Qwen 3.5 9B vs Qwen 3.8 Flash: which is better at resume extraction?
Qwen 3.8 Flash leads Qwen 3.5 9B by 8.0 points (84.1 vs 76.0 F1). Cost per 1,000 pages: $1.31 vs $2.98. Median time per resume: 138 s vs 122 s.
Qwen 3.5 9B wins on 66 of 297 resumes, Qwen 3.8 Flash on 213, 18 tied.
76.19
Qwen 3.5 9B
A · Vision LLM
84.17
Qwen 3.8 Flash
B · Vision LLM
Qwen 3.8 Flash is ahead by 7.99 points on these 297 resumes.
- Many entries 82A 73.9 · B 82.4−8.5
- Long bullet lists 100A 73.2 · B 82.3−9.1
- Split across pages 84A 73.4 · B 82.9−9.5
- Big skills list 59A 76.3 · B 84.0−7.6
- Custom sections 104A 73.0 · B 81.1−8.1
- Combined headings 37A 72.2 · B 82.6−10.5
- Promotions 86A 75.3 · B 83.8−8.5
- Partial dates 138A 75.6 · B 83.3−7.7
- Overlapping jobs 68A 72.2 · B 81.4−9.2
- Accented text 13A 83.7 · B 86.9−3.2
Side by side
| Qwen 3.5 9B | Qwen 3.8 Flash | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | Alibaba | Alibaba |
| Rank | 15 of 18 | 8 of 18 |
| F1 | 76.0 | 84.1 |
| Precision | 78.9 | 85.4 |
| Recall | 80.3 | 86.2 |
| Cost per 1,000 pages | $1.31 | $2.98 |
| Median time per resume | 138 s | 122 s |
| Resumes won | 66 | 213 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Qwen 3.5 9B | Qwen 3.8 Flash | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 17 | 65.9 | 86.3 | −20.5 |
| Hidden text & prompt injections (distractors) | 13 | 70.2 | 89.0 | −18.8 |
| Rotated & warped pages (distractors) | 19 | 67.9 | 86.0 | −18.1 |
| Image only (reading the page) | 26 | 61.9 | 74.7 | −12.9 |
| Typos kept on purpose (distractors) | 35 | 73.3 | 85.7 | −12.4 |
| Busy page (noise) | 18 | 67.6 | 79.8 | −12.3 |
| Two-column sidebar (layout) | 52 | 70.3 | 81.6 | −11.3 |
| Photo (reading the page) | 44 | 75.8 | 86.3 | −10.6 |
| Combined headings (extraction challenges) | 37 | 72.2 | 82.6 | −10.5 |
| Sidebar or grid (reading the page) | 80 | 74.0 | 83.8 | −9.8 |
| Split across pages (extraction challenges) | 84 | 73.4 | 82.9 | −9.5 |
| Overlapping jobs (extraction challenges) | 68 | 72.2 | 81.4 | −9.2 |
| Section | Resumes | Qwen 3.5 9B | Qwen 3.8 Flash | Gap |
|---|---|---|---|---|
| Contact details | 297 | 92.5 | 93.7 | −1.2 |
| Experience | 297 | 77.3 | 87.9 | −10.6 |
| Education | 294 | 81.1 | 90.1 | −8.9 |
| Summary | 288 | 78.1 | 81.5 | −3.4 |
| Certifications | 245 | 74.7 | 82.2 | −7.5 |
| Awards | 125 | 55.5 | 75.7 | −20.2 |
| Skills | 291 | 83.5 | 89.9 | −6.5 |
| Languages | 170 | 79.8 | 84.6 | −4.8 |
| Custom sections | 212 | 36.5 | 54.0 | −17.5 |
| Interests | 66 | 85.3 | 91.1 | −5.8 |
| Volunteering | 79 | 80.5 | 83.9 | −3.4 |
| Profiles | 38 | 51.3 | 75.5 | −24.2 |
| Projects | 79 | 75.5 | 84.7 | −9.2 |
| Publications | 53 | 76.3 | 81.8 | −5.5 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Qwen 3.5 9B | Qwen 3.8 Flash | Gap |
|---|---|---|---|
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 54.9 | 83.8 | −28.9 |
| Deputy Head of Mission: Rotated scan with hidden identity text | 65.0 | 88.0 | −22.9 |
| Senior Controls Engineer: Five custom sections | 65.2 | 85.3 | −20.1 |
| Senior Customer Service Representative: Busy Canva page | 75.6 | 94.9 | −19.3 |
Which to pick
Qwen 3.8 Flash is more accurate. Qwen 3.5 9B costs 56% less for 8.0 fewer points.
Qwen 3.8 Flash does better on scanned & rotated, hidden text & prompt injections and rotated & warped pages.
Qwen 3.5 9B against every other model
- Qwen 3.5 9B vs LlamaParse Extract Agentic Plus
- Qwen 3.5 9B vs LlamaParse Extract Agentic
- Qwen 3.5 9B vs LlamaParse Extract Cost-Effective
- Qwen 3.5 9B vs Reducto Deep Extract
- Qwen 3.5 9B vs Reducto Extract
- Qwen 3.5 9B vs Extend Max Context
- Qwen 3.5 9B vs Extend Extract
- Qwen 3.5 9B vs GPT-5.6
- Qwen 3.5 9B vs GPT-5.5
- Qwen 3.5 9B vs GPT-5.4 Nano
- Qwen 3.5 9B vs Claude Opus 4.8
- Qwen 3.5 9B vs Gemini 3.5 Flash
- Qwen 3.5 9B vs Gemma 4 26B
- Qwen 3.5 9B vs Qwen 3.8 27B
- Qwen 3.5 9B vs Qwen 3.5 35B
- Qwen 3.5 9B vs GLM 5.3 Flash
Qwen 3.8 Flash against every other model
- Qwen 3.8 Flash vs LlamaParse Extract Agentic Plus
- Qwen 3.8 Flash vs LlamaParse Extract Agentic
- Qwen 3.8 Flash vs LlamaParse Extract Cost-Effective
- Qwen 3.8 Flash vs Reducto Deep Extract
- Qwen 3.8 Flash vs Reducto Extract
- Qwen 3.8 Flash vs Extend Max Context
- Qwen 3.8 Flash vs Extend Extract
- Qwen 3.8 Flash vs GPT-5.6
- Qwen 3.8 Flash vs GPT-5.5
- Qwen 3.8 Flash vs GPT-5.4 Nano
- Qwen 3.8 Flash vs Claude Opus 4.8
- Qwen 3.8 Flash vs Gemini 3.5 Flash
- Qwen 3.8 Flash vs Gemma 4 26B
- Qwen 3.8 Flash vs Qwen 3.8 27B
- Qwen 3.8 Flash vs Qwen 3.5 35B
- Qwen 3.8 Flash 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