{"text":"Exa reports 81.4% soft recall on WANDR at a $18.53 median cost per task, 58.9% on WideSearch at $3.85 per task, and 93.9% on DeepSearchQA.","source_urls":["https://exa.ai/blog/exa-agent-ultra"]},{"text":"Exa says the agent divides tasks into subtasks, assigns subagents across domains, and mixes frontier and cost-effective models.","source_urls":["https://exa.ai/blog/exa-agent-ultra"]}],"mechanism":"The mechanism is task decomposition into subagents researching multiple domains in parallel, combined with a mix of frontier and cost-effective models chosen for the task.","why_now":"This matters now because Exa is attaching explicit benchmark and per-task-cost claims to that orchestration layer: it reports 81.4% soft recall on WANDR, 58.9% on WideSearch, and 93.9% on DeepSearchQA.","implications":["The immediate implication is that research-agent competition is shifting toward auditable, row-level coverage for diligence, monitoring, and market mapping, not only answer quality.","Teams evaluating agents may need to measure evidence quality, deduplication, and review burden alongside answer-level benchmark scores."],"evidence_boundary":"The evidence boundary is that these performance and cost comparisons come from Exa’s launch post; the company says it used published competitor results where available and ran some comparisons itself, while its Find-All Company benchmark is internal.","watch_conditions":["Watch whether independent evaluations reproduce the reported recall-and-cost advantages on real customer workloads with stated completeness criteria.","Watch whether cited output, exclusion of previously found rows, and error rates hold up in production workflows."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this launch starts a distinct record for high-effort, exhaustive research-agent orchestration and its benchmark claims."}
routes: {"html":"https://newruntime.com/posts/introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5/","markdown":"https://newruntime.com/posts/introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5.md","json":"https://newruntime.com/posts/introducing-exa-agent-ultra-lists-without-limits-analysis-ad81c5.json"}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
---

# Introducing Exa Agent Ultra – Lists without Limits

## Retrieval answer

Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search.

Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search. The product is a new highest-effort mode for Exa Agent, and Exa says it is intended for deep research, list-building, and entity enrichment where completeness outweighs latency or cost.

Exa Agent Ultra frames exhaustive list construction as a distinct agent workload, rather than a variant of conversational search. The mechanism is task decomposition into subagents researching multiple domains in parallel, combined with a mix of frontier and cost-effective models chosen for the task. This matters now because Exa is attaching explicit benchmark and per-task-cost claims to that orchestration layer: it reports 81.4% soft recall on WANDR, 58.9% on WideSearch, and 93.9% on DeepSearchQA.

The immediate implication is that research-agent competition is shifting toward auditable, row-level coverage for diligence, monitoring, and market mapping, not only answer quality. The evidence boundary is that these performance and cost comparisons come from Exa’s launch post; the company says it used published competitor results where available and ran some comparisons itself, while its Find-All Company benchmark is internal. Watch whether independent evaluations reproduce the reported recall-and-cost advantages on real customer workloads with stated completeness criteria. The product’s practical test will be whether teams can obtain cited, deduplicated lists that remain cheaper to review than manually assembled research.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)