vs
GPT-5.6 vs Qwen 3.5 9B: which is better at resume extraction?
GPT-5.6 leads Qwen 3.5 9B by 8.7 points (84.7 vs 76.0 F1). Cost per 1,000 pages: $74 vs $1.31. Median time per resume: 38 s vs 138 s.
GPT-5.6 wins on 233 of 299 resumes, Qwen 3.5 9B on 60, 6 tied.
84.81
GPT-5.6
A · Vision LLM
+8.78A − B, points of F1
76.03
Qwen 3.5 9B
B · Vision LLM
A better on 233level on 6B better on 60
GPT-5.6 is ahead by 8.78 points on these 299 resumes.
- Many entries 82A 84.7 · B 73.9+10.8
- Long bullet lists 100A 85.1 · B 73.2+11.9
- Split across pages 84A 85.1 · B 73.4+11.7
- Big skills list 59A 85.5 · B 76.3+9.2
- Custom sections 104A 82.3 · B 73.0+9.3
- Combined headings 37A 84.1 · B 72.2+11.9
- Promotions 88A 85.0 · B 74.8+10.2
- Partial dates 138A 85.7 · B 75.6+10.1
- Overlapping jobs 69A 83.2 · B 72.2+11.0
- Accented text 13A 86.5 · B 83.7+2.8
Side by side
| GPT-5.6 | Qwen 3.5 9B | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | OpenAI | Alibaba |
| Rank | 6 of 18 | 15 of 18 |
| F1 | 84.7 | 76.0 |
| Precision | 85.9 | 78.9 |
| Recall | 87.2 | 80.3 |
| Cost per 1,000 pages | $74 | $1.31 |
| Median time per resume | 38 s | 138 s |
| Resumes won | 233 | 60 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | GPT-5.6 | Qwen 3.5 9B | Gap |
|---|---|---|---|---|
| Busy page (noise) | 18 | 84.0 | 67.6 | +16.4 |
| Long bullet lists (extraction challenges) | 100 | 85.1 | 73.2 | +11.9 |
| Combined headings (extraction challenges) | 37 | 84.1 | 72.2 | +11.9 |
| Split across pages (extraction challenges) | 84 | 85.1 | 73.4 | +11.7 |
| Wording traps (distractors) | 16 | 89.1 | 77.7 | +11.4 |
| Overlapping jobs (extraction challenges) | 69 | 83.2 | 72.2 | +11.0 |
| Many entries (extraction challenges) | 82 | 84.7 | 73.9 | +10.8 |
| Photo (reading the page) | 45 | 85.5 | 74.9 | +10.6 |
| Repeated header (noise) | 67 | 86.8 | 76.4 | +10.4 |
| References (noise) | 55 | 87.9 | 77.6 | +10.3 |
| Promotions (extraction challenges) | 88 | 85.0 | 74.8 | +10.2 |
| Partial dates (extraction challenges) | 138 | 85.7 | 75.6 | +10.1 |
| Section | Resumes | GPT-5.6 | Qwen 3.5 9B | Gap |
|---|---|---|---|---|
| Contact details | 299 | 94.1 | 92.6 | +1.5 |
| Experience | 299 | 88.4 | 77.2 | +11.2 |
| Education | 296 | 91.9 | 80.8 | +11.1 |
| Summary | 289 | 83.7 | 77.9 | +5.8 |
| Certifications | 254 | 78.2 | 74.7 | +3.5 |
| Awards | 94 | 81.0 | 55.2 | +25.8 |
| Skills | 286 | 93.2 | 83.2 | +10.0 |
| Languages | 168 | 85.9 | 80.0 | +5.8 |
| Custom sections | 211 | 51.2 | 36.4 | +14.8 |
| Interests | 64 | 91.4 | 84.3 | +7.1 |
| Volunteering | 77 | 85.7 | 80.2 | +5.5 |
| Profiles | 36 | 68.5 | 51.3 | +17.2 |
| Projects | 79 | 89.2 | 75.5 | +13.7 |
| Publications | 51 | 77.8 | 76.3 | +1.5 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | GPT-5.6 | Qwen 3.5 9B | Gap |
|---|---|---|---|
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 78.6 | 54.9 | +23.7 |
| Senior Controls Engineer: Five custom sections | 86.8 | 65.2 | +21.6 |
| Senior Game Designer: Game designer's Canva resume | 75.9 | 55.1 | +20.8 |
| Lead Respiratory Therapist (Adult ICU): Long resume, 34 entries | 88.7 | 69.4 | +19.3 |
Which to pick
GPT-5.6 is more accurate. Qwen 3.5 9B costs 98% less for 8.7 fewer points.
GPT-5.6 does better on busy page, long bullet lists and combined headings.
GPT-5.6 against every other model
- GPT-5.6 vs LlamaParse Extract Agentic Plus
- GPT-5.6 vs LlamaParse Extract Agentic
- GPT-5.6 vs LlamaParse Extract Cost-Effective
- GPT-5.6 vs Reducto Deep Extract
- GPT-5.6 vs Reducto Extract
- GPT-5.6 vs Extend Max Context
- GPT-5.6 vs Extend Extract
- GPT-5.6 vs GPT-5.5
- GPT-5.6 vs GPT-5.4 Nano
- GPT-5.6 vs Claude Opus 4.8
- GPT-5.6 vs Gemini 3.5 Flash
- GPT-5.6 vs Gemma 4 26B
- GPT-5.6 vs Qwen 3.8 Flash
- GPT-5.6 vs Qwen 3.8 27B
- GPT-5.6 vs Qwen 3.5 35B
- GPT-5.6 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-5.5
- 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 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