---
type: "raw_signal"
id: "introducing-extract-v2-5-e730c9"
slug: "introducing-extract-v2-5-e730c9"
title: "LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents"
description: "LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents."
retrieval_nugget: "LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents."
observed_at: "2026-10-01"
record_date: "2026-10-01"
date_kind: "observed_at"
why_it_matters: "LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents. The company reports higher ExtractBench value F1 scores across all three tiers: Cost Effective increased from 87.1 to 93.9, Agentic from 89.8 to 95.8, and Agentic Plus from 95.1 to 96.4. It says per-page pricing is unchanged. The release also extends improved grounding features to Agentic, including boundi"
novelty: "notable"
verification_level: "source-linked"
signal_type: "release"
source_platform: "llamaindex.ai"
topics: []
entities: []
related_patterns: []
source_url: "https://www.llamaindex.ai/blog/introducing-extract-v2-5"
source_urls: ["https://www.llamaindex.ai/blog/introducing-extract-v2-5"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:introducing-extract-v2-5-e730c9"
status: "published"
visuals: []
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"]}
routes: {"html":"https://newruntime.com/signals/introducing-extract-v2-5-e730c9/","markdown":"https://newruntime.com/signals/introducing-extract-v2-5-e730c9.md","json":"https://newruntime.com/signals/introducing-extract-v2-5-e730c9.json"}
---

# LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents

## Retrieval answer

LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents.

LlamaIndex announced Extract v2.5, an update to its schema-based document extraction agents. The company reports higher ExtractBench value F1 scores across all three tiers: Cost Effective increased from 87.1 to 93.9, Agentic from 89.8 to 95.8, and Agentic Plus from 95.1 to 96.4. It says per-page pricing is unchanged. The release also extends improved grounding features to Agentic, including bounding-box citations for supporting evidence, and describes gains on long lists, cross-page records, and scanned forms. These figures and examples are vendor-reported; the announcement does not provide an independent evaluation.

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: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- 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)