GLM 5.3 Flash vs GPT-5.4 Nano: which is better at resume extraction?
GPT-5.4 Nano leads GLM 5.3 Flash by 2.6 points (61.9 vs 59.2 F1). Cost per 1,000 pages: $0.80 vs $2.76. Median time per resume: 46 s vs 20 s.
GLM 5.3 Flash wins on 157 of 298 resumes, GPT-5.4 Nano on 138, 3 tied.
59.24
GLM 5.3 Flash
A · Vision LLM
61.96
GPT-5.4 Nano
B · Vision LLM
GPT-5.4 Nano is ahead by 2.72 points on these 298 resumes.
- Many entries 82A 43.1 · B 59.8−16.8
- Long bullet lists 100A 42.8 · B 61.1−18.4
- Split across pages 84A 51.2 · B 61.8−10.7
- Big skills list 60A 64.0 · B 64.9−0.9
- Custom sections 104A 48.2 · B 59.6−11.4
- Combined headings 37A 65.7 · B 64.6+1.1
- Promotions 87A 59.2 · B 58.8+0.4
- Partial dates 137A 52.2 · B 62.7−10.5
- Overlapping jobs 68A 50.4 · B 58.8−8.4
- Accented text 13A 51.1 · B 55.9−4.7
Side by side
| GLM 5.3 Flash | GPT-5.4 Nano | |
|---|---|---|
| Type | Vision LLM | Vision LLM |
| Maker | Zhipu AI | OpenAI |
| Rank | 18 of 18 | 17 of 18 |
| F1 | 59.2 | 61.9 |
| Precision | 71.8 | 70.1 |
| Recall | 62.0 | 66.3 |
| Cost per 1,000 pages | $0.80 | $2.76 |
| Median time per resume | 46 s | 20 s |
| Resumes won | 157 | 138 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | GLM 5.3 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|---|
| Content as images (distractors) | 8 | 10.4 | 61.3 | −50.9 |
| Scanned & rotated (source) | 18 | 54.8 | 4.2 | +50.7 |
| Distractor layouts (source) | 18 | 17.8 | 62.2 | −44.4 |
| Wording traps (distractors) | 16 | 20.2 | 62.8 | −42.6 |
| Rotated & warped pages (distractors) | 20 | 50.3 | 10.0 | +40.3 |
| Medium (length) | 90 | 24.2 | 60.2 | −36.0 |
| Image only (reading the page) | 28 | 53.3 | 21.0 | +32.3 |
| Hidden text & prompt injections (distractors) | 13 | 46.4 | 20.2 | +26.2 |
| Resume-builder templates (source) | 10 | 48.8 | 67.6 | −18.7 |
| Long bullet lists (extraction challenges) | 100 | 42.8 | 61.1 | −18.4 |
| Expert-built (Canva / LaTeX) (source) | 32 | 42.9 | 60.3 | −17.4 |
| Many entries (extraction challenges) | 82 | 43.1 | 59.8 | −16.8 |
| Section | Resumes | GLM 5.3 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|---|
| Contact details | 298 | 73.3 | 81.3 | −8.1 |
| Experience | 298 | 55.4 | 63.7 | −8.3 |
| Education | 296 | 64.2 | 62.3 | +1.9 |
| Summary | 288 | 50.7 | 71.9 | −21.2 |
| Certifications | 244 | 57.3 | 53.8 | +3.5 |
| Awards | 103 | 54.3 | 45.2 | +9.2 |
| Skills | 285 | 66.0 | 74.8 | −8.8 |
| Languages | 178 | 65.8 | 68.8 | −3.0 |
| Custom sections | 212 | 32.0 | 26.6 | +5.4 |
| Interests | 74 | 44.3 | 70.0 | −25.7 |
| Volunteering | 79 | 43.0 | 60.2 | −17.2 |
| Profiles | 38 | 41.2 | 23.7 | +17.5 |
| Projects | 81 | 60.5 | 48.8 | +11.8 |
| Publications | 50 | 65.0 | 30.7 | +34.3 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | GLM 5.3 Flash | GPT-5.4 Nano | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 87.0 | 1.7 | +85.3 |
| Lead Solutions Architect: Indian format with CGPA | 17.2 | 77.6 | −60.4 |
| Head of Events & Public Space Operations: Table grid with emoji bullets | 10.8 | 61.5 | −50.6 |
| Senior Manufacturing Engineer: Third-person prose, no bullets | 25.4 | 75.2 | −49.7 |
Which to pick
GPT-5.4 Nano is more accurate. GLM 5.3 Flash costs 71% less for 2.6 fewer points.
GLM 5.3 Flash does better on scanned & rotated, rotated & warped pages and image only. GPT-5.4 Nano does better on content as images, distractor layouts and wording traps.
GLM 5.3 Flash against every other model
- GLM 5.3 Flash vs LlamaParse Extract Agentic Plus
- GLM 5.3 Flash vs LlamaParse Extract Agentic
- GLM 5.3 Flash vs LlamaParse Extract Cost-Effective
- GLM 5.3 Flash vs Reducto Deep Extract
- GLM 5.3 Flash vs Reducto Extract
- GLM 5.3 Flash vs Extend Max Context
- GLM 5.3 Flash vs Extend Extract
- GLM 5.3 Flash vs GPT-5.6
- GLM 5.3 Flash vs GPT-5.5
- GLM 5.3 Flash vs Claude Opus 4.8
- GLM 5.3 Flash vs Gemini 3.5 Flash
- GLM 5.3 Flash vs Gemma 4 26B
- GLM 5.3 Flash vs Qwen 3.8 Flash
- GLM 5.3 Flash vs Qwen 3.8 27B
- GLM 5.3 Flash vs Qwen 3.5 35B
- GLM 5.3 Flash vs Qwen 3.5 9B
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 35B
- GPT-5.4 Nano vs Qwen 3.5 9B
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