Skip to content
vs

Extend Extract vs GPT-5.5: which is better at resume extraction?

GPT-5.5 leads Extend Extract by 2.5 points (83.5 vs 81.0 F1). Cost per 1,000 pages: $12 vs $87. Median time per resume: 31 s vs 48 s.

Extend Extract wins on 84 of 300 resumes, GPT-5.5 on 192, 24 tied.

80.97

Extend Extract

A · Document API

−2.53A − B, points of F1

83.50

GPT-5.5

B · Vision LLM

A better on 84level on 24B better on 192

GPT-5.5 is ahead by 2.53 points on these 300 resumes.

  • Many entries 83A 80.2 · B 83.8−3.6
  • Long bullet lists 101A 79.8 · B 84.3−4.5
  • Split across pages 85A 79.6 · B 85.0−5.4
  • Big skills list 60A 81.4 · B 85.5−4.1
  • Custom sections 105A 77.3 · B 80.8−3.5
  • Combined headings 37A 78.0 · B 83.4−5.4
  • Promotions 88A 81.7 · B 84.0−2.3
  • Partial dates 138A 80.1 · B 84.0−4.0
  • Overlapping jobs 69A 81.2 · B 82.1−0.8
  • Accented text 13A 84.3 · B 83.9+0.4

Side by side

Extend Extract vs GPT-5.5
Extend ExtractGPT-5.5
TypeDocument APIVision LLM
MakerExtendOpenAI
Rank13 of 1811 of 18
F181.083.5
Precision82.984.9
Recall83.286.0
Cost per 1,000 pages$12$87
Median time per resume31 s48 s
Resumes won84192

Extend Extract in detailGPT-5.5 in detail

Where each one wins

Where the two differ most.

Extend Extract vs GPT-5.5 by kind of resume
Kind of resume# of resumesExtend ExtractGPT-5.5Gap
Scanned & rotated (source)1979.560.7+18.8
Rotated & warped pages (distractors)2179.863.4+16.5
Image only (reading the page)2870.757.0+13.7
Hidden text & prompt injections (distractors)1482.469.1+13.3
Typos kept on purpose (distractors)3782.373.3+9.0
Two-column (layout)2381.388.3−7.1
Table grid (layout)889.896.8−6.9
Expert-built (Canva / LaTeX) (source)3277.183.6−6.5
Photo (reading the page)4578.584.0−5.5
Combined headings (extraction challenges)3778.083.4−5.4
Split across pages (extraction challenges)8579.685.0−5.4
Repeated header (noise)6782.286.7−4.5
Extend Extract vs GPT-5.5 by section
SectionResumesExtend ExtractGPT-5.5Gap
Contact details30091.193.5−2.4
Experience30081.087.4−6.4
Education29790.288.7+1.4
Summary29080.781.8−1.1
Certifications25475.777.3−1.6
Awards9973.875.3−1.4
Skills28788.791.5−2.8
Languages16785.185.7−0.6
Custom sections21244.449.7−5.3
Interests6490.992.4−1.5
Volunteering7784.585.0−0.5
Profiles3642.668.5−25.9
Projects8280.286.4−6.2
Publications5185.182.5+2.6

Published resumes they disagree on most

Open one to see both outputs field by field.

Which to pick

GPT-5.5 is more accurate. Extend Extract costs 86% less for 2.5 fewer points.

Extend Extract does better on scanned & rotated, rotated & warped pages and image only. GPT-5.5 does better on two-column, table grid and expert-built (canva / latex).

Extend Extract against every other model

GPT-5.5 against every other model

Popular comparisons

Frequently asked questions (FAQs)

Which is better for resume extraction, Extend Extract or GPT-5.5?

GPT-5.5: 83.5 F1 against 81.0.

Which is cheaper, Extend Extract or GPT-5.5?

Extend Extract: $12 per 1,000 pages against $87.

Where does each one do better?

Extend Extract: scanned & rotated, rotated & warped pages and image only. GPT-5.5: two-column, table grid and expert-built (canva / latex).