GPT-5.5 vs Qwen 3.8 Flash: which is better at resume extraction?
Qwen 3.8 Flash leads GPT-5.5 by 0.6 points (84.1 vs 83.5 F1). Cost per 1,000 pages: $87 vs $2.98. Median time per resume: 48 s vs 122 s.
GPT-5.5 wins on 139 of 298 resumes, Qwen 3.8 Flash on 117, 42 tied.
83.67
GPT-5.5
A · Vision LLM
84.08
Qwen 3.8 Flash
B · Vision LLM
Qwen 3.8 Flash is ahead by 0.41 points on these 298 resumes.
- Many entries 83A 83.8 · B 82.1+1.7
- Long bullet lists 101A 84.3 · B 82.1+2.2
- Split across pages 85A 85.0 · B 82.6+2.4
- Big skills list 60A 85.5 · B 83.5+2.0
- Custom sections 105A 80.8 · B 80.8−0.0
- Combined headings 37A 83.4 · B 82.6+0.8
- Promotions 86A 84.6 · B 83.8+0.8
- Partial dates 138A 84.0 · B 83.3+0.7
- Overlapping jobs 68A 82.3 · B 81.4+0.9
- Accented text 13A 83.9 · B 86.9−3.0
Side by side
| GPT-5.5 | Qwen 3.8 Flash | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | OpenAI | Alibaba |
| Rank | 11 of 18 | 8 of 18 |
| F1 | 83.5 | 84.1 |
| Precision | 84.9 | 85.4 |
| Recall | 86.0 | 86.2 |
| Cost per 1,000 pages | $87 | $2.98 |
| Median time per resume | 48 s | 122 s |
| Resumes won | 139 | 117 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | GPT-5.5 | Qwen 3.8 Flash | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 17 | 61.0 | 86.3 | −25.3 |
| Rotated & warped pages (distractors) | 19 | 63.9 | 86.0 | −22.1 |
| Hidden text & prompt injections (distractors) | 13 | 69.4 | 89.0 | −19.7 |
| Image only (reading the page) | 26 | 56.9 | 74.7 | −17.8 |
| Typos kept on purpose (distractors) | 35 | 74.2 | 85.7 | −11.6 |
| Content as images (distractors) | 8 | 92.6 | 88.3 | +4.3 |
| Two-column sidebar (layout) | 52 | 78.0 | 81.6 | −3.7 |
| References (noise) | 55 | 88.5 | 85.4 | +3.1 |
| Accented text (extraction challenges) | 13 | 83.9 | 86.9 | −3.0 |
| Sidebar or grid (reading the page) | 80 | 81.0 | 83.8 | −2.8 |
| Busy page (noise) | 18 | 82.5 | 79.8 | +2.7 |
| Handwritten (reading the page) | 9 | 51.0 | 53.5 | −2.5 |
| Section | Resumes | GPT-5.5 | Qwen 3.8 Flash | Gap |
|---|---|---|---|---|
| Contact details | 298 | 93.7 | 93.6 | +0.0 |
| Experience | 298 | 87.6 | 88.0 | −0.3 |
| Education | 295 | 89.0 | 90.1 | −1.1 |
| Summary | 289 | 82.4 | 81.3 | +1.1 |
| Certifications | 253 | 77.5 | 82.2 | −4.7 |
| Awards | 96 | 74.7 | 75.5 | −0.8 |
| Skills | 285 | 91.8 | 90.0 | +1.8 |
| Languages | 166 | 85.5 | 84.6 | +0.9 |
| Custom sections | 208 | 49.9 | 53.9 | −4.0 |
| Interests | 63 | 92.3 | 91.1 | +1.2 |
| Volunteering | 76 | 84.8 | 83.9 | +0.9 |
| Profiles | 36 | 68.5 | 73.3 | −4.8 |
| Projects | 79 | 86.4 | 83.7 | +2.7 |
| Publications | 51 | 82.5 | 81.8 | +0.7 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | GPT-5.5 | Qwen 3.8 Flash | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 64.7 | 88.0 | −23.3 |
| Creative Director: Accented text in a styled CV | 90.2 | 76.8 | +13.5 |
| Senior Data Scientist (Lead): Fifty skills, seven pages | 71.2 | 83.9 | −12.7 |
| Senior Customer Service Representative: Busy Canva page | 85.8 | 94.9 | −9.1 |
Which to pick
Qwen 3.8 Flash is more accurate. GPT-5.5 is the faster of the two.
GPT-5.5 does better on content as images, references and busy page. Qwen 3.8 Flash does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections.
GPT-5.5 against every other model
- GPT-5.5 vs LlamaParse Extract Agentic Plus
- GPT-5.5 vs LlamaParse Extract Agentic
- GPT-5.5 vs LlamaParse Extract Cost-Effective
- GPT-5.5 vs Reducto Deep Extract
- GPT-5.5 vs Reducto Extract
- GPT-5.5 vs Extend Max Context
- GPT-5.5 vs Extend Extract
- GPT-5.5 vs GPT-5.6
- GPT-5.5 vs GPT-5.4 Nano
- GPT-5.5 vs Claude Opus 4.8
- GPT-5.5 vs Gemini 3.5 Flash
- GPT-5.5 vs Gemma 4 26B
- GPT-5.5 vs Qwen 3.8 27B
- GPT-5.5 vs Qwen 3.5 35B
- GPT-5.5 vs Qwen 3.5 9B
- GPT-5.5 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-5.4 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 Qwen 3.5 9B
- 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