Gemini 3.5 Flash vs GPT-5.4 Nano: which is better at resume extraction?
Gemini 3.5 Flash leads GPT-5.4 Nano by 23.6 points (85.5 vs 61.9 F1). Cost per 1,000 pages: $35 vs $2.76. Median time per resume: 43 s vs 20 s.
Gemini 3.5 Flash wins on 277 of 300 resumes, GPT-5.4 Nano on 20, 3 tied.
85.45
Gemini 3.5 Flash
A · Vision LLM
61.86
GPT-5.4 Nano
B · Vision LLM
Gemini 3.5 Flash is ahead by 23.59 points on these 300 resumes.
- Many entries 83A 84.3 · B 59.7+24.6
- Long bullet lists 101A 83.9 · B 61.0+22.8
- Split across pages 85A 84.6 · B 61.7+22.9
- Big skills list 60A 86.0 · B 64.9+21.2
- Custom sections 105A 83.6 · B 59.5+24.1
- Combined headings 37A 84.2 · B 64.6+19.6
- Promotions 88A 87.3 · B 58.7+28.7
- Partial dates 138A 84.9 · B 62.6+22.4
- Overlapping jobs 69A 86.4 · B 58.6+27.8
- Accented text 13A 86.4 · B 55.9+30.5
Side by side
| Gemini 3.5 Flash | GPT-5.4 Nano | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | OpenAI | |
| Rank | 5 of 18 | 17 of 18 |
| F1 | 85.5 | 61.9 |
| Precision | 86.9 | 70.1 |
| Recall | 87.9 | 66.3 |
| Cost per 1,000 pages | $35 | $2.76 |
| Median time per resume | 43 s | 20 s |
| Resumes won | 277 | 20 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Gemini 3.5 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 87.2 | 6.3 | +80.9 |
| Rotated & warped pages (distractors) | 21 | 87.4 | 11.7 | +75.7 |
| Hidden text & prompt injections (distractors) | 14 | 86.7 | 22.0 | +64.7 |
| Image only (reading the page) | 28 | 75.1 | 21.0 | +54.1 |
| Typos kept on purpose (distractors) | 37 | 84.4 | 33.5 | +50.9 |
| Two-column sidebar (layout) | 54 | 83.4 | 47.4 | +36.0 |
| Accented text (extraction challenges) | 13 | 86.4 | 55.9 | +30.5 |
| Sidebar or grid (reading the page) | 82 | 85.1 | 54.9 | +30.2 |
| Promotions (extraction challenges) | 88 | 87.3 | 58.7 | +28.7 |
| Overlapping jobs (extraction challenges) | 69 | 86.4 | 58.6 | +27.8 |
| Photo (reading the page) | 45 | 85.1 | 59.4 | +25.7 |
| Medium (length) | 90 | 85.0 | 60.2 | +24.8 |
| Section | Resumes | Gemini 3.5 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|---|
| Contact details | 300 | 95.3 | 81.2 | +14.1 |
| Experience | 300 | 86.5 | 63.7 | +22.8 |
| Education | 297 | 91.5 | 62.2 | +29.3 |
| Summary | 290 | 89.3 | 71.7 | +17.6 |
| Certifications | 245 | 84.0 | 53.6 | +30.4 |
| Awards | 92 | 80.3 | 44.9 | +35.4 |
| Skills | 289 | 93.5 | 74.8 | +18.7 |
| Languages | 170 | 84.3 | 69.0 | +15.3 |
| Custom sections | 216 | 50.1 | 26.3 | +23.8 |
| Interests | 63 | 94.2 | 70.0 | +24.2 |
| Volunteering | 79 | 83.7 | 59.9 | +23.9 |
| Profiles | 35 | 64.8 | 23.7 | +41.0 |
| Projects | 80 | 80.4 | 48.8 | +31.6 |
| Publications | 52 | 83.4 | 30.7 | +52.7 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Gemini 3.5 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 82.6 | 1.7 | +80.9 |
| Partner: Consulting CV, six pages | 79.1 | 27.8 | +51.3 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 82.7 | 32.5 | +50.2 |
| Head of Digital Marketing: Canva design with a photo | 83.5 | 39.3 | +44.2 |
Which to pick
Gemini 3.5 Flash is more accurate. GPT-5.4 Nano costs 92% less for 23.6 fewer points.
Gemini 3.5 Flash does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections. GPT-5.4 Nano does better on handwritten.
Gemini 3.5 Flash against every other model
- Gemini 3.5 Flash vs LlamaParse Extract Agentic Plus
- Gemini 3.5 Flash vs LlamaParse Extract Agentic
- Gemini 3.5 Flash vs LlamaParse Extract Cost-Effective
- Gemini 3.5 Flash vs Reducto Deep Extract
- Gemini 3.5 Flash vs Reducto Extract
- Gemini 3.5 Flash vs Extend Max Context
- Gemini 3.5 Flash vs Extend Extract
- Gemini 3.5 Flash vs GPT-5.6
- Gemini 3.5 Flash vs GPT-5.5
- Gemini 3.5 Flash vs Claude Opus 4.8
- Gemini 3.5 Flash vs Gemma 4 26B
- Gemini 3.5 Flash vs Qwen 3.8 Flash
- Gemini 3.5 Flash vs Qwen 3.8 27B
- Gemini 3.5 Flash vs Qwen 3.5 35B
- Gemini 3.5 Flash vs Qwen 3.5 9B
- Gemini 3.5 Flash vs GLM 5.3 Flash
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 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 35B
- GPT-5.4 Nano vs Qwen 3.5 9B
- GPT-5.4 Nano 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