Qwen 3.8 27B vs Reducto Deep Extract: which is better at resume extraction?
Reducto Deep Extract leads Qwen 3.8 27B by 1.8 points (85.6 vs 83.8 F1). Cost per 1,000 pages: $9.59 vs $40. Median time per resume: 36 s vs 65 s.
Qwen 3.8 27B wins on 102 of 300 resumes, Reducto Deep Extract on 152, 46 tied.
83.77
Qwen 3.8 27B
A · Vision LLM
85.55
Reducto Deep Extract
B · Document API
Reducto Deep Extract is ahead by 1.78 points on these 300 resumes.
- Many entries 83A 81.6 · B 84.3−2.7
- Long bullet lists 101A 80.8 · B 84.5−3.7
- Split across pages 85A 81.8 · B 84.1−2.3
- Big skills list 60A 82.9 · B 85.8−2.9
- Custom sections 105A 81.0 · B 82.6−1.6
- Combined headings 37A 83.7 · B 82.7+1.0
- Promotions 88A 83.2 · B 86.0−2.7
- Partial dates 138A 83.6 · B 85.2−1.6
- Overlapping jobs 69A 82.4 · B 85.9−3.5
- Accented text 13A 84.9 · B 85.5−0.5
Side by side
| Qwen 3.8 27B | Reducto Deep Extract | |
|---|---|---|
| Type | Vision LLM | Document API |
| Maker | Alibaba | Reducto |
| Rank | 9 of 18 | 2 of 18 |
| F1 | 83.8 | 85.6 |
| Precision | 85.4 | 87.5 |
| Recall | 85.5 | 87.3 |
| Cost per 1,000 pages | $9.59 | $40 |
| Median time per resume | 36 s | 65 s |
| Resumes won | 102 | 152 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Qwen 3.8 27B | Reducto Deep Extract | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 78.4 | 90.1 | −11.7 |
| Rotated & warped pages (distractors) | 21 | 78.7 | 89.3 | −10.5 |
| Resume-builder templates (source) | 10 | 86.7 | 77.6 | +9.1 |
| Hidden text & prompt injections (distractors) | 14 | 82.4 | 90.1 | −7.7 |
| Image only (reading the page) | 28 | 70.5 | 76.4 | −5.9 |
| Two-column sidebar (layout) | 54 | 78.4 | 83.8 | −5.4 |
| References (noise) | 55 | 84.9 | 90.2 | −5.3 |
| Typos kept on purpose (distractors) | 37 | 82.9 | 87.9 | −5.0 |
| Long (length) | 62 | 73.4 | 77.5 | −4.1 |
| Long bullet lists (extraction challenges) | 101 | 80.8 | 84.5 | −3.7 |
| Overlapping jobs (extraction challenges) | 69 | 82.4 | 85.9 | −3.5 |
| Sidebar or grid (reading the page) | 82 | 81.8 | 85.2 | −3.5 |
| Section | Resumes | Qwen 3.8 27B | Reducto Deep Extract | Gap |
|---|---|---|---|---|
| Contact details | 300 | 93.2 | 94.2 | −1.1 |
| Experience | 300 | 86.2 | 90.6 | −4.4 |
| Education | 297 | 87.2 | 92.0 | −4.8 |
| Summary | 290 | 81.7 | 85.2 | −3.4 |
| Certifications | 244 | 82.7 | 81.4 | +1.3 |
| Awards | 97 | 78.4 | 79.4 | −1.0 |
| Skills | 284 | 92.7 | 93.9 | −1.2 |
| Languages | 168 | 85.4 | 81.8 | +3.6 |
| Custom sections | 216 | 51.2 | 51.3 | −0.1 |
| Interests | 63 | 94.6 | 93.8 | +0.8 |
| Volunteering | 79 | 84.7 | 89.5 | −4.9 |
| Profiles | 34 | 72.5 | 69.6 | +2.9 |
| Projects | 78 | 85.5 | 88.1 | −2.6 |
| Publications | 51 | 77.2 | 79.3 | −2.1 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Qwen 3.8 27B | Reducto Deep Extract | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 65.6 | 89.2 | −23.6 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 88.5 | 66.6 | +21.9 |
| Senior Game Designer: Game designer's Canva resume | 74.0 | 52.9 | +21.1 |
| Creative Director: Accented text in a styled CV | 78.3 | 61.2 | +17.2 |
Which to pick
Reducto Deep Extract is more accurate. Qwen 3.8 27B costs 76% less for 1.8 fewer points.
Qwen 3.8 27B does better on resume-builder templates, wording traps and table grid. Reducto Deep Extract does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections.
Qwen 3.8 27B against every other model
- Qwen 3.8 27B vs LlamaParse Extract Agentic Plus
- Qwen 3.8 27B vs LlamaParse Extract Agentic
- Qwen 3.8 27B vs LlamaParse Extract Cost-Effective
- Qwen 3.8 27B vs Reducto Extract
- Qwen 3.8 27B vs Extend Max Context
- Qwen 3.8 27B vs Extend Extract
- Qwen 3.8 27B vs GPT-5.6
- Qwen 3.8 27B vs GPT-5.5
- Qwen 3.8 27B vs GPT-5.4 Nano
- Qwen 3.8 27B vs Claude Opus 4.8
- Qwen 3.8 27B vs Gemini 3.5 Flash
- Qwen 3.8 27B vs Gemma 4 26B
- Qwen 3.8 27B vs Qwen 3.8 Flash
- Qwen 3.8 27B vs Qwen 3.5 35B
- Qwen 3.8 27B vs Qwen 3.5 9B
- Qwen 3.8 27B 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 Gemma 4 26B
- Reducto Deep Extract vs Qwen 3.8 Flash
- 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