Skip to content

LlamaIndex · Document API

LlamaParse Extract Cost-Effective on ResumeBench: 83.5 F1, rank 10 of 18

LlamaParse Extract Cost-Effective ranks 10th of 18 on ResumeBench: 83.5 F1, $8.18 per 1,000 pages, 29 seconds per resume. It is 3.6 points behind LlamaParse Extract Agentic Plus. Qwen 3.8 Flash scores 84.1 for $2.98.

5th of 7 document APIs. Strongest on scanned & rotated (+7.4 points), rotated & warped pages (+6.7 points) and hidden text & prompt injections (+6.0 points). Weakest on busy page (−3.2 points), references (−2.3 points) and many entries (−2.2 points).

Rank
10/18
F1
83.5
Precision
85.1
Recall
85.3
Perfect resumes
26/300
Per 1,000 pages
$8.18
Median per resume
29 s

Averaged over 300 resumes. Dates (scored separately): 97.9. 90% finish within 42 s.

Where LlamaParse Extract Cost-Effective wins and loses

F1 by kind of resume, against the median system.

Reading the page

LlamaParse Extract Cost-Effective by reading the page
Kind of resume# of resumesF1MedianGapRank
Handwritten949.651.5−1.915th
Image only2876.371.4+4.96th
Sidebar or grid8284.381.9+2.36th
Photo4583.184.0−0.911th

Distractors

LlamaParse Extract Cost-Effective by distractors
Kind of resume# of resumesF1MedianGapRank
Rotated & warped pages2189.682.9+6.74th
Hidden text & prompt injections1490.584.5+6.04th
Content as images888.288.2−0.010th
Wording traps1685.084.5+0.58th
Typos kept on purpose3788.083.7+4.34th

Extraction challenges

LlamaParse Extract Cost-Effective by extraction challenges
Kind of resume# of resumesF1MedianGapRank
Many entries8379.681.9−2.212th
Long bullet lists10180.481.4−1.111th
Split across pages8581.882.2−0.411th
Big skills list6084.183.6+0.58th
Custom sections10580.480.8−0.411th
Combined headings3782.882.7+0.18th
Promotions8882.983.5−0.611th
Partial dates13882.583.4−1.011th
Overlapping jobs6979.781.7−2.013th
Accented text1387.085.5+1.55th

Noise

LlamaParse Extract Cost-Effective by noise
Kind of resume# of resumesF1MedianGapRank
Repeated header6784.385.2−0.811th
References5582.885.2−2.313th
Busy page1876.980.1−3.212th

Source

LlamaParse Extract Cost-Effective by source
Kind of resume# of resumesF1MedianGapRank
Everyday resumes21284.685.1−0.611th
Expert-built (Canva / LaTeX)3281.883.6−1.812th
Resume-builder templates1080.381.2−0.911th
Scanned & rotated1990.482.9+7.44th
Distractor layouts1885.585.3+0.19th

Layout

LlamaParse Extract Cost-Effective by layout
Kind of resume# of resumesF1MedianGapRank
Single column19582.884.0−1.111th
Two-column2386.786.8−0.110th
Two-column sidebar5482.380.0+2.36th
Three-column2085.984.8+1.14th
Table grid893.794.1−0.410th

Length

LlamaParse Extract Cost-Effective by length
Kind of resume# of resumesF1MedianGapRank
Short14887.787.1+0.67th
Medium9083.084.7−1.711th
Long6274.374.6−0.310th

By section of the resume

F1 by resume section, against the median system.

LlamaParse Extract Cost-Effective by resume section
SectionResumesF1MedianGap
Contact details30093.693.5+0.1
Experience30084.586.3−1.8
Education29787.689.4−1.8
Summary29081.781.7+0.0
Certifications24682.581.1+1.4
Awards9384.175.4+8.8
Skills28694.392.1+2.2
Languages16782.284.3−2.1
Custom sections21548.849.9−1.1
Interests6492.191.7+0.4
Volunteering7787.484.5+2.8
Profiles3769.466.7+2.7
Projects7882.783.2−0.5
Publications5172.278.6−6.3

On the published resumes

Best and worst of the 20 public resumes.

LlamaParse Extract Cost-Effective on published resumes
ResumeWhat makes it hardF1
Head of Developer Experience (Internal title: Jedi Master of Analytics) Unusual job titles92.7
Deputy Head of MissionRotated scan with hidden identity text88.9
Head of Digital MarketingCanva design with a photo85.6
Senior Game DesignerGame designer's Canva resume55.0
Senior ocularist and practice ownerHandwritten resume, six pages51.4
senior Consultant - package ImplementationHandwritten resume44.1

How it was run

LlamaParse Extract API, extraction mode cost_effective. 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 Cost-Effective 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_cost_effective --split test
$ resume-bench grade llamaextract_cost_effective --split test

Closest alternatives

Nearest in F1.

Embed this result

Shows its current rank. Updates with each run.

LlamaParse Extract Cost-Effective: #10 on ResumeBench resume extraction benchmark
<a href="https://www.resumebench.ai/systems/llamaparse-extract-cost-effective"><img src="https://www.resumebench.ai/badge/llamaparse-extract-cost-effective.svg" alt="LlamaParse Extract Cost-Effective: #10 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 Cost-Effective at resume extraction?

83.5 F1 on ResumeBench's 300 resumes, 10th of 18 models.

How much does LlamaParse Extract Cost-Effective cost to parse resumes?

$8.18 per 1,000 pages, about $0.023 per resume.

How fast is LlamaParse Extract Cost-Effective?

29 seconds per resume (median).

Which resumes is LlamaParse Extract Cost-Effective best at?

Scanned & rotated, rotated & warped pages and hidden text & prompt injections. Weakest on busy page, references and many entries.

What is a cheaper alternative to LlamaParse Extract Cost-Effective?

Qwen 3.8 Flash: 84.1 F1 for $2.98 per 1,000 pages.