Gemma 4 26B vs Reducto Deep Extract: which is better at resume extraction?
Reducto Deep Extract leads Gemma 4 26B by 22.7 points (85.6 vs 62.8 F1). Cost per 1,000 pages: $0.56 vs $40. Median time per resume: 29 s vs 65 s.
Gemma 4 26B wins on 24 of 300 resumes, Reducto Deep Extract on 272, 4 tied.
62.82
Gemma 4 26B
A · Vision LLM
85.55
Reducto Deep Extract
B · Document API
Reducto Deep Extract is ahead by 22.73 points on these 300 resumes.
- Many entries 83A 58.0 · B 84.3−26.3
- Long bullet lists 101A 57.4 · B 84.5−27.1
- Split across pages 85A 60.0 · B 84.1−24.2
- Big skills list 60A 61.9 · B 85.8−23.8
- Custom sections 105A 58.9 · B 82.6−23.7
- Combined headings 37A 60.9 · B 82.7−21.9
- Promotions 88A 59.5 · B 86.0−26.5
- Partial dates 138A 62.9 · B 85.2−22.3
- Overlapping jobs 69A 57.5 · B 85.9−28.4
- Accented text 13A 59.3 · B 85.5−26.2
Side by side
| Gemma 4 26B | Reducto Deep Extract | |
|---|---|---|
| Type | Vision LLM | Document API |
| Maker | Reducto | |
| Rank | 16 of 18 | 2 of 18 |
| F1 | 62.8 | 85.6 |
| Precision | 69.1 | 87.5 |
| Recall | 66.2 | 87.3 |
| Cost per 1,000 pages | $0.56 | $40 |
| Median time per resume | 29 s | 65 s |
| Resumes won | 24 | 272 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Gemma 4 26B | Reducto Deep Extract | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 40.7 | 90.1 | −49.4 |
| Rotated & warped pages (distractors) | 21 | 43.4 | 89.3 | −45.8 |
| Hidden text & prompt injections (distractors) | 14 | 48.6 | 90.1 | −41.5 |
| Typos kept on purpose (distractors) | 37 | 51.3 | 87.9 | −36.6 |
| Image only (reading the page) | 28 | 45.1 | 76.4 | −31.3 |
| Two-column sidebar (layout) | 54 | 55.1 | 83.8 | −28.6 |
| Overlapping jobs (extraction challenges) | 69 | 57.5 | 85.9 | −28.4 |
| References (noise) | 55 | 62.5 | 90.2 | −27.7 |
| Long bullet lists (extraction challenges) | 101 | 57.4 | 84.5 | −27.1 |
| Promotions (extraction challenges) | 88 | 59.5 | 86.0 | −26.5 |
| Many entries (extraction challenges) | 83 | 58.0 | 84.3 | −26.3 |
| Accented text (extraction challenges) | 13 | 59.3 | 85.5 | −26.2 |
| Section | Resumes | Gemma 4 26B | Reducto Deep Extract | Gap |
|---|---|---|---|---|
| Contact details | 300 | 84.6 | 94.2 | −9.6 |
| Experience | 300 | 56.0 | 90.6 | −34.6 |
| Education | 297 | 68.0 | 92.0 | −24.0 |
| Summary | 290 | 35.5 | 85.2 | −49.7 |
| Certifications | 244 | 63.0 | 81.4 | −18.4 |
| Awards | 110 | 55.7 | 79.4 | −23.7 |
| Skills | 290 | 81.9 | 93.9 | −12.0 |
| Languages | 171 | 80.9 | 81.8 | −0.8 |
| Custom sections | 223 | 32.3 | 51.3 | −19.1 |
| Interests | 67 | 71.8 | 93.8 | −21.9 |
| Volunteering | 78 | 59.4 | 89.5 | −30.1 |
| Profiles | 37 | 57.2 | 69.6 | −12.4 |
| Projects | 83 | 55.0 | 88.1 | −33.1 |
| Publications | 53 | 54.5 | 79.3 | −24.8 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Gemma 4 26B | Reducto Deep Extract | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 34.1 | 89.2 | −55.1 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 29.4 | 66.6 | −37.2 |
| Senior Manufacturing Engineer: Third-person prose, no bullets | 62.3 | 93.9 | −31.5 |
| Head of Developer Experience (Internal title: Jedi Master of Analytics) : Unusual job titles | 53.7 | 83.4 | −29.7 |
Which to pick
Reducto Deep Extract is more accurate. Gemma 4 26B costs 99% less for 22.7 fewer points.
Gemma 4 26B does better on handwritten. Reducto Deep Extract does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections.
Gemma 4 26B against every other model
- Gemma 4 26B vs LlamaParse Extract Agentic Plus
- Gemma 4 26B vs LlamaParse Extract Agentic
- Gemma 4 26B vs LlamaParse Extract Cost-Effective
- Gemma 4 26B vs Reducto Extract
- Gemma 4 26B vs Extend Max Context
- Gemma 4 26B vs Extend Extract
- Gemma 4 26B vs GPT-5.6
- Gemma 4 26B vs GPT-5.5
- Gemma 4 26B vs GPT-5.4 Nano
- Gemma 4 26B vs Claude Opus 4.8
- Gemma 4 26B vs Gemini 3.5 Flash
- Gemma 4 26B vs Qwen 3.8 Flash
- Gemma 4 26B vs Qwen 3.8 27B
- Gemma 4 26B vs Qwen 3.5 35B
- Gemma 4 26B vs Qwen 3.5 9B
- Gemma 4 26B vs GLM 5.3 Flash
Reducto Deep Extract against every other model
- Reducto Deep Extract vs LlamaParse Extract Agentic Plus
- Reducto Deep Extract vs LlamaParse Extract Agentic
- Reducto Deep Extract vs LlamaParse Extract Cost-Effective
- Reducto Deep Extract vs Reducto Extract
- Reducto Deep Extract vs Extend Max Context
- Reducto Deep Extract vs Extend Extract
- Reducto Deep Extract vs GPT-5.6
- Reducto Deep Extract vs GPT-5.5
- Reducto Deep Extract vs GPT-5.4 Nano
- Reducto Deep Extract vs Claude Opus 4.8
- Reducto Deep Extract vs Gemini 3.5 Flash
- Reducto Deep Extract vs Qwen 3.8 Flash
- Reducto Deep Extract vs Qwen 3.8 27B
- Reducto Deep Extract vs Qwen 3.5 35B
- Reducto Deep Extract vs Qwen 3.5 9B
- Reducto Deep Extract 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