Gemma 4 26B vs Qwen 3.5 9B: which is better at resume extraction?
Qwen 3.5 9B leads Gemma 4 26B by 13.2 points (76.0 vs 62.8 F1). Cost per 1,000 pages: $0.56 vs $1.31. Median time per resume: 29 s vs 138 s.
Gemma 4 26B wins on 61 of 299 resumes, Qwen 3.5 9B on 236, 2 tied.
62.98
Gemma 4 26B
A · Vision LLM
76.03
Qwen 3.5 9B
B · Vision LLM
Qwen 3.5 9B is ahead by 13.05 points on these 299 resumes.
- Many entries 82A 58.6 · B 73.9−15.3
- Long bullet lists 100A 57.9 · B 73.2−15.3
- Split across pages 84A 60.5 · B 73.4−12.9
- Big skills list 59A 62.8 · B 76.3−13.6
- Custom sections 104A 59.4 · B 73.0−13.7
- Combined headings 37A 60.9 · B 72.2−11.3
- Promotions 88A 59.5 · B 74.8−15.3
- Partial dates 138A 62.9 · B 75.6−12.7
- Overlapping jobs 69A 57.5 · B 72.2−14.7
- Accented text 13A 59.3 · B 83.7−24.5
Side by side
| Gemma 4 26B | Qwen 3.5 9B | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | Alibaba | |
| Rank | 16 of 18 | 15 of 18 |
| F1 | 62.8 | 76.0 |
| Precision | 69.1 | 78.9 |
| Recall | 66.2 | 80.3 |
| Cost per 1,000 pages | $0.56 | $1.31 |
| Median time per resume | 29 s | 138 s |
| Resumes won | 61 | 236 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Gemma 4 26B | Qwen 3.5 9B | Gap |
|---|---|---|---|---|
| Accented text (extraction challenges) | 13 | 59.3 | 83.7 | −24.5 |
| Scanned & rotated (source) | 19 | 40.7 | 64.5 | −23.8 |
| Rotated & warped pages (distractors) | 21 | 43.4 | 66.5 | −23.1 |
| Hidden text & prompt injections (distractors) | 14 | 48.6 | 70.2 | −21.6 |
| Typos kept on purpose (distractors) | 37 | 51.3 | 72.2 | −21.0 |
| Content as images (distractors) | 8 | 65.6 | 85.9 | −20.3 |
| Resume-builder templates (source) | 10 | 58.6 | 78.5 | −19.8 |
| Distractor layouts (source) | 18 | 62.4 | 80.4 | −18.0 |
| Wording traps (distractors) | 16 | 60.6 | 77.7 | −17.1 |
| Image only (reading the page) | 28 | 45.1 | 61.2 | −16.1 |
| Long bullet lists (extraction challenges) | 100 | 57.9 | 73.2 | −15.3 |
| Many entries (extraction challenges) | 82 | 58.6 | 73.9 | −15.3 |
| Section | Resumes | Gemma 4 26B | Qwen 3.5 9B | Gap |
|---|---|---|---|---|
| Contact details | 299 | 84.6 | 92.6 | −7.9 |
| Experience | 299 | 56.1 | 77.2 | −21.0 |
| Education | 296 | 68.2 | 80.8 | −12.6 |
| Summary | 289 | 35.6 | 77.9 | −42.3 |
| Certifications | 243 | 63.2 | 74.7 | −11.5 |
| Awards | 109 | 56.2 | 55.2 | +1.0 |
| Skills | 289 | 81.9 | 83.2 | −1.3 |
| Languages | 171 | 80.9 | 80.0 | +0.9 |
| Custom sections | 222 | 32.4 | 36.4 | −4.0 |
| Interests | 66 | 72.9 | 84.3 | −11.4 |
| Volunteering | 78 | 59.4 | 80.2 | −20.8 |
| Profiles | 36 | 58.8 | 51.3 | +7.5 |
| Projects | 82 | 55.6 | 75.5 | −19.9 |
| Publications | 53 | 54.5 | 76.3 | −21.8 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Gemma 4 26B | Qwen 3.5 9B | Gap |
|---|---|---|---|
| Creative Director: Accented text in a styled CV | 39.3 | 75.3 | −36.1 |
| Deputy Head of Mission: Rotated scan with hidden identity text | 34.1 | 65.0 | −30.9 |
| Graphic Designer: Graphic designer's Canva CV | 49.1 | 78.4 | −29.3 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 29.4 | 54.9 | −25.5 |
Which to pick
Qwen 3.5 9B is more accurate. Gemma 4 26B costs 58% less for 13.2 fewer points.
Qwen 3.5 9B does better on accented text, scanned & rotated and rotated & warped pages.
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 Deep Extract
- 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 GLM 5.3 Flash
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 Qwen 3.8 Flash
- 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
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