LlamaParse Extract Agentic vs Qwen 3.8 27B: which is better at resume extraction?
LlamaParse Extract Agentic leads Qwen 3.8 27B by 1.8 points (85.5 vs 83.8 F1). Cost per 1,000 pages: $23 vs $9.59. Median time per resume: 55 s vs 36 s.
LlamaParse Extract Agentic wins on 145 of 300 resumes, Qwen 3.8 27B on 103, 52 tied.
85.55
LlamaParse Extract Agentic
A · Document API
83.77
Qwen 3.8 27B
B · Vision LLM
LlamaParse Extract Agentic is ahead by 1.78 points on these 300 resumes.
- Many entries 83A 83.0 · B 81.6+1.4
- Long bullet lists 101A 83.1 · B 80.8+2.3
- Split across pages 85A 83.8 · B 81.8+2.0
- Big skills list 60A 83.7 · B 82.9+0.9
- Custom sections 105A 82.5 · B 81.0+1.5
- Combined headings 37A 83.2 · B 83.7−0.5
- Promotions 88A 85.5 · B 83.2+2.3
- Partial dates 138A 85.6 · B 83.6+2.0
- Overlapping jobs 69A 83.5 · B 82.4+1.1
- Accented text 13A 91.1 · B 84.9+6.2
Side by side
| LlamaParse Extract Agentic | Qwen 3.8 27B | |
|---|---|---|
| Type | Document API | Vision LLM |
| Maker | LlamaIndex | Alibaba |
| Rank | 3 of 18 | 9 of 18 |
| F1 | 85.5 | 83.8 |
| Precision | 86.8 | 85.4 |
| Recall | 87.3 | 85.5 |
| Cost per 1,000 pages | $23 | $9.59 |
| Median time per resume | 55 s | 36 s |
| Resumes won | 145 | 103 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | LlamaParse Extract Agentic | Qwen 3.8 27B | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 91.6 | 78.4 | +13.2 |
| Rotated & warped pages (distractors) | 21 | 90.9 | 78.7 | +12.2 |
| Handwritten (reading the page) | 9 | 63.5 | 51.6 | +11.9 |
| Image only (reading the page) | 28 | 81.5 | 70.5 | +11.0 |
| Hidden text & prompt injections (distractors) | 14 | 91.6 | 82.4 | +9.2 |
| Typos kept on purpose (distractors) | 37 | 90.0 | 82.9 | +7.1 |
| Accented text (extraction challenges) | 13 | 91.1 | 84.9 | +6.2 |
| Two-column sidebar (layout) | 54 | 84.0 | 78.4 | +5.7 |
| Long (length) | 62 | 78.0 | 73.4 | +4.6 |
| Sidebar or grid (reading the page) | 82 | 85.5 | 81.8 | +3.7 |
| Resume-builder templates (source) | 10 | 83.2 | 86.7 | −3.5 |
| Busy page (noise) | 18 | 83.5 | 80.4 | +3.2 |
| Section | Resumes | LlamaParse Extract Agentic | Qwen 3.8 27B | Gap |
|---|---|---|---|---|
| Contact details | 300 | 94.5 | 93.2 | +1.4 |
| Experience | 300 | 89.6 | 86.2 | +3.4 |
| Education | 297 | 91.3 | 87.2 | +4.1 |
| Summary | 290 | 84.5 | 81.7 | +2.8 |
| Certifications | 252 | 80.9 | 82.7 | −1.8 |
| Awards | 93 | 80.3 | 78.4 | +1.9 |
| Skills | 286 | 95.1 | 92.7 | +2.4 |
| Languages | 167 | 84.2 | 85.4 | −1.2 |
| Custom sections | 208 | 52.1 | 51.2 | +0.9 |
| Interests | 64 | 93.0 | 94.6 | −1.6 |
| Volunteering | 77 | 87.7 | 84.7 | +3.0 |
| Profiles | 35 | 67.6 | 72.5 | −4.9 |
| Projects | 78 | 86.2 | 85.5 | +0.7 |
| Publications | 51 | 80.5 | 77.2 | +3.3 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | LlamaParse Extract Agentic | Qwen 3.8 27B | Gap |
|---|---|---|---|
| Senior ocularist and practice owner: Handwritten resume, six pages | 70.4 | 41.2 | +29.2 |
| senior Consultant - package Implementation: Handwritten resume | 74.0 | 46.5 | +27.5 |
| Deputy Head of Mission: Rotated scan with hidden identity text | 89.5 | 65.6 | +23.9 |
| Senior Controls Engineer: Five custom sections | 89.6 | 73.5 | +16.2 |
Which to pick
LlamaParse Extract Agentic is more accurate. Qwen 3.8 27B costs 58% less for 1.8 fewer points.
LlamaParse Extract Agentic does better on scanned & rotated, rotated & warped pages and handwritten. Qwen 3.8 27B does better on resume-builder templates and medium.
LlamaParse Extract Agentic against every other model
- LlamaParse Extract Agentic vs LlamaParse Extract Agentic Plus
- LlamaParse Extract Agentic vs LlamaParse Extract Cost-Effective
- LlamaParse Extract Agentic vs Reducto Deep Extract
- LlamaParse Extract Agentic vs Reducto Extract
- LlamaParse Extract Agentic vs Extend Max Context
- LlamaParse Extract Agentic vs Extend Extract
- LlamaParse Extract Agentic vs GPT-5.6
- LlamaParse Extract Agentic vs GPT-5.5
- LlamaParse Extract Agentic vs GPT-5.4 Nano
- LlamaParse Extract Agentic vs Claude Opus 4.8
- LlamaParse Extract Agentic vs Gemini 3.5 Flash
- LlamaParse Extract Agentic vs Gemma 4 26B
- LlamaParse Extract Agentic vs Qwen 3.8 Flash
- LlamaParse Extract Agentic vs Qwen 3.5 35B
- LlamaParse Extract Agentic vs Qwen 3.5 9B
- LlamaParse Extract Agentic vs GLM 5.3 Flash
Qwen 3.8 27B against every other model
- Qwen 3.8 27B vs LlamaParse Extract Agentic Plus
- Qwen 3.8 27B vs LlamaParse Extract Cost-Effective
- Qwen 3.8 27B vs Reducto Deep Extract
- 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
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