Skip to content

LlamaIndex · Document API

LlamaParse Extract Agentic Plus on ResumeBench: 87.2 F1, rank 1 of 18

LlamaParse Extract Agentic Plus ranks 1st of 18 on ResumeBench: 87.2 F1, $54 per 1,000 pages, 70 seconds per resume. LlamaParse Extract Agentic scores 85.5 for $23, 57% less.

1st of 7 document APIs. Strongest on handwritten (+16.4 points), image only (+12.8 points) and scanned & rotated (+9.8 points).

Rank
1/18
F1
87.2
Precision
88.4
Recall
88.3
Perfect resumes
46/300
Per 1,000 pages
$54
Median per resume
70 s

Averaged over 300 resumes. Dates (scored separately): 98.3. 90% finish within 164 s.

Where LlamaParse Extract Agentic Plus wins and loses

F1 by kind of resume, against the median system.

Reading the page

LlamaParse Extract Agentic Plus by reading the page
Kind of resume# of resumesF1MedianGapRank
Handwritten967.851.5+16.41st
Image only2884.271.4+12.81st
Sidebar or grid8286.881.9+4.81st
Photo4588.184.0+4.11st

Distractors

LlamaParse Extract Agentic Plus by distractors
Kind of resume# of resumesF1MedianGapRank
Rotated & warped pages2192.582.9+9.51st
Hidden text & prompt injections1493.284.5+8.71st
Content as images893.988.2+5.71st
Wording traps1689.084.5+4.52nd
Typos kept on purpose3791.483.7+7.71st

Extraction challenges

LlamaParse Extract Agentic Plus by extraction challenges
Kind of resume# of resumesF1MedianGapRank
Many entries8385.081.9+3.22nd
Long bullet lists10184.981.4+3.51st
Split across pages8585.382.2+3.11st
Big skills list6086.083.6+2.42nd
Custom sections10585.180.8+4.31st
Combined headings3784.682.7+1.92nd
Promotions8887.483.5+3.91st
Partial dates13887.083.4+3.51st
Overlapping jobs6984.181.7+2.34th
Accented text1392.085.5+6.51st

Noise

LlamaParse Extract Agentic Plus by noise
Kind of resume# of resumesF1MedianGapRank
Repeated header6789.385.2+4.11st
References5587.685.2+2.47th
Busy page1882.880.1+2.75th

Source

LlamaParse Extract Agentic Plus by source
Kind of resume# of resumesF1MedianGapRank
Everyday resumes21287.485.1+2.31st
Expert-built (Canva / LaTeX)3286.483.6+2.71st
Resume-builder templates1085.681.2+4.43rd
Scanned & rotated1992.882.9+9.81st
Distractor layouts1889.985.3+4.62nd

Layout

LlamaParse Extract Agentic Plus by layout
Kind of resume# of resumesF1MedianGapRank
Single column19587.084.0+3.01st
Two-column2390.386.8+3.41st
Two-column sidebar5484.880.0+4.81st
Three-column2087.684.8+2.81st
Table grid898.194.1+4.01st

Length

LlamaParse Extract Agentic Plus by length
Kind of resume# of resumesF1MedianGapRank
Short14890.187.1+3.01st
Medium9087.484.7+2.71st
Long6279.774.6+5.11st

By section of the resume

F1 by resume section, against the median system.

LlamaParse Extract Agentic Plus by resume section
SectionResumesF1MedianGap
Contact details30094.393.5+0.8
Experience30091.586.3+5.2
Education29793.389.4+3.9
Summary29084.881.7+3.1
Certifications24085.881.1+4.7
Awards9384.175.4+8.7
Skills28397.692.1+5.5
Languages16784.684.3+0.3
Custom sections21152.249.9+2.3
Interests6493.091.7+1.3
Volunteering7789.684.5+5.0
Profiles3573.366.7+6.6
Projects7887.083.2+3.8
Publications5285.578.6+7.0

On the published resumes

Best and worst of the 20 public resumes.

LlamaParse Extract Agentic Plus on published resumes
ResumeWhat makes it hardF1
Head of Events & Public Space OperationsTable grid with emoji bullets98.4
Senior Data Scientist (Lead)Fifty skills, seven pages97.7
Head of Developer Experience (Internal title: Jedi Master of Analytics) Unusual job titles96.3
Senior Full-Stack EngineerLaTeX CV with a running header77.0
Senior ocularist and practice ownerHandwritten resume, six pages74.0
Senior Game DesignerGame designer's Canva resume64.8

How it was run

LlamaParse Extract API, extraction mode agentic_plus. Same PDF, schema and instructions as every other system, one run per resume (4 November 2026).

LlamaParse Extract (formerly LlamaExtract) is the structured-extraction API of LlamaIndex’s LlamaParse platform.

LlamaParse Extract Agentic Plus on LlamaIndex’s site

Reproduce it with the open-source harness (GitHub):

$ pip install "resume-bench[runners] @ git+https://github.com/careerflow/ResumeBench"
$ hf auth login   # the dataset is gated: accept its terms on Hugging Face first
$ resume-bench download
$ resume-bench run llamaextract_agentic_plus --split test
$ resume-bench grade llamaextract_agentic_plus --split test

Closest alternatives

Nearest in F1.

Embed this result

Shows its current rank. Updates with each run.

LlamaParse Extract Agentic Plus: #1 on ResumeBench resume extraction benchmark
<a href="https://www.resumebench.ai/systems/llamaparse-extract-agentic-plus"><img src="https://www.resumebench.ai/badge/llamaparse-extract-agentic-plus.svg" alt="LlamaParse Extract Agentic Plus: #1 on ResumeBench resume extraction benchmark" height="28"></a>

Choosing a resume extraction model

All 18 models · Leaderboard

Frequently asked questions (FAQs)

How accurate is LlamaParse Extract Agentic Plus at resume extraction?

87.2 F1 on ResumeBench's 300 resumes, 1st of 18 models.

How much does LlamaParse Extract Agentic Plus cost to parse resumes?

$54 per 1,000 pages, about $0.151 per resume.

How fast is LlamaParse Extract Agentic Plus?

70 seconds per resume (median).

Which resumes is LlamaParse Extract Agentic Plus best at?

Handwritten, image only and scanned & rotated. Weakest on none.

What is a cheaper alternative to LlamaParse Extract Agentic Plus?

Reducto Deep Extract: 85.6 F1 for $40 per 1,000 pages.