Skip to content

Extend · Document API

Extend Max Context on ResumeBench: 84.7 F1, rank 7 of 18

Extend Max Context ranks 7th of 18 on ResumeBench: 84.7 F1, $50 per 1,000 pages, 43 seconds per resume. It is 2.4 points behind LlamaParse Extract Agentic Plus. Reducto Deep Extract scores 85.6 for $40.

4th of 7 document APIs. Strongest on scanned & rotated (+8.5 points), rotated & warped pages (+7.8 points) and typos kept on purpose (+5.6 points). Weakest on handwritten (−4.9 points), combined headings (−2.3 points) and two-column (−1.5 points).

Rank
7/18
F1
84.7
Precision
86.6
Recall
86.4
Perfect resumes
35/300
Per 1,000 pages
$50
Median per resume
43 s

Averaged over 300 resumes. Dates (scored separately): 97.8. 90% finish within 82 s. Cost from list price.

Where Extend Max Context wins and loses

F1 by kind of resume, against the median system.

Reading the page

Extend Max Context by reading the page
Kind of resume# of resumesF1MedianGapRank
Handwritten946.551.5−4.917th
Image only2876.571.4+5.14th
Sidebar or grid8284.481.9+2.45th
Photo4585.684.0+1.55th

Distractors

Extend Max Context by distractors
Kind of resume# of resumesF1MedianGapRank
Rotated & warped pages2190.882.9+7.83rd
Hidden text & prompt injections1489.684.5+5.16th
Content as images889.488.2+1.27th
Wording traps1685.684.5+1.16th
Typos kept on purpose3789.383.7+5.63rd

Extraction challenges

Extend Max Context by extraction challenges
Kind of resume# of resumesF1MedianGapRank
Many entries8385.681.9+3.71st
Long bullet lists10184.881.4+3.33rd
Split across pages8583.182.2+0.98th
Big skills list6085.883.6+2.23rd
Custom sections10582.080.8+1.27th
Combined headings3780.482.7−2.311th
Promotions8885.883.5+2.34th
Partial dates13884.883.4+1.46th
Overlapping jobs6985.281.7+3.53rd
Accented text1388.885.5+3.33rd

Noise

Extend Max Context by noise
Kind of resume# of resumesF1MedianGapRank
Repeated header6785.585.2+0.38th
References5590.385.2+5.12nd
Busy page1883.280.1+3.14th

Source

Extend Max Context by source
Kind of resume# of resumesF1MedianGapRank
Everyday resumes21285.785.1+0.68th
Expert-built (Canva / LaTeX)3284.883.6+1.24th
Resume-builder templates1080.681.2−0.610th
Scanned & rotated1991.482.9+8.53rd
Distractor layouts1887.085.3+1.75th

Layout

Extend Max Context by layout
Kind of resume# of resumesF1MedianGapRank
Single column19584.884.0+0.87th
Two-column2385.386.8−1.511th
Two-column sidebar5483.080.0+3.05th
Three-column2084.884.8−0.110th
Table grid892.694.1−1.513th

Length

Extend Max Context by length
Kind of resume# of resumesF1MedianGapRank
Short14887.987.1+0.76th
Medium9085.184.7+0.47th
Long6276.774.6+2.18th

By section of the resume

F1 by resume section, against the median system.

Extend Max Context by resume section
SectionResumesF1MedianGap
Contact details30095.593.5+2.0
Experience30087.986.3+1.6
Education29791.289.4+1.7
Summary29087.681.7+5.9
Certifications25181.881.1+0.7
Awards9278.675.4+3.2
Skills28588.792.1−3.4
Languages16784.684.3+0.3
Custom sections21250.249.9+0.3
Interests6491.391.7−0.4
Volunteering7687.384.5+2.8
Profiles3469.666.7+2.9
Projects8084.583.2+1.3
Publications5191.278.6+12.6

On the published resumes

Best and worst of the 20 public resumes.

Extend Max Context on published resumes
ResumeWhat makes it hardF1
Senior Controls EngineerFive custom sections96.2
Senior Data Scientist (Lead)Fifty skills, seven pages95.9
Senior Manufacturing EngineerThird-person prose, no bullets92.1
Senior Game DesignerGame designer's Canva resume66.4
senior Consultant - package ImplementationHandwritten resume28.6
Senior ocularist and practice ownerHandwritten resume, six pages26.9

How it was run

Extend extraction with the large_array_max_context array strategy. Same PDF, schema and instructions as every other system, one run per resume (4 November 2026).

Max Context makes several passes through the document for long lists (the large_array_max_context strategy), at about twice the credits.

Extend Max Context on Extend’s site

Run the open-source harness: list its pipelines, or grade your own predictions (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 providers
$ resume-bench grade-file my_predictions.jsonl --split test

Closest alternatives

Nearest in F1.

Embed this result

Shows its current rank. Updates with each run.

Extend Max Context: #7 on ResumeBench resume extraction benchmark
<a href="https://www.resumebench.ai/systems/extend-max-context"><img src="https://www.resumebench.ai/badge/extend-max-context.svg" alt="Extend Max Context: #7 on ResumeBench resume extraction benchmark" height="28"></a>

Choosing a resume extraction model

All 18 models · Leaderboard

Frequently asked questions (FAQs)

How accurate is Extend Max Context at resume extraction?

84.7 F1 on ResumeBench's 300 resumes, 7th of 18 models.

How much does Extend Max Context cost to parse resumes?

$50 per 1,000 pages, about $0.140 per resume.

How fast is Extend Max Context?

43 seconds per resume (median).

Which resumes is Extend Max Context best at?

Scanned & rotated, rotated & warped pages and typos kept on purpose. Weakest on handwritten, combined headings and two-column.

What is a cheaper alternative to Extend Max Context?

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