Gemma 4 26B vs LlamaParse Extract Agentic: which is better at resume extraction?
LlamaParse Extract Agentic leads Gemma 4 26B by 22.7 points (85.5 vs 62.8 F1). Cost per 1,000 pages: $0.56 vs $23. Median time per resume: 29 s vs 55 s.
Gemma 4 26B wins on 14 of 300 resumes, LlamaParse Extract Agentic on 283, 3 tied.
62.82
Gemma 4 26B
A · Vision LLM
85.55
LlamaParse Extract Agentic
B · Document API
LlamaParse Extract Agentic is ahead by 22.73 points on these 300 resumes.
- Many entries 83A 58.0 · B 83.0−25.0
- Long bullet lists 101A 57.4 · B 83.1−25.7
- Split across pages 85A 60.0 · B 83.8−23.8
- Big skills list 60A 61.9 · B 83.7−21.8
- Custom sections 105A 58.9 · B 82.5−23.6
- Combined headings 37A 60.9 · B 83.2−22.3
- Promotions 88A 59.5 · B 85.5−26.1
- Partial dates 138A 62.9 · B 85.6−22.7
- Overlapping jobs 69A 57.5 · B 83.5−26.0
- Accented text 13A 59.3 · B 91.1−31.9
Side by side
| Gemma 4 26B | LlamaParse Extract Agentic | |
|---|---|---|
| Type | Vision LLM | Document API |
| Maker | LlamaIndex | |
| Rank | 16 of 18 | 3 of 18 |
| F1 | 62.8 | 85.5 |
| Precision | 69.1 | 86.8 |
| Recall | 66.2 | 87.3 |
| Cost per 1,000 pages | $0.56 | $23 |
| Median time per resume | 29 s | 55 s |
| Resumes won | 14 | 283 |
Where each one wins
Where the two differ most.
| Kind of resume | # of resumes | Gemma 4 26B | LlamaParse Extract Agentic | Gap |
|---|---|---|---|---|
| Scanned & rotated (source) | 19 | 40.7 | 91.6 | −50.9 |
| Rotated & warped pages (distractors) | 21 | 43.4 | 90.9 | −47.5 |
| Hidden text & prompt injections (distractors) | 14 | 48.6 | 91.6 | −43.0 |
| Typos kept on purpose (distractors) | 37 | 51.3 | 90.0 | −38.8 |
| Image only (reading the page) | 28 | 45.1 | 81.5 | −36.3 |
| Accented text (extraction challenges) | 13 | 59.3 | 91.1 | −31.9 |
| Two-column sidebar (layout) | 54 | 55.1 | 84.0 | −28.9 |
| Wording traps (distractors) | 16 | 60.6 | 87.7 | −27.1 |
| Promotions (extraction challenges) | 88 | 59.5 | 85.5 | −26.1 |
| Overlapping jobs (extraction challenges) | 69 | 57.5 | 83.5 | −26.0 |
| Content as images (distractors) | 8 | 65.6 | 91.6 | −26.0 |
| Distractor layouts (source) | 18 | 62.4 | 88.3 | −26.0 |
| Section | Resumes | Gemma 4 26B | LlamaParse Extract Agentic | Gap |
|---|---|---|---|---|
| Contact details | 300 | 84.6 | 94.5 | −10.0 |
| Experience | 300 | 56.0 | 89.6 | −33.6 |
| Education | 297 | 68.0 | 91.3 | −23.3 |
| Summary | 290 | 35.5 | 84.5 | −49.0 |
| Certifications | 244 | 63.0 | 80.9 | −18.0 |
| Awards | 110 | 55.7 | 80.3 | −24.6 |
| Skills | 290 | 81.9 | 95.1 | −13.2 |
| Languages | 171 | 80.9 | 84.2 | −3.2 |
| Custom sections | 223 | 32.3 | 52.1 | −19.8 |
| Interests | 67 | 71.8 | 93.0 | −21.1 |
| Volunteering | 78 | 59.4 | 87.7 | −28.3 |
| Profiles | 37 | 57.2 | 67.6 | −10.4 |
| Projects | 83 | 55.0 | 86.2 | −31.2 |
| Publications | 53 | 54.5 | 80.5 | −26.0 |
Published resumes they disagree on most
Open one to see both outputs field by field.
| Resume | Gemma 4 26B | LlamaParse Extract Agentic | Gap |
|---|---|---|---|
| Deputy Head of Mission: Rotated scan with hidden identity text | 34.1 | 89.5 | −55.4 |
| Supervisory Human Resources Specialist, GS-0201-15: Resume-builder template | 29.4 | 83.9 | −54.5 |
| Creative Director: Accented text in a styled CV | 39.3 | 84.2 | −44.9 |
| Head of Developer Experience (Internal title: Jedi Master of Analytics) : Unusual job titles | 53.7 | 91.9 | −38.2 |
Which to pick
LlamaParse Extract Agentic is more accurate. Gemma 4 26B costs 98% less for 22.7 fewer points.
LlamaParse Extract Agentic does better on scanned & rotated, rotated & warped pages and hidden text & prompt injections.
Gemma 4 26B against every other model
- Gemma 4 26B vs LlamaParse Extract Agentic Plus
- Gemma 4 26B vs LlamaParse Extract Cost-Effective
- Gemma 4 26B vs Reducto Deep Extract
- Gemma 4 26B vs Reducto Extract
- Gemma 4 26B vs Extend Max Context
- Gemma 4 26B vs Extend Extract
- Gemma 4 26B vs GPT-5.6
- Gemma 4 26B vs GPT-5.5
- Gemma 4 26B vs GPT-5.4 Nano
- Gemma 4 26B vs Claude Opus 4.8
- Gemma 4 26B vs Gemini 3.5 Flash
- Gemma 4 26B vs Qwen 3.8 Flash
- Gemma 4 26B vs Qwen 3.8 27B
- Gemma 4 26B vs Qwen 3.5 35B
- Gemma 4 26B vs Qwen 3.5 9B
- Gemma 4 26B vs GLM 5.3 Flash
LlamaParse Extract Agentic against every other model
- LlamaParse Extract Agentic vs LlamaParse Extract Agentic Plus
- LlamaParse Extract Agentic vs LlamaParse Extract Cost-Effective
- LlamaParse Extract Agentic vs Reducto Deep Extract
- LlamaParse Extract Agentic vs Reducto Extract
- LlamaParse Extract Agentic vs Extend Max Context
- LlamaParse Extract Agentic vs Extend Extract
- LlamaParse Extract Agentic vs GPT-5.6
- LlamaParse Extract Agentic vs GPT-5.5
- LlamaParse Extract Agentic vs GPT-5.4 Nano
- LlamaParse Extract Agentic vs Claude Opus 4.8
- LlamaParse Extract Agentic vs Gemini 3.5 Flash
- LlamaParse Extract Agentic vs Qwen 3.8 Flash
- LlamaParse Extract Agentic vs Qwen 3.8 27B
- LlamaParse Extract Agentic vs Qwen 3.5 35B
- LlamaParse Extract Agentic vs Qwen 3.5 9B
- LlamaParse Extract Agentic 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