Open Source11B
Llama-3.2-Vision-11B
Meta131K context2025-09-25
Overall Rank
#13
of 16 models
Overall Score
50.1
avg across benchmarks
Best Task
Key Information Extraction
76.1
Weakest Task
Table Understanding
44.8
Benchmark Performance
OlmOCR Benchv1.0
13/16
| Overall | Math | Table | Present | Absent | Order |
|---|---|---|---|---|---|
| 47.2 | 39.4 | 60.8 | 30.5 | 69.9 | 51.9 |
OmniDocBenchv1.5
15/16
| Overall | Text Edit↓ | Formula CDM↑ | Table TEDS↑ | TEDS-S↑ | Read Order↓ |
|---|---|---|---|---|---|
| 44.6 | 0.541 | 55.4 | 32.6 | 42.9 | 0.340 |
IDP Core Benchv1.0
13/16
| Overall | KIE | OCR | Table | VQA |
|---|---|---|---|---|
| 58.6 | 76.1 | 65.8 | 41.1 | 51.5 |
Capability Profile
Strength Analysis
Auto-generated from benchmark scores
Strengths
- Key Information Extraction76.1
- Layout & Order58.9
Weaknesses
- Table Understanding44.8
- Text Extraction45.9