GLM-OCR
Zhipu AI2025-06-01
Overall Rank
#19
of 27 models
Overall Score
64.2
avg across benchmarks
Best Task
Text Extraction
85.6
Weakest Task
Table Understanding
40.3
Benchmark Performance
OlmOCR Benchv1.0
14/27
| Overall | ArXiv Math | H&F | Long/Tiny | Multi-Col | Old Scans | Scans Math | Tables |
|---|---|---|---|---|---|---|---|
| 68.4 | 67.3 | 89.2 | 35.7 | 80.3 | 41.3 | 75.5 | 59.0 |
OmniDocBenchv1.5
19/27
| Overall | Text Edit↓ | CDM↑ | TEDS↑ | TEDS-S↑ | Read Order↓ |
|---|---|---|---|---|---|
| 69.2 | 0.144 | 84.7 | 37.4 | 39.3 | 0.141 |
IDP Core Benchv1.0
24/27
| Overall | KIE | OCR | Table | VQA |
|---|---|---|---|---|
| 54.9 | 83.5 | 66.7 | 24.5 | 44.9 |
Capability Profile
Strength Analysis
Auto-generated from benchmark scores
Strengths
- Text Extraction85.6
- Formula84.7
Weaknesses
- Table Understanding40.3
- Visual QA44.9