Extend Extract vs GPT-5.6: which is better at resume extraction?
GPT-5.6 leads Extend Extract by 3.8 points (84.7 vs 81.0 F1). Cost per 1,000 pages: $12 vs $74. Median time per resume: 31 s vs 38 s.
Extend Extract wins on 75 of 300 resumes, GPT-5.6 on 204, 21 tied.
80.97
Extend Extract
A · Document API
84.75
GPT-5.6
B · Vision LLM
GPT-5.6 is ahead by 3.78 points on these 300 resumes.
- Many entries 83A 80.2 · B 84.5−4.2
- Long bullet lists 101A 79.8 · B 84.9−5.1
- Split across pages 85A 79.6 · B 84.9−5.3
- Big skills list 60A 81.4 · B 85.2−3.7
- Custom sections 105A 77.3 · B 82.2−4.8
- Combined headings 37A 78.0 · B 84.1−6.1
- Promotions 88A 81.7 · B 85.0−3.3
- Partial dates 138A 80.1 · B 85.7−5.6
- Overlapping jobs 69A 81.2 · B 83.2−1.9
- Accented text 13A 84.3 · B 86.5−2.3
Side by side
| Extend Extract | GPT-5.6 | |
|---|---|---|
| Type | Document API | Vision LLM |
| Maker | Extend | OpenAI |
| Rank | 13 of 18 | 6 of 18 |
| F1 | 81.0 | 84.7 |
| Precision | 82.9 | 85.9 |
| Recall | 83.2 | 87.2 |
| Cost per 1,000 pages | $12 | $74 |
| Median time per resume | 31 s | 38 s |
| Resumes won | 75 | 204 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Extend Extract | GPT-5.6 | Gap |
|---|---|---|---|---|
| Expert-built (Canva / LaTeX) (source) | 32 | 77.1 | 85.2 | −8.1 |
| Scanned & rotated (source) | 19 | 79.5 | 71.6 | +8.0 |
| Two-column (layout) | 23 | 81.3 | 89.0 | −7.7 |
| Photo (reading the page) | 45 | 78.5 | 85.5 | −7.0 |
| Rotated & warped pages (distractors) | 21 | 79.8 | 73.2 | +6.6 |
| Wording traps (distractors) | 16 | 82.5 | 89.1 | −6.6 |
| Combined headings (extraction challenges) | 37 | 78.0 | 84.1 | −6.1 |
| Busy page (noise) | 18 | 78.2 | 84.0 | −5.8 |
| Partial dates (extraction challenges) | 138 | 80.1 | 85.7 | −5.6 |
| Image only (reading the page) | 28 | 70.7 | 65.4 | +5.3 |
| Split across pages (extraction challenges) | 85 | 79.6 | 84.9 | −5.3 |
| Long bullet lists (extraction challenges) | 101 | 79.8 | 84.9 | −5.1 |
| Section | Resumes | Extend Extract | GPT-5.6 | Gap |
|---|---|---|---|---|
| Contact details | 300 | 91.1 | 94.1 | −3.0 |
| Experience | 300 | 81.0 | 88.4 | −7.4 |
| Education | 297 | 90.2 | 91.9 | −1.7 |
| Summary | 290 | 80.7 | 83.4 | −2.8 |
| Certifications | 254 | 75.7 | 77.9 | −2.2 |
| Awards | 99 | 73.8 | 81.2 | −7.3 |
| Skills | 287 | 88.7 | 93.2 | −4.5 |
| Languages | 167 | 85.1 | 85.9 | −0.7 |
| Custom sections | 212 | 44.4 | 51.2 | −6.9 |
| Interests | 64 | 90.9 | 91.4 | −0.4 |
| Volunteering | 77 | 84.5 | 85.7 | −1.2 |
| Profiles | 36 | 42.6 | 69.4 | −26.8 |
| Projects | 82 | 80.2 | 88.2 | −8.0 |
| Publications | 51 | 85.1 | 77.8 | +7.3 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Extend Extract | GPT-5.6 | Gap |
|---|---|---|---|
| Head of Developer Experience (Internal title: Jedi Master of Analytics) : Unusual job titles | 55.1 | 93.6 | −38.5 |
| Postdoctoral Researcher: Academic CV, seven pages | 54.2 | 72.3 | −18.1 |
| Senior Data Scientist (Lead): Fifty skills, seven pages | 86.3 | 72.8 | +13.5 |
| Head of Digital Marketing: Canva design with a photo | 74.0 | 86.7 | −12.7 |
Which to pick
GPT-5.6 is more accurate. Extend Extract costs 84% less for 3.8 fewer points.
Extend Extract does better on scanned & rotated, rotated & warped pages and image only. GPT-5.6 does better on expert-built (canva / latex), two-column and photo.
Extend Extract against every other model
- Extend Extract vs LlamaParse Extract Agentic Plus
- Extend Extract vs LlamaParse Extract Agentic
- Extend Extract vs LlamaParse Extract Cost-Effective
- Extend Extract vs Reducto Deep Extract
- Extend Extract vs Reducto Extract
- Extend Extract vs Extend Max Context
- Extend Extract vs GPT-5.5
- Extend Extract vs GPT-5.4 Nano
- Extend Extract vs Claude Opus 4.8
- Extend Extract vs Gemini 3.5 Flash
- Extend Extract vs Gemma 4 26B
- Extend Extract vs Qwen 3.8 Flash
- Extend Extract vs Qwen 3.8 27B
- Extend Extract vs Qwen 3.5 35B
- Extend Extract vs Qwen 3.5 9B
- Extend Extract vs GLM 5.3 Flash
GPT-5.6 against every other model
- GPT-5.6 vs LlamaParse Extract Agentic Plus
- GPT-5.6 vs LlamaParse Extract Agentic
- GPT-5.6 vs LlamaParse Extract Cost-Effective
- GPT-5.6 vs Reducto Deep Extract
- GPT-5.6 vs Reducto Extract
- GPT-5.6 vs Extend Max Context
- GPT-5.6 vs GPT-5.5
- GPT-5.6 vs GPT-5.4 Nano
- GPT-5.6 vs Claude Opus 4.8
- GPT-5.6 vs Gemini 3.5 Flash
- GPT-5.6 vs Gemma 4 26B
- GPT-5.6 vs Qwen 3.8 Flash
- GPT-5.6 vs Qwen 3.8 27B
- GPT-5.6 vs Qwen 3.5 35B
- GPT-5.6 vs Qwen 3.5 9B
- GPT-5.6 vs GLM 5.3 Flash
Popular comparisons
- LlamaParse Extract Agentic Plus vs Reducto Deep Extract
- Extend Max Context vs Reducto Deep Extract
- Extend Max Context vs LlamaParse Extract Agentic Plus
- Claude Opus 4.8 vs GPT-5.6
- Gemini 3.5 Flash vs GPT-5.6
- Claude Opus 4.8 vs Gemini 3.5 Flash
- LlamaParse Extract Agentic vs LlamaParse Extract Agentic Plus
- Reducto Deep Extract vs Reducto Extract
- Gemma 4 26B vs Qwen 3.8 27B
- Document APIs vs vision LLMs