GPT-5.4 Nano vs Qwen 3.5 35B: which is better at resume extraction?
Qwen 3.5 35B leads GPT-5.4 Nano by 16.1 points (77.9 vs 61.9 F1). Cost per 1,000 pages: $2.76 vs $3.91. Median time per resume: 20 s vs 78 s.
GPT-5.4 Nano wins on 67 of 300 resumes, Qwen 3.5 35B on 232, 1 tied.
61.86
GPT-5.4 Nano
A · Vision LLM
77.95
Qwen 3.5 35B
B · Vision LLM
Qwen 3.5 35B is ahead by 16.09 points on these 300 resumes.
- Many entries 83A 59.7 · B 78.9−19.3
- Long bullet lists 101A 61.0 · B 76.3−15.3
- Split across pages 85A 61.7 · B 76.5−14.8
- Big skills list 60A 64.9 · B 76.6−11.7
- Custom sections 105A 59.5 · B 77.0−17.5
- Combined headings 37A 64.6 · B 74.6−10.0
- Promotions 88A 58.7 · B 76.7−18.0
- Partial dates 138A 62.6 · B 78.9−16.3
- Overlapping jobs 69A 58.6 · B 76.2−17.6
- Accented text 13A 55.9 · B 80.4−24.6
Side by side
| GPT-5.4 Nano | Qwen 3.5 35B | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | OpenAI | Alibaba |
| Rank | 17 of 18 | 14 of 18 |
| F1 | 61.9 | 77.9 |
| Precision | 70.1 | 78.8 |
| Recall | 66.3 | 83.7 |
| Cost per 1,000 pages | $2.76 | $3.91 |
| Median time per resume | 20 s | 78 s |
| Resumes won | 67 | 232 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | GPT-5.4 Nano | Qwen 3.5 35B | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 6.3 | 75.3 | −69.0 |
| Rotated & warped pages (distractors) | 21 | 11.7 | 76.5 | −64.9 |
| Hidden text & prompt injections (distractors) | 14 | 22.0 | 79.2 | −57.2 |
| Image only (reading the page) | 28 | 21.0 | 69.7 | −48.7 |
| Typos kept on purpose (distractors) | 37 | 33.5 | 80.4 | −46.8 |
| Content as images (distractors) | 8 | 61.3 | 90.3 | −29.0 |
| Accented text (extraction challenges) | 13 | 55.9 | 80.4 | −24.6 |
| Two-column sidebar (layout) | 54 | 47.4 | 71.5 | −24.1 |
| Distractor layouts (source) | 18 | 62.2 | 85.7 | −23.5 |
| Expert-built (Canva / LaTeX) (source) | 32 | 60.3 | 83.7 | −23.4 |
| Wording traps (distractors) | 16 | 62.8 | 84.8 | −22.0 |
| Photo (reading the page) | 45 | 59.4 | 80.0 | −20.6 |
| Section | Resumes | GPT-5.4 Nano | Qwen 3.5 35B | Gap |
|---|---|---|---|---|
| Contact details | 300 | 81.2 | 92.1 | −10.9 |
| Experience | 300 | 63.7 | 79.3 | −15.6 |
| Education | 297 | 62.2 | 81.8 | −19.6 |
| Summary | 290 | 71.7 | 80.4 | −8.7 |
| Certifications | 244 | 53.6 | 79.1 | −25.5 |
| Awards | 102 | 44.9 | 61.9 | −16.9 |
| Skills | 291 | 74.8 | 87.8 | −13.0 |
| Languages | 170 | 69.0 | 79.4 | −10.4 |
| Custom sections | 231 | 26.3 | 40.2 | −13.9 |
| Interests | 66 | 70.0 | 89.7 | −19.7 |
| Volunteering | 80 | 59.9 | 83.1 | −23.3 |
| Profiles | 66 | 23.7 | 65.8 | −42.1 |
| Projects | 81 | 48.8 | 76.7 | −28.0 |
| Publications | 51 | 30.7 | 75.0 | −44.3 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | GPT-5.4 Nano | Qwen 3.5 35B | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 1.7 | 79.6 | −78.0 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 32.5 | 82.8 | −50.3 |
| Partner: Consulting CV, six pages | 27.8 | 75.5 | −47.7 |
| Head of Digital Marketing: Canva design with a photo | 39.3 | 84.5 | −45.2 |
Which to pick
Qwen 3.5 35B is more accurate. GPT-5.4 Nano costs 29% less for 16.1 fewer points.
Qwen 3.5 35B does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections.
GPT-5.4 Nano against every other model
- GPT-5.4 Nano vs LlamaParse Extract Agentic Plus
- GPT-5.4 Nano vs LlamaParse Extract Agentic
- GPT-5.4 Nano vs LlamaParse Extract Cost-Effective
- GPT-5.4 Nano vs Reducto Deep Extract
- GPT-5.4 Nano vs Reducto Extract
- GPT-5.4 Nano vs Extend Max Context
- GPT-5.4 Nano vs Extend Extract
- GPT-5.4 Nano vs GPT-5.6
- GPT-5.4 Nano vs GPT-5.5
- GPT-5.4 Nano vs Claude Opus 4.8
- GPT-5.4 Nano vs Gemini 3.5 Flash
- GPT-5.4 Nano vs Gemma 4 26B
- GPT-5.4 Nano vs Qwen 3.8 Flash
- GPT-5.4 Nano vs Qwen 3.8 27B
- GPT-5.4 Nano vs Qwen 3.5 9B
- GPT-5.4 Nano vs GLM 5.3 Flash
Qwen 3.5 35B against every other model
- Qwen 3.5 35B vs LlamaParse Extract Agentic Plus
- Qwen 3.5 35B vs LlamaParse Extract Agentic
- Qwen 3.5 35B vs LlamaParse Extract Cost-Effective
- Qwen 3.5 35B vs Reducto Deep Extract
- Qwen 3.5 35B vs Reducto Extract
- Qwen 3.5 35B vs Extend Max Context
- Qwen 3.5 35B vs Extend Extract
- Qwen 3.5 35B vs GPT-5.6
- Qwen 3.5 35B vs GPT-5.5
- Qwen 3.5 35B vs Claude Opus 4.8
- Qwen 3.5 35B vs Gemini 3.5 Flash
- Qwen 3.5 35B vs Gemma 4 26B
- Qwen 3.5 35B vs Qwen 3.8 Flash
- Qwen 3.5 35B vs Qwen 3.8 27B
- Qwen 3.5 35B vs Qwen 3.5 9B
- Qwen 3.5 35B 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